{"id":31037,"date":"2025-10-05T01:02:00","date_gmt":"2025-10-04T22:02:00","guid":{"rendered":"https:\/\/fatihsoysal.com\/blog\/luckfox-lyra-zero-w-ile-gelisime-baslangic-rehberi\/"},"modified":"2025-10-05T01:02:00","modified_gmt":"2025-10-04T22:02:00","slug":"luckfox-lyra-zero-w-ile-gelisime-baslangic-rehberi","status":"publish","type":"post","link":"https:\/\/fatihsoysal.com\/blog\/luckfox-lyra-zero-w-ile-gelisime-baslangic-rehberi\/","title":{"rendered":"Luckfox Lyra Zero W ile Geli\u015fime Ba\u015flang\u0131\u00e7 Rehberi"},"content":{"rendered":"<p>\n    IoT ve g\u00f6m\u00fcl\u00fc yapay zeka projeleriniz i\u00e7in uygun maliyetli, g\u00fc\u00e7l\u00fc ve k\u00fc\u00e7\u00fck bir \u00e7\u00f6z\u00fcm m\u00fc ar\u0131yorsunuz? Luckfox Lyra Zero W, inovatif fikirlerinizi ger\u00e7e\u011fe d\u00f6n\u00fc\u015ft\u00fcrmek i\u00e7in size harika bir ba\u015flang\u0131\u00e7 noktas\u0131 sunuyor. Bu rehberle, s\u0131f\u0131rdan ba\u015flayarak Luckfox Lyra Zero W d\u00fcnyas\u0131na ad\u0131m atacak ve kendi ak\u0131ll\u0131 cihazlar\u0131n\u0131z\u0131 in\u015fa etmenin keyfini ya\u015fayacaks\u0131n\u0131z.\n<\/p>\n<p>\n    G\u00fcn\u00fcm\u00fczde nesnelerin interneti (IoT) ve yapay zeka (AI) uygulamalar\u0131 hayat\u0131m\u0131z\u0131n her alan\u0131na h\u0131zla entegre oluyor. Ak\u0131ll\u0131 ev cihazlar\u0131ndan end\u00fcstriyel otomasyona, giyilebilir teknolojilerden ak\u0131ll\u0131 \u015fehirlere kadar geni\u015f bir yelpazede, g\u00f6m\u00fcl\u00fc sistemlerin \u00f6nemi her ge\u00e7en g\u00fcn artmaktad\u0131r. Ancak, bu t\u00fcr projeler i\u00e7in uygun maliyetli, y\u00fcksek performansl\u0131 ve kompakt donan\u0131mlar bulmak bazen zorlay\u0131c\u0131 olabilir. \u0130\u015fte tam da bu noktada, Luckfox Lyra Zero W devreye giriyor ve \u00f6zellikle b\u00fct\u00e7e dostu yap\u0131s\u0131 ve zengin \u00f6zellikleri ile dikkat \u00e7ekiyor. Peki, bu k\u00fc\u00e7\u00fck kart\u0131 bu kadar pop\u00fcler yapan nedir?\n<\/p>\n<p>\n    Luckfox Lyra Zero W, Rockchip&#8217;in g\u00fc\u00e7l\u00fc RV1103 System-on-Chip (SoC) i\u015flemcisi \u00fczerine in\u015fa edilmi\u015f, kredi kart\u0131ndan bile k\u00fc\u00e7\u00fck bir geli\u015ftirme kart\u0131d\u0131r. En b\u00fcy\u00fck avantaj\u0131, entegre bir NPU (Neural Processing Unit) yani yapay zeka h\u0131zland\u0131r\u0131c\u0131s\u0131na sahip olmas\u0131d\u0131r. Bu sayede, cihaz \u00fczerinde do\u011frudan yapay zeka \u00e7\u0131kar\u0131mlar\u0131 yapabilir, nesne tan\u0131ma, y\u00fcz alg\u0131lama veya ses i\u015fleme gibi g\u00f6revleri d\u00fc\u015f\u00fck g\u00fc\u00e7 t\u00fcketimiyle ve y\u00fcksek performansla ger\u00e7ekle\u015ftirebilirsiniz. Bu \u00f6zellik, \u00f6zellikle kenar bili\u015fim (edge computing) uygulamalar\u0131 i\u00e7in kritik bir avantaj sunar. Geleneksel olarak bu t\u00fcr AI g\u00f6revleri bulut sunucular\u0131na ba\u011f\u0131ml\u0131yken, Luckfox Lyra Zero W gibi kartlar verilerin yerel olarak i\u015flenmesini sa\u011flayarak gecikmeyi azalt\u0131r, gizlili\u011fi art\u0131r\u0131r ve bant geni\u015fli\u011fi maliyetlerinden tasarruf etmenizi sa\u011flar.\n<\/p>\n<p>\n    Sadece yapay zeka yetenekleriyle de\u011fil, ayn\u0131 zamanda kompakt boyutlar\u0131yla da \u00f6n plana \u00e7\u0131kar. K\u00fc\u00e7\u00fck boyutu sayesinde, dronlardan giyilebilir cihazlara, ak\u0131ll\u0131 kap\u0131 zillerinden g\u00fcvenlik kameralar\u0131na kadar pek \u00e7ok alanda kolayca entegre edilebilir. Dahili Wi-Fi ve Bluetooth ba\u011flant\u0131 \u00f6zellikleri sayesinde de kablosuz ileti\u015fim gerektiren projelerinizde ek mod\u00fcl ihtiyac\u0131n\u0131 ortadan kald\u0131r\u0131r. D\u00fc\u015f\u00fck g\u00fc\u00e7 t\u00fcketimi ise batarya ile \u00e7al\u0131\u015fan uygulamalar i\u00e7in ideal bir se\u00e7im olmas\u0131n\u0131 sa\u011flar. T\u00fcm bu \u00f6zellikler bir araya geldi\u011finde, Luckfox Lyra Zero W, hem hobi ama\u00e7l\u0131 projeler geli\u015ftiren merakl\u0131lar hem de profesyonel IoT \u00fcr\u00fcnleri tasarlayan m\u00fchendisler i\u00e7in cazip bir se\u00e7enek haline gelmektedir. Bu rehber boyunca, bu kart\u0131n potansiyelini nas\u0131l ortaya \u00e7\u0131karaca\u011f\u0131n\u0131z\u0131 ad\u0131m ad\u0131m ke\u015ffedece\u011fiz.\n<\/p>\n<h2>Temel Kavramlar: Luckfox Lyra Zero W Nedir ve Ne Sunar?<\/h2>\n<p>\n    Luckfox Lyra Zero W ile derinlemesine bir yolculu\u011fa \u00e7\u0131kmadan \u00f6nce, bu minik ama g\u00fc\u00e7l\u00fc kart\u0131n temel \u00f6zelliklerini ve mimarisini anlamak olduk\u00e7a \u00f6nemlidir. Bu bilgiler, projenizin gereksinimlerine en uygun \u00e7\u00f6z\u00fcmleri geli\u015ftirmenize yard\u0131mc\u0131 olacakt\u0131r. Temelinde, Rockchip RV1103 adl\u0131 entegre bir yonga setini bar\u0131nd\u0131r\u0131r. Bu SoC (System-on-Chip), tek bir yonga \u00fczerinde bir mikroi\u015flemci, bellek, grafik i\u015flem birimi (varsa), \u00e7evre birimleri ve en \u00f6nemlisi bir NPU&#8217;yu bir araya getirir.\n<\/p>\n<p>\n    RV1103&#8217;\u00fcn kalbinde, g\u00f6m\u00fcl\u00fc uygulamalar i\u00e7in optimize edilmi\u015f bir ARM Cortex-A7 i\u015flemci \u00e7ekirde\u011fi bulunur. Bu \u00e7ekirdek, i\u015fletim sistemi ve genel uygulama mant\u0131\u011f\u0131n\u0131 \u00e7al\u0131\u015ft\u0131rmak i\u00e7in yeterli i\u015flem g\u00fcc\u00fcn\u00fc sunar. Ancak, kart\u0131n as\u0131l parlad\u0131\u011f\u0131 nokta, entegre NPU&#8217;sudur. Bu yapay zeka h\u0131zland\u0131r\u0131c\u0131s\u0131, geleneksel CPU&#8217;lara k\u0131yasla \u00e7ok daha verimli bir \u015fekilde yapay zeka \u00e7\u0131kar\u0131m g\u00f6revlerini (inference) ger\u00e7ekle\u015ftirebilir. Bu sayede, g\u00f6r\u00fcnt\u00fc i\u015fleme, ses tan\u0131ma veya anlamsal segmentasyon gibi karma\u015f\u0131k AI algoritmalar\u0131n\u0131 d\u00fc\u015f\u00fck g\u00fc\u00e7le ve ger\u00e7ek zamanl\u0131ya yak\u0131n h\u0131zlarda \u00e7al\u0131\u015ft\u0131rabilir. Bu durum, \u00f6zellikle pil \u00f6mr\u00fcn\u00fcn kritik oldu\u011fu mobil veya batarya destekli IoT cihazlar\u0131 i\u00e7in b\u00fcy\u00fck bir avantajd\u0131r. \u00d6rne\u011fin, bir ak\u0131ll\u0131 kap\u0131 zili, gelen ziyaret\u00e7iyi h\u0131zl\u0131ca alg\u0131lay\u0131p analiz edebilir, bu bilgiyi yerel olarak i\u015fleyip sadece gerekli durumlarda bir bildirim g\u00f6nderebilir.\n<\/p>\n<p>\n    Bellek taraf\u0131nda ise Luckfox Lyra Zero W, genellikle dahili bir miktar RAM ile birlikte gelir, bu da i\u015fletim sisteminin ve uygulamalar\u0131n \u00e7al\u0131\u015fmas\u0131 i\u00e7in gerekli alan\u0131 sa\u011flar. Harici depolama i\u00e7in ise bir microSD kart yuvas\u0131 bulunur, bu da kullan\u0131c\u0131lara i\u015fletim sistemi imajlar\u0131n\u0131 ve uygulama verilerini depolama esnekli\u011fi sunar. Ba\u011flant\u0131 se\u00e7enekleri de olduk\u00e7a geni\u015ftir. Dahili Wi-Fi mod\u00fcl\u00fc sayesinde internete ve di\u011fer a\u011f cihazlar\u0131na kolayca ba\u011flanabilir. Bluetooth deste\u011fi ise kablosuz sens\u00f6rlerle veya di\u011fer d\u00fc\u015f\u00fck g\u00fc\u00e7l\u00fc cihazlarla ileti\u015fim kurmak i\u00e7in idealdir. Ayr\u0131ca, MIPI CSI aray\u00fcz\u00fc sayesinde y\u00fcksek \u00e7\u00f6z\u00fcn\u00fcrl\u00fckl\u00fc kamera mod\u00fclleri kolayca ba\u011flanabilir; bu da onu yapay zeka destekli g\u00f6r\u00fcnt\u00fc i\u015fleme projeleri i\u00e7in m\u00fckemmel bir se\u00e7enek yapar. GPIO (Genel Ama\u00e7l\u0131 Giri\u015f\/\u00c7\u0131k\u0131\u015f) pinleri, I2C, SPI, UART gibi standart \u00e7evre birimi aray\u00fczleri de kart\u0131n d\u0131\u015f d\u00fcnya ile etkile\u015fim kurmas\u0131n\u0131 ve \u00e7e\u015fitli sens\u00f6rler, akt\u00fcat\u00f6rler veya di\u011fer donan\u0131mlarla birle\u015ftirilmesini sa\u011flar.\n<\/p>\n<p>\n    Raspberry Pi Zero W gibi benzer boyutlu kartlarla kar\u015f\u0131la\u015ft\u0131r\u0131ld\u0131\u011f\u0131nda, Luckfox Lyra Zero W&#8217;nin en belirgin fark\u0131, yukar\u0131da bahsedilen entegre NPU&#8217;sudur. Raspberry Pi Zero W genel ama\u00e7l\u0131 bir bilgisayar olarak daha uygunken, Luckfox Lyra Zero W \u00f6zellikle yerel AI i\u015fleme ve d\u00fc\u015f\u00fck g\u00fc\u00e7 t\u00fcketimi gerektiren AIoT (Yapay Zeka Nesnelerin \u0130nterneti) ve g\u00f6m\u00fcl\u00fc g\u00f6r\u00fc\u015f (embedded vision) uygulamalar\u0131 i\u00e7in tasarlanm\u0131\u015ft\u0131r. Bu \u00f6zel donan\u0131m h\u0131zland\u0131r\u0131c\u0131s\u0131, belirli AI g\u00f6revlerinde \u00e7ok daha \u00fcst\u00fcn performans sergilemesini sa\u011flar. Ak\u0131ll\u0131 ev otomasyonu, end\u00fcstriyel izleme sistemleri, giyilebilir yapay zeka cihazlar\u0131 ve ak\u0131ll\u0131 \u015fehir uygulamalar\u0131 gibi alanlar, Luckfox Lyra Zero W&#8217;nin potansiyelini tam anlam\u0131yla ortaya \u00e7\u0131karabilece\u011fi ba\u015fl\u0131ca kullan\u0131m alanlar\u0131d\u0131r.\n<\/p>\n<h2>Ba\u015flang\u0131\u00e7 \u0130\u00e7in Gerekli Malzemeler ve \u0130lk Kurulum Ad\u0131mlar\u0131 Nelerdir?<\/h2>\n<p>\n    Luckfox Lyra Zero W ile geli\u015ftirme yapmaya ba\u015flamak, do\u011fru malzemelerle ve ad\u0131m ad\u0131m bir yakla\u015f\u0131mla olduk\u00e7a kolayd\u0131r. T\u0131pk\u0131 herhangi bir mikrodenetleyici veya geli\u015ftirme kart\u0131nda oldu\u011fu gibi, ilk kurulum ad\u0131mlar\u0131 projenizin temelini olu\u015fturur. \u0130\u015fte ba\u015flamak i\u00e7in ihtiyac\u0131n\u0131z olanlar ve ilk kurulum s\u00fcreci:\n<\/p>\n<h3>Gerekli Malzemeler<\/h3>\n<ul>\n<li><strong>Luckfox Lyra Zero W Geli\u015ftirme Kart\u0131:<\/strong> Elbette projenizin kalbi.<\/li>\n<li><strong>microSD Kart (8GB veya \u00fczeri):<\/strong> \u0130\u015fletim sistemi ve uygulamalar\u0131n\u0131z i\u00e7in depolama alan\u0131. H\u0131zl\u0131 (Class 10 veya \u00fczeri) bir kart tercih etmek performans a\u00e7\u0131s\u0131ndan \u00f6nemlidir.<\/li>\n<li><strong>5V\/2A USB-C G\u00fc\u00e7 Kayna\u011f\u0131:<\/strong> Kart\u0131 beslemek i\u00e7in g\u00fcvenilir bir g\u00fc\u00e7 kayna\u011f\u0131. D\u00fc\u015f\u00fck kaliteli veya yetersiz g\u00fc\u00e7 kaynaklar\u0131 stabilite sorunlar\u0131na yol a\u00e7abilir.<\/li>\n<li><strong>USB-C Veri Kablosu:<\/strong> Kart\u0131 bilgisayar\u0131n\u0131za ba\u011flamak ve ilk kez g\u00f6r\u00fcnt\u00fc yazmak i\u00e7in gereklidir.<\/li>\n<li><strong>Bilgisayar:<\/strong> \u0130\u015fletim sistemi imaj\u0131n\u0131 microSD karta yazmak ve daha sonra kartla etkile\u015fim kurmak i\u00e7in.<\/li>\n<li><strong>\u0130ste\u011fe Ba\u011fl\u0131:<\/strong>\n<ul>\n<li><strong>TTL Seri Konsol Adapt\u00f6r\u00fc:<\/strong> Ba\u015flang\u0131\u00e7ta hata ay\u0131klama veya ilk \u00f6ny\u00fckleme sorunlar\u0131n\u0131 giderme i\u00e7in \u00e7ok faydal\u0131d\u0131r.<\/li>\n<li><strong>MIPI CSI Kamera Mod\u00fcl\u00fc:<\/strong> G\u00f6r\u00fcnt\u00fc i\u015fleme veya ak\u0131ll\u0131 kamera projeleri i\u00e7in.<\/li>\n<li><strong>Mini HDMI &#8211; HDMI Kablosu ve Monit\u00f6r:<\/strong> E\u011fer g\u00f6rsel bir aray\u00fcze ihtiya\u00e7 duyuyorsan\u0131z. Ancak \u00e7o\u011funlukla headless (ekrans\u0131z) modda kullan\u0131labilir.<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n<h3>\u0130lk Kurulum Ad\u0131mlar\u0131<\/h3>\n<p>\n    Luckfox Lyra Zero W&#8217;yi \u00e7al\u0131\u015ft\u0131rman\u0131n ilk ad\u0131m\u0131, kart i\u00e7in \u00f6zel olarak haz\u0131rlanm\u0131\u015f bir i\u015fletim sistemi imaj\u0131n\u0131 microSD karta yazmakt\u0131r. Bu imajlar genellikle Luckfox&#8217;un resmi GitHub sayfas\u0131ndan veya topluluk taraf\u0131ndan sa\u011flanan kaynaklardan temin edilebilir.\n<\/p>\n<ol>\n<li>\n        <strong>\u0130\u015fletim Sistemi \u0130maj\u0131n\u0131 \u0130ndirin:<\/strong> Luckfox&#8217;un resmi sitesinden veya GitHub deposundan RV1103 tabanl\u0131 Lyra Zero W i\u00e7in uygun olan i\u015fletim sistemi imaj\u0131n\u0131 indirin. Genellikle bu bir <code>.img<\/code> dosyas\u0131 olacakt\u0131r.\n    <\/li>\n<li>\n        <strong>\u0130maj\u0131 microSD Karta Yaz\u0131n:<\/strong> \u0130ndirdi\u011finiz imaj dosyas\u0131n\u0131 microSD karta yazmak i\u00e7in <a href=\"https:\/\/www.balena.io\/etcher\/\" target=\"_blank\">Balena Etcher<\/a> veya <a href=\"https:\/\/rufus.ie\/tr\/\" target=\"_blank\">Rufus<\/a> gibi bir ara\u00e7 kullanmal\u0131s\u0131n\u0131z. Bu ara\u00e7lar, imaj dosyas\u0131n\u0131 do\u011fru bir \u015fekilde kart\u0131n\u0131za kopyalaman\u0131z\u0131 ve \u00f6ny\u00fcklenebilir hale getirmenizi sa\u011flar.<\/p>\n<pre><code class=\"language-bash\">\n# Linux \u00fczerinde dd komutu ile imaj yazma (Dikkatli olun, yanl\u0131\u015f disk se\u00e7imi veri kayb\u0131na neden olabilir!)\n# \u0130lk olarak microSD kart\u0131n\u0131z\u0131n ba\u011fl\u0131 oldu\u011fu diski bulun (\u00f6rne\u011fin \/dev\/sdX)\n# sudo fdisk -l\n# Ard\u0131ndan imaj\u0131 yaz\u0131n\n# sudo dd if=\/path\/to\/your\/image.img of=\/dev\/sdX bs=4M status=progress && sync\n        <\/pre>\n<p><\/code><\/p>\n<p>\n            Etcher gibi ara\u00e7lar grafik aray\u00fcz\u00fc sundu\u011fu i\u00e7in daha kullan\u0131c\u0131 dostudur ve hatal\u0131 disk se\u00e7imi riskini azalt\u0131r.\n        <\/p>\n<\/li>\n<li>\n        <strong>microSD Kart\u0131 Luckfox Lyra Zero W'ye Tak\u0131n:<\/strong> \u0130maj\u0131 ba\u015far\u0131yla yazd\u0131ktan sonra, microSD kart\u0131 kart \u00fczerindeki yuvaya nazik\u00e7e tak\u0131n.\n    <\/li>\n<li>\n        <strong>Kart\u0131 \u00c7al\u0131\u015ft\u0131r\u0131n:<\/strong> USB-C g\u00fc\u00e7 kablosunu kart\u0131n g\u00fc\u00e7 giri\u015fine tak\u0131n. Kart\u0131n durum \u0131\u015f\u0131klar\u0131n\u0131n yand\u0131\u011f\u0131n\u0131 g\u00f6rmelisiniz. \u0130lk \u00f6ny\u00fckleme biraz zaman alabilir, sab\u0131rl\u0131 olun.\n    <\/li>\n<li>\n        <strong>\u0130lk Ba\u011flant\u0131 ve Yap\u0131land\u0131rma (Headless SSH):<\/strong><\/p>\n<p>\n            \u00c7o\u011fu durumda, Luckfox Lyra Zero W'yi bir monit\u00f6re ba\u011flamak yerine SSH \u00fczerinden headless (ekrans\u0131z) modda kullanmak tercih edilir.\n        <\/p>\n<ol>\n<li>\n                <strong>A\u011f Ba\u011flant\u0131s\u0131:<\/strong> E\u011fer imaj\u0131n\u0131z Wi-Fi a\u011f\u0131na otomatik ba\u011flanacak \u015fekilde \u00f6nceden yap\u0131land\u0131r\u0131lmad\u0131ysa, ilk \u00f6ny\u00fcklemede USB-C kablosu \u00fczerinden bilgisayar\u0131n\u0131za ba\u011flayarak veya Ethernet adapt\u00f6r\u00fc kullanarak a\u011fa ba\u011flaman\u0131z gerekebilir. Baz\u0131 imajlar USB \u00fczerinden otomatik olarak bir a\u011f aray\u00fcz\u00fc olu\u015fturabilir.\n            <\/li>\n<li>\n                <strong>IP Adresini Bulun:<\/strong> Kart\u0131n\u0131z a\u011fa ba\u011fland\u0131\u011f\u0131nda, yerel a\u011f\u0131n\u0131zdaki IP adresini bulman\u0131z gerekir. Bunu genellikle router'\u0131n\u0131z\u0131n aray\u00fcz\u00fcnden veya Nmap gibi a\u011f tarama ara\u00e7lar\u0131yla yapabilirsiniz:<\/p>\n<pre><code class=\"language-bash\">\n# Linux\/macOS \u00fczerinde\n# nmap -sn 192.168.1.0\/24  # Kendi a\u011f aral\u0131\u011f\u0131n\u0131za g\u00f6re d\u00fczenleyin\n                <\/pre>\n<p><\/code>\n            <\/li>\n<li>\n                <strong>SSH ile Ba\u011flan\u0131n:<\/strong> IP adresini bulduktan sonra, bir terminal em\u00fclat\u00f6r\u00fc (Linux\/macOS i\u00e7in yerle\u015fik terminal, Windows i\u00e7in PuTTY veya Termius) kullanarak SSH ile kart\u0131n\u0131za ba\u011flanabilirsiniz:<\/p>\n<pre><code class=\"language-bash\">\n# SSH ba\u011flant\u0131s\u0131\n# ssh root@<Luckfox_IP_Adresi>\n# Varsay\u0131lan \u015fifre genellikle \"root\" veya \"luckfox\" gibidir, imaj\u0131n\u0131z\u0131n belgelerini kontrol edin.\n                <\/pre>\n<p><\/code>\n            <\/li>\n<li>\n                <strong>\u0130lk Yap\u0131land\u0131rma:<\/strong> Ba\u011fland\u0131ktan sonra, \u015fifrenizi de\u011fi\u015ftirmek, yerel saat dilimini ayarlamak ve sistem g\u00fcncellemelerini yapmak gibi temel yap\u0131land\u0131rmalar\u0131 ger\u00e7ekle\u015ftirmeniz \u00f6nerilir.<\/p>\n<pre><code class=\"language-bash\">\n# Sistem paketlerini g\u00fcncelleme\n# apt update\n# apt upgrade\n                <\/pre>\n<p><\/code>\n            <\/li>\n<\/ol>\n<\/li>\n<\/ol>\n<p>\n    Bu ad\u0131mlar\u0131 tamamlad\u0131ktan sonra Luckfox Lyra Zero W'niz geli\u015ftirme i\u00e7in haz\u0131r hale gelecektir. Art\u0131k kart\u0131n\u0131z \u00fczerinde uygulamalar\u0131n\u0131z\u0131 \u00e7al\u0131\u015ft\u0131rmaya, sens\u00f6rleri ba\u011flamaya ve yapay zeka modellerini denemeye ba\u015flayabilirsiniz. Bu s\u00fcre\u00e7, temelleri sa\u011flam bir \u015fekilde atman\u0131z\u0131 sa\u011flayarak ileri d\u00fczey projelere ge\u00e7i\u015finizi kolayla\u015ft\u0131racakt\u0131r.\n<\/p>\n<h2>G\u00f6m\u00fcl\u00fc Yapay Zeka ile Tan\u0131\u015f\u0131n: NPU Kullanarak Nesne Alg\u0131lama Nas\u0131l Yap\u0131l\u0131r?<\/h2>\n<p>\n    Luckfox Lyra Zero W'nin en heyecan verici \u00f6zelliklerinden biri, entegre NPU (Neural Processing Unit) sayesinde do\u011frudan cihaz \u00fczerinde yapay zeka \u00e7\u0131kar\u0131mlar\u0131 yapabilme yetene\u011fidir. Bu, bulut tabanl\u0131 AI \u00e7\u00f6z\u00fcmlerine g\u00f6re daha h\u0131zl\u0131, daha g\u00fcvenli ve \u00e7o\u011fu zaman daha enerji verimli bir yakla\u015f\u0131md\u0131r. Nesne alg\u0131lama, bu yetene\u011fi sergilemek i\u00e7in harika bir ba\u015flang\u0131\u00e7 noktas\u0131d\u0131r. Peki, bu k\u00fc\u00e7\u00fck kart \u00fczerinde nesne alg\u0131lamay\u0131 nas\u0131l ger\u00e7ekle\u015ftirebiliriz?\n<\/p>\n<h3>NPU Neden \u00d6nemli?<\/h3>\n<p>\n    Geleneksel CPU'lar, genel hesaplama g\u00f6revleri i\u00e7in tasarlanm\u0131\u015ft\u0131r. Ancak, yapay zeka modellerinin yo\u011fun matris \u00e7arp\u0131m\u0131 ve toplama i\u015flemleri CPU'lar i\u00e7in olduk\u00e7a maliyetlidir. NPU'lar ise bu t\u00fcr i\u015flemleri paralel olarak ve \u00f6zel donan\u0131m h\u0131zland\u0131rmas\u0131yla ger\u00e7ekle\u015ftirmek \u00fczere optimize edilmi\u015ftir. Bu sayede, Luckfox Lyra Zero W'nin NPU'su, \u00f6rne\u011fin bir YOLO (You Only Look Once) modeli gibi karma\u015f\u0131k bir nesne alg\u0131lama modelini saniyede kare h\u0131zlar\u0131nda (FPS) \u00e7al\u0131\u015ft\u0131rabilirken, ayn\u0131 g\u00f6revi CPU \u00fczerinde \u00e7al\u0131\u015ft\u0131rmak \u00e7ok daha yava\u015f ve enerji yo\u011fun olacakt\u0131r. Bu durum, ger\u00e7ek zamanl\u0131 video analizi gerektiren uygulamalar i\u00e7in hayati \u00f6nem ta\u015f\u0131r.\n<\/p>\n<h3>Nesne Alg\u0131lama Uygulamas\u0131 Ad\u0131m Ad\u0131m<\/h3>\n<p>\n    Bu senaryoda, Luckfox Lyra Zero W'ye ba\u011fl\u0131 bir kamera mod\u00fcl\u00fcnden gelen g\u00f6r\u00fcnt\u00fcler \u00fczerinde \u00f6nceden e\u011fitilmi\u015f bir nesne alg\u0131lama modeli (\u00f6rne\u011fin MobileNet-SSD veya k\u00fc\u00e7\u00fck bir YOLO versiyonu) kullanarak insanlar\u0131, ara\u00e7lar\u0131 veya belirli nesneleri alg\u0131layaca\u011f\u0131z.\n<\/p>\n<ol>\n<li>\n        <strong>Gerekli Donan\u0131m\u0131 Ba\u011flay\u0131n:<\/strong><\/p>\n<p>\n            E\u011fer hen\u00fcz yapmad\u0131ysan\u0131z, bir MIPI CSI kamera mod\u00fcl\u00fcn\u00fc Luckfox Lyra Zero W'nin kamera portuna ba\u011flay\u0131n. Bu ba\u011flant\u0131 genellikle bir \u015ferit kablo arac\u0131l\u0131\u011f\u0131yla yap\u0131l\u0131r ve do\u011fru y\u00f6nlendirmeye dikkat etmek \u00f6nemlidir.\n        <\/p>\n<\/li>\n<li>\n        <strong>Sistemi Haz\u0131rlay\u0131n:<\/strong><\/p>\n<p>\n            Kart\u0131n\u0131z\u0131n g\u00fcncel oldu\u011fundan emin olun ve gerekli k\u00fct\u00fcphaneleri y\u00fckleyin. Luckfox SDK's\u0131 genellikle NPU'ya eri\u015fim i\u00e7in \u00f6zel k\u00fct\u00fcphaneler (\u00f6rne\u011fin Rockchip'in RKNPU veya RV1103 \u00f6zel k\u00fct\u00fcphaneleri) sa\u011flar. Ayr\u0131ca OpenCV gibi g\u00f6r\u00fcnt\u00fc i\u015fleme k\u00fct\u00fcphanelerine de ihtiyac\u0131n\u0131z olabilir.\n        <\/p>\n<pre><code class=\"language-bash\">\n# K\u00fct\u00fcphane ba\u011f\u0131ml\u0131l\u0131klar\u0131n\u0131 y\u00fckleyin (\u00d6rnek Python i\u00e7in)\n# sudo apt update\n# sudo apt install python3-pip -y\n# pip3 install opencv-python numpy\n# Luckfox\/Rockchip NPU SDK's\u0131na \u00f6zg\u00fc ad\u0131mlar burada de\u011fi\u015febilir, dok\u00fcmantasyonu kontrol edin.\n# Genellikle SDK, model d\u00f6n\u00fc\u015ft\u00fcrme ara\u00e7lar\u0131 ve \u00e7al\u0131\u015fma zaman\u0131 k\u00fct\u00fcphaneleri i\u00e7erir.\n        <\/pre>\n<p><\/code>\n    <\/li>\n<li>\n        <strong>\u00d6nceden E\u011fitilmi\u015f Modeli \u0130ndirin ve D\u00f6n\u00fc\u015ft\u00fcr\u00fcn:<\/strong><\/p>\n<p>\n            NPU'lar genellikle belirli bir formatta e\u011fitilmi\u015f modelleri (\u00f6rne\u011fin ONNX, TensorFlow Lite) do\u011frudan \u00e7al\u0131\u015ft\u0131ramaz. Rockchip NPU'lar\u0131 i\u00e7in modeller genellikle RKNN (Rockchip Neural Network) format\u0131na d\u00f6n\u00fc\u015ft\u00fcr\u00fclmelidir. Rockchip SDK's\u0131, pop\u00fcler model formatlar\u0131n\u0131 RKNN'ye d\u00f6n\u00fc\u015ft\u00fcrmek i\u00e7in ara\u00e7lar sa\u011flar.\n        <\/p>\n<p>\n            \u00d6rne\u011fin, bir MobileNet-SSD modelini ( TensorFlow Lite format\u0131nda) indirip RKNN format\u0131na d\u00f6n\u00fc\u015ft\u00fcrebilirsiniz.\n        <\/p>\n<pre><code class=\"language-python\">\n# \u00d6rnek: Model d\u00f6n\u00fc\u015ft\u00fcrme (genellikle bilgisayar\u0131n\u0131zda yap\u0131l\u0131r, kart \u00fczerinde de\u011fil)\n# from rknn.api import RKNN\n#\n# rknn = RKNN(verbose=True)\n# rknn.config(mean_values=[128, 128, 128], std_values=[128, 128, 128], channel_mean_value='0 0 0', channel_std_value='1 1 1')\n#\n# # TFLite modelini y\u00fckle\n# print('--> Loading model')\n# rknn.load_tflite(model='.\/model.tflite',\n#                 inputs=['input'],\n#                 outputs=['output'],\n#                 input_size_list=[[1, 300, 300, 3]]) # Modelinize g\u00f6re boyutlar\u0131 ayarlay\u0131n\n# print('done')\n#\n# print('--> Building model')\n# rknn.build(do_quantization=True, dataset='.\/dataset.txt') # Kuantizasyon i\u00e7in veri k\u00fcmesi gerekebilir\n# print('done')\n#\n# print('--> Export rknn model')\n# rknn.export_rknn(export_path='.\/model.rknn')\n# print('done')\n#\n# rknn.release()\n        <\/pre>\n<p><\/code><\/p>\n<p>\n            D\u00f6n\u00fc\u015ft\u00fcr\u00fclen <code>.rknn<\/code> dosyas\u0131n\u0131 microSD kart\u0131n\u0131za veya Luckfox Lyra Zero W'nin dahili depolama birimine aktar\u0131n.\n        <\/p>\n<\/li>\n<li>\n        <strong>Python ile Nesne Alg\u0131lama Kodu Yaz\u0131n:<\/strong><\/p>\n<p>\n            \u015eimdi, kamera g\u00f6r\u00fcnt\u00fcs\u00fcn\u00fc alacak, NPU \u00fczerinde \u00e7\u0131kar\u0131m yapacak ve alg\u0131lama sonu\u00e7lar\u0131n\u0131 g\u00f6sterecek bir Python beti\u011fi yazal\u0131m.\n        <\/p>\n<pre><code class=\"language-python\">\nimport cv2\nimport numpy as np\nfrom rknpu.api import RKNN # Bu k\u00fct\u00fcphane Luckfox SDK's\u0131ndan gelir ve NPU ile etkile\u015fimi sa\u011flar\n\n# Modeli y\u00fckle\nRKNN_MODEL = '.\/model.rknn'\nrknn = RKNN(verbose=True)\nprint('--> Loading rknn model')\nret = rknn.load_rknn(RKNN_MODEL)\nif ret != 0:\n    print('Load rknn model failed!')\n    exit(ret)\nprint('done')\n\nprint('--> Init runtime environment')\nret = rknn.init_runtime(core_mask=RKNN.NPU_CORE_0) # NPU \u00e7ekirde\u011fini belirtin\nif ret != 0:\n    print('Init runtime environment failed!')\n    exit(ret)\nprint('done')\n\n# Kamera ak\u0131\u015f\u0131n\u0131 ba\u015flat\ncap = cv2.VideoCapture(0) # 0, genellikle varsay\u0131lan kamera demektir. Luckfox'ta farkl\u0131 olabilir.\nif not cap.isOpened():\n    print(\"Kamera a\u00e7\u0131lamad\u0131!\")\n    exit()\n\nwhile True:\n    ret, frame = cap.read()\n    if not ret:\n        break\n\n    # G\u00f6r\u00fcnt\u00fcy\u00fc modele uygun boyuta yeniden boyutland\u0131r ve normalize et\n    # Modelinizin bekledi\u011fi giri\u015f boyutuna ve normalizasyonuna dikkat edin (\u00f6rn: 300x300, 0-1 veya -1-1 aral\u0131\u011f\u0131)\n    img_resized = cv2.resize(frame, (300, 300))\n    img_normalized = (img_resized \/ 255.0).astype(np.float32) # \u00d6rnek normalizasyon\n    img_input = img_normalized[np.newaxis, :, :, :] # Batch boyutu ekle (1, H, W, C)\n\n    # NPU \u00fczerinde \u00e7\u0131kar\u0131m yap\n    outputs = rknn.inference(inputs=[img_input])\n\n    # \u00c7\u0131kar\u0131m sonu\u00e7lar\u0131n\u0131 i\u015fle (Modelden modele de\u011fi\u015fir)\n    # outputs[0] genellikle alg\u0131lanan nesne kutular\u0131n\u0131, etiketlerini ve g\u00fcven skorlar\u0131n\u0131 i\u00e7erir.\n    # Bu k\u0131s\u0131m, kulland\u0131\u011f\u0131n\u0131z modelin \u00e7\u0131kt\u0131 format\u0131na g\u00f6re \u00f6zelle\u015ftirilmelidir.\n    # \u00d6rnek olarak, basitle\u015ftirilmi\u015f bir kutu \u00e7izimi:\n    for detection in outputs[0]: # Varsay\u0131msal \u00e7\u0131kt\u0131 format\u0131\n        score = detection[4]\n        if score > 0.5: # G\u00fcven skoru %50'den y\u00fcksekse\n            x1, y1, x2, y2 = (detection[:4] * np.array([frame.shape[1], frame.shape[0], frame.shape[1], frame.shape[0]])).astype(int)\n            cv2.rectangle(frame, (x1, y1), (x2, y2), (0, 255, 0), 2)\n            cv2.putText(frame, f'Nesne: {score:.2f}', (x1, y1 - 10), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 255, 0), 2)\n\n    # Sonucu ekranda g\u00f6ster (e\u011fer monit\u00f6r ba\u011fl\u0131ysa veya bir web ak\u0131\u015f\u0131 ile)\n    # cv2.imshow('Nesne Algilama', frame) # Sadece masa\u00fcst\u00fc ortam\u0131nda \u00e7al\u0131\u015f\u0131r\n\n    # Web ak\u0131\u015f\u0131 i\u00e7in kareleri bir HTTP sunucusuna g\u00f6nderebilirsiniz\n    # veya sadece \u00e7\u0131kt\u0131lar\u0131 bir dosyaya kaydedebilirsiniz.\n\n    if cv2.waitKey(1) & 0xFF == ord('q'):\n        break\n\ncap.release()\ncv2.destroyAllWindows()\nrknn.release()\n        <\/pre>\n<p><\/code><\/p>\n<blockquote>\n<p><strong>Uzman \u0130pucu:<\/strong> NPU performans\u0131n\u0131 en \u00fcst d\u00fczeye \u00e7\u0131karmak i\u00e7in, modelinizi kuantizasyon (nicemleme) i\u015flemiyle daha d\u00fc\u015f\u00fck hassasiyetli (\u00f6rn. INT8) formatlara d\u00f6n\u00fc\u015ft\u00fcrmeyi d\u00fc\u015f\u00fcnebilirsiniz. Bu, model boyutunu k\u00fc\u00e7\u00fclt\u00fcr ve \u00e7\u0131kar\u0131m h\u0131z\u0131n\u0131 \u00f6nemli \u00f6l\u00e7\u00fcde art\u0131r\u0131rken, \u00e7o\u011fu durumda do\u011frulukta kabul edilebilir bir d\u00fc\u015f\u00fc\u015fe neden olur. Luckfox\/Rockchip NPU ara\u00e7lar\u0131 genellikle bu \u00f6zelli\u011fi destekler.<\/p>\n<\/blockquote>\n<\/li>\n<li>\n        <strong>Uygulamay\u0131 \u00c7al\u0131\u015ft\u0131r\u0131n:<\/strong><\/p>\n<p>\n            Beti\u011fi Luckfox Lyra Zero W'ye kopyalay\u0131n ve SSH \u00fczerinden \u00e7al\u0131\u015ft\u0131r\u0131n:\n        <\/p>\n<pre><code class=\"language-bash\">\n# scp your_script.py root@<Luckfox_IP_Adresi>:\/home\/\n# ssh root@<Luckfox_IP_Adresi>\n# python3 \/home\/your_script.py\n        <\/pre>\n<p><\/code>\n    <\/li>\n<\/ol>\n<p>\n    Bu ad\u0131mlar\u0131 takip ederek, Luckfox Lyra Zero W'nin NPU'sunu kullanarak ger\u00e7ek zamanl\u0131 nesne alg\u0131lama yapabilen bir sistem kurabilirsiniz. Bu yetenek, ak\u0131ll\u0131 g\u00fcvenlik kameralar\u0131, otonom robotlar veya perakende analizi gibi bir\u00e7ok ger\u00e7ek d\u00fcnya uygulamas\u0131n\u0131n temelini olu\u015fturabilir.\n<\/p>\n<h2>Ger\u00e7ek D\u00fcnya Uygulamalar\u0131: Bir Ak\u0131ll\u0131 Kamera Sistemi Nas\u0131l Olu\u015fturulur?<\/h2>\n<p>\n    Luckfox Lyra Zero W'nin NPU \u00f6zelliklerini kullanarak nesne alg\u0131lama temellerini \u00f6\u011frendikten sonra, bu bilgiyi daha kapsaml\u0131 bir ger\u00e7ek d\u00fcnya uygulamas\u0131na d\u00f6n\u00fc\u015ft\u00fcrmenin zaman\u0131 geldi: kendi ak\u0131ll\u0131 kamera sisteminizi olu\u015fturmak. Bu t\u00fcr bir sistem, yaln\u0131zca g\u00f6r\u00fcnt\u00fc yakalamakla kalmayacak, ayn\u0131 zamanda yakalanan verileri analiz edecek ve belirli olaylara tepki verecektir. Hayal edin: evinize yakla\u015fan bir ki\u015fiyi otomatik olarak alg\u0131layan, bir bildirim g\u00f6nderen ve gerekirse g\u00f6r\u00fcnt\u00fcy\u00fc kaydeden bir sistem. \u0130\u015fte ad\u0131m ad\u0131m b\u00f6yle bir sistemi nas\u0131l kurabilece\u011finiz.\n<\/p>\n<h3>Vaka Analizi: Ak\u0131ll\u0131 G\u00fcvenlik Kameras\u0131<\/h3>\n<p>\n    Hedefimiz, Luckfox Lyra Zero W'yi kullanarak, hareket alg\u0131lad\u0131\u011f\u0131nda (\u00f6zellikle bir insan alg\u0131lad\u0131\u011f\u0131nda) kullan\u0131c\u0131ya anl\u0131k bildirim g\u00f6nderen ve olay\u0131n k\u0131sa bir videosunu kaydeden basit bir g\u00fcvenlik kameras\u0131 olu\u015fturmak.\n<\/p>\n<ol>\n<li>\n        <strong>Donan\u0131m Haz\u0131rl\u0131\u011f\u0131:<\/strong><\/p>\n<ul>\n<li><strong>Luckfox Lyra Zero W:<\/strong> Elbette ana i\u015flemcimiz.<\/li>\n<li><strong>MIPI CSI Kamera Mod\u00fcl\u00fc:<\/strong> Y\u00fcksek \u00e7\u00f6z\u00fcn\u00fcrl\u00fckl\u00fc g\u00f6r\u00fcnt\u00fc yakalama i\u00e7in.<\/li>\n<li><strong>microSD Kart:<\/strong> \u0130\u015fletim sistemi, model ve kaydedilen videolar i\u00e7in yeterli depolama (minimum 16GB \u00f6nerilir).<\/li>\n<li><strong>G\u00fc\u00e7 Kayna\u011f\u0131:<\/strong> Stabil \u00e7al\u0131\u015fma i\u00e7in 5V\/2A.<\/li>\n<li><strong>\u0130nternet Ba\u011flant\u0131s\u0131:<\/strong> Bildirim g\u00f6ndermek i\u00e7in Wi-Fi.<\/li>\n<\/ul>\n<\/li>\n<li>\n        <strong>Yaz\u0131l\u0131m Ortam\u0131n\u0131n Kurulumu:<\/strong><\/p>\n<p>\n            \u00d6nceki b\u00f6l\u00fcmde bahsedildi\u011fi gibi, Luckfox Lyra Zero W \u00fczerinde temel i\u015fletim sistemini kurun ve SSH ba\u011flant\u0131s\u0131n\u0131 sa\u011flay\u0131n. Ard\u0131ndan, a\u015fa\u011f\u0131daki k\u00fct\u00fcphaneleri y\u00fckleyin:\n        <\/p>\n<ul>\n<li><strong>OpenCV:<\/strong> G\u00f6r\u00fcnt\u00fc ve video i\u015fleme i\u00e7in.<\/li>\n<li><strong>Numpy:<\/strong> Say\u0131sal i\u015flemler i\u00e7in.<\/li>\n<li><strong>RKNPU API:<\/strong> NPU ile etkile\u015fim i\u00e7in (Luckfox SDK's\u0131 ile birlikte gelir).<\/li>\n<li><strong>Python Requests:<\/strong> HTTP istekleri g\u00f6ndererek bildirim g\u00f6ndermek i\u00e7in (iste\u011fe ba\u011fl\u0131, farkl\u0131 bildirim y\u00f6ntemleri de kullan\u0131labilir).<\/li>\n<li><strong>FFmpeg:<\/strong> Video kay\u0131t ve s\u0131k\u0131\u015ft\u0131rma i\u00e7in (genellikle Linux sistemlerinde mevcuttur).<\/li>\n<\/ul>\n<pre><code class=\"language-bash\">\n# Gerekli Python k\u00fct\u00fcphanelerini y\u00fckleyin\n# sudo apt install python3-pip -y\n# pip3 install opencv-python numpy requests\n\n# FFmpeg kurulumu (e\u011fer mevcut de\u011filse)\n# sudo apt install ffmpeg -y\n        <\/pre>\n<p><\/code>\n    <\/li>\n<li>\n        <strong>Nesne Alg\u0131lama Modelinin Entegrasyonu:<\/strong><\/p>\n<p>\n            \u00d6nceki b\u00f6l\u00fcmde d\u00f6n\u00fc\u015ft\u00fcrd\u00fc\u011f\u00fcm\u00fcz <code>.rknn<\/code> format\u0131ndaki nesne alg\u0131lama modelini (\u00f6rne\u011fin insan alg\u0131lamas\u0131 i\u00e7in e\u011fitilmi\u015f MobileNet-SSD veya YOLO-nano) Luckfox Lyra Zero W'ye aktar\u0131n. Modelin alg\u0131layabilece\u011fi etiketleri i\u00e7eren bir <code>labels.txt<\/code> dosyas\u0131na da ihtiyac\u0131n\u0131z olabilir.\n        <\/p>\n<\/li>\n<li>\n        <strong>Python Komut Dosyas\u0131n\u0131n Geli\u015ftirilmesi:<\/strong><\/p>\n<p>\n            \u015eimdi, t\u00fcm bile\u015fenleri bir araya getiren ana Python komut dosyas\u0131n\u0131 (<code>smart_cam.py<\/code>) olu\u015ftural\u0131m.\n        <\/p>\n<pre><code class=\"language-python\">\nimport cv2\nimport numpy as np\nimport time\nimport requests\nimport subprocess\nfrom rknpu.api import RKNN # NPU API\n\n# --- Yap\u0131land\u0131rma ---\nRKNN_MODEL_PATH = '.\/model.rknn'\nLABELS_PATH = '.\/labels.txt' # Modelinizin alg\u0131lad\u0131\u011f\u0131 nesne etiketleri\nCAMERA_INDEX = 0 # Varsay\u0131lan kamera indeksi (MIPI CSI i\u00e7in kontrol edin)\nCONF_THRESHOLD = 0.6 # Alg\u0131lama g\u00fcven skoru e\u015fi\u011fi\nRECORD_DURATION = 10 # Kaydedilecek video s\u00fcresi (saniye)\nNOTIFICATION_URL = \"https:\/\/api.telegram.org\/bot<YOUR_BOT_TOKEN>\/sendMessage\" # \u00d6rnek Telegram Bot API\nCHAT_ID = \"<YOUR_CHAT_ID>\" # Telegram sohbet kimli\u011finiz\n\n# --- Fonksiyonlar ---\ndef send_notification(message):\n    try:\n        data = {\"chat_id\": CHAT_ID, \"text\": message}\n        requests.post(NOTIFICATION_URL, data=data)\n        print(f\"Bildirim g\u00f6nderildi: {message}\")\n    except Exception as e:\n        print(f\"Bildirim g\u00f6nderme hatas\u0131: {e}\")\n\ndef record_video(filename, duration, camera_index):\n    print(f\"Video kayd\u0131 ba\u015flat\u0131l\u0131yor: {filename}\")\n    # FFmpeg kullanarak videoyu kaydetme\n    # Kamera indeksini ve \u00e7\u00f6z\u00fcn\u00fcrl\u00fc\u011f\u00fc sisteminize g\u00f6re ayarlay\u0131n\n    command = [\n        'ffmpeg',\n        '-f', 'v4l2',          # Video for Linux 2 input\n        '-framerate', '30',    # Kare h\u0131z\u0131\n        '-video_size', '1280x720', # Kamera \u00e7\u00f6z\u00fcn\u00fcrl\u00fc\u011f\u00fc\n        '-i', f'\/dev\/video{camera_index}', # Kamera cihaz\u0131 (farkl\u0131 olabilir)\n        '-c:v', 'libx264',     # Video codec\n        '-preset', 'ultrafast',# H\u0131zl\u0131 s\u0131k\u0131\u015ft\u0131rma\n        '-t', str(duration),   # Kay\u0131t s\u00fcresi\n        '-y',                  # Var olan dosyay\u0131 \u00fczerine yaz\n        filename\n    ]\n    try:\n        subprocess.run(command, check=True)\n        print(f\"Video kayd\u0131 tamamland\u0131: {filename}\")\n    except subprocess.CalledProcessError as e:\n        print(f\"Video kayd\u0131 hatas\u0131: {e}\")\n\n# --- Ana Program ---\ndef main():\n    # 1. RKNN Modelini Y\u00fckle\n    rknn = RKNN(verbose=True)\n    ret = rknn.load_rknn(RKNN_MODEL_PATH)\n    if ret != 0:\n        print('RKNN modeli y\u00fcklenemedi!')\n        return\n\n    ret = rknn.init_runtime(core_mask=RKNN.NPU_CORE_0)\n    if ret != 0:\n        print('RKNN \u00e7al\u0131\u015fma zaman\u0131 ba\u015flat\u0131lamad\u0131!')\n        return\n\n    with open(LABELS_PATH, 'r') as f:\n        labels = [line.strip() for line in f.readlines()]\n\n    # 2. Kamera Ak\u0131\u015f\u0131n\u0131 Ba\u015flat\n    cap = cv2.VideoCapture(CAMERA_INDEX)\n    if not cap.isOpened():\n        print(\"Kamera a\u00e7\u0131lamad\u0131! L\u00fctfen CAMERA_INDEX veya kamera ba\u011flant\u0131s\u0131n\u0131 kontrol edin.\")\n        return\n\n    last_detection_time = 0\n    record_active = False\n\n    print(\"Ak\u0131ll\u0131 Kamera Sistemi Ba\u015flat\u0131ld\u0131...\")\n    while True:\n        ret, frame = cap.read()\n        if not ret:\n            print(\"Kareden okuma hatas\u0131!\")\n            break\n\n        # G\u00f6r\u00fcnt\u00fcy\u00fc modele uygun boyuta yeniden boyutland\u0131r ve normalize et\n        img_resized = cv2.resize(frame, (300, 300)) # Modelinizin bekledi\u011fi boyut\n        img_normalized = (img_resized \/ 255.0 - 0.5) * 2.0 # \u00d6rnek normalizasyon (-1 ile 1 aras\u0131)\n        img_input = img_normalized.astype(np.float32)[np.newaxis, :, :, :]\n\n        # 3. NPU ile \u00c7\u0131kar\u0131m Yap\n        outputs = rknn.inference(inputs=[img_input])\n        detections = outputs[0] # Model \u00e7\u0131kt\u0131n\u0131z\u0131n format\u0131na g\u00f6re bu k\u0131sm\u0131 \u00f6zelle\u015ftirin\n\n        human_detected = False\n        for detection in detections:\n            # \u00d6rnek \u00e7\u0131kt\u0131 format\u0131: [x1, y1, x2, y2, score, class_id]\n            score = detection[4]\n            class_id = int(detection[5])\n            \n            if score > CONF_THRESHOLD and labels[class_id] == 'person': # \"person\" etiketi i\u00e7in kontrol\n                x1, y1, x2, y2 = (detection[:4] * np.array([frame.shape[1], frame.shape[0], frame.shape[1], frame.shape[0]])).astype(int)\n                cv2.rectangle(frame, (x1, y1), (x2, y2), (0, 255, 0), 2)\n                cv2.putText(frame, f'{labels[class_id]}: {score:.2f}', (x1, y1 - 10), cv2.FONT_HERSHEY_SIMPLEX, 0.5, (0, 255, 0), 2)\n                human_detected = True\n\n        # 4. Olay Y\u00f6netimi (Bildirim ve Kay\u0131t)\n        current_time = time.time()\n        if human_detected and not record_active:\n            if current_time - last_detection_time > 30: # Son bildirimden 30 saniye sonra tekrar bildirim g\u00f6nder\n                send_notification(\"Hareket alg\u0131land\u0131! Bir ki\u015fi g\u00f6r\u00fcld\u00fc.\")\n                filename = f\"detection_{int(current_time)}.mp4\"\n                # Video kayd\u0131n\u0131 ayr\u0131 bir i\u015flemde ba\u015flat\n                import threading\n                record_thread = threading.Thread(target=record_video, args=(filename, RECORD_DURATION, CAMERA_INDEX))\n                record_thread.start()\n                record_active = True\n                last_detection_time = current_time\n        elif not human_detected and record_active and not record_thread.is_alive():\n             record_active = False # Kay\u0131t tamamland\u0131 ve insan yoksa durumu s\u0131f\u0131rla\n\n        # E\u011fer bir monit\u00f6r ba\u011fl\u0131ysa g\u00f6r\u00fcnt\u00fcy\u00fc g\u00f6ster\n        # cv2.imshow('Ak\u0131ll\u0131 Guvenlik Kamerasi', frame)\n        # if cv2.waitKey(1) & 0xFF == ord('q'):\n        #     break\n\n    cap.release()\n    # cv2.destroyAllWindows()\n    rknn.release()\n\nif __name__ == \"__main__\":\n    main()\n        <\/pre>\n<p><\/code><\/p>\n<p>\n            <strong>Not:<\/strong> Bu kod \u00f6rne\u011fi, varsay\u0131msal bir model \u00e7\u0131kt\u0131s\u0131 ve Telegram API kullan\u0131m\u0131 i\u00e7erir. Kendi modelinizin \u00e7\u0131kt\u0131 format\u0131na ve tercih etti\u011finiz bildirim mekanizmas\u0131na (e-posta, di\u011fer anl\u0131k mesajla\u015fma servisleri, MQTT vb.) g\u00f6re uyarlaman\u0131z gerekecektir. Telegram botu i\u00e7in <code><YOUR_BOT_TOKEN><\/code> ve <code><YOUR_CHAT_ID><\/code> k\u0131s\u0131mlar\u0131n\u0131 kendi bilgilerinizle doldurman\u0131z gerekmektedir. FFmpeg komutundaki <code>\/dev\/video{camera_index}<\/code> yolu da sisteminize g\u00f6re farkl\u0131l\u0131k g\u00f6sterebilir (\u00f6rne\u011fin <code>\/dev\/video0<\/code> veya <code>\/dev\/video1<\/code>).\n        <\/p>\n<\/li>\n<li>\n        <strong>Uygulamay\u0131 \u00c7al\u0131\u015ft\u0131rma ve Test Etme:<\/strong><\/p>\n<p>\n            Komut dosyas\u0131n\u0131 Luckfox Lyra Zero W'ye aktar\u0131n ve \u00e7al\u0131\u015ft\u0131r\u0131n. Kameran\u0131n g\u00f6r\u00fc\u015f alan\u0131nda bir hareket (veya ki\u015fi) alg\u0131land\u0131\u011f\u0131nda, bildirim almal\u0131 ve bir video kayd\u0131n\u0131n ba\u015flad\u0131\u011f\u0131n\u0131 g\u00f6rmelisiniz.\n        <\/p>\n<pre><code class=\"language-bash\">\n# scp smart_cam.py labels.txt model.rknn root@<Luckfox_IP_Adresi>:\/home\/smart_cam\/\n# ssh root@<Luckfox_IP_Adresi>\n# cd \/home\/smart_cam\/\n# python3 smart_cam.py\n        <\/pre>\n<p><\/code>\n    <\/li>\n<\/ol>\n<p>\n    Bu ak\u0131ll\u0131 kamera sistemi, Luckfox Lyra Zero W'nin AI yeteneklerini ger\u00e7ek d\u00fcnya sorunlar\u0131n\u0131 \u00e7\u00f6zmek i\u00e7in nas\u0131l kullan\u0131labilece\u011finin harika bir \u00f6rne\u011fidir. Benzer mant\u0131kla, farkl\u0131 sens\u00f6r verilerini birle\u015ftirerek (\u00f6rne\u011fin s\u0131cakl\u0131k, nem) ve farkl\u0131 AI modelleri kullanarak (\u00f6rne\u011fin ses tan\u0131ma) \u00e7ok daha karma\u015f\u0131k ve ak\u0131ll\u0131 IoT cihazlar\u0131 tasarlayabilirsiniz. Bu proje, hem donan\u0131m hem de yaz\u0131l\u0131m entegrasyonu becerilerinizi geli\u015ftirmeniz i\u00e7in sa\u011flam bir temel sunar.\n<\/p>\n<h3>Mobil Uyumlu Aray\u00fczler Geli\u015ftirmek: Basit Bir Web Sunucusu Kurulumu<\/h3>\n<p>\n    Ak\u0131ll\u0131 kamera sisteminiz art\u0131k olaylar\u0131 alg\u0131lay\u0131p bildirim g\u00f6nderebiliyor. Peki ya kamera beslemesini canl\u0131 olarak izlemek veya sistem ayarlar\u0131n\u0131 uzaktan de\u011fi\u015ftirmek isterseniz? \u0130\u015fte burada mobil uyumlu bir web aray\u00fcz\u00fc devreye giriyor. Luckfox Lyra Zero W \u00fczerinde basit bir web sunucusu kurarak, projenizi bir web taray\u0131c\u0131s\u0131 \u00fczerinden kontrol edebilir ve izleyebilirsiniz. Flask gibi hafif bir Python web \u00e7er\u00e7evesi bu i\u015f i\u00e7in olduk\u00e7a uygundur.\n<\/p>\n<h4>Flask ile Basit Bir Web Sunucusu Kurulumu<\/h4>\n<p>\n    Flask, Python ile web uygulamalar\u0131 geli\u015ftirmek i\u00e7in minimalist ve esnek bir mikro-\u00e7er\u00e7evedir. Luckfox Lyra Zero W \u00fczerinde de kolayca \u00e7al\u0131\u015ft\u0131r\u0131labilir.\n<\/p>\n<ol>\n<li>\n        <strong>Flask'\u0131 Y\u00fckleyin:<\/strong><\/p>\n<p>\n            Luckfox Lyra Zero W \u00fczerinde SSH ile ba\u011flan\u0131p Flask'\u0131 pip ile y\u00fckleyin:\n        <\/p>\n<pre><code class=\"language-bash\">\n# pip3 install Flask\n        <\/pre>\n<p><\/code>\n    <\/li>\n<li>\n        <strong>Web Sunucusu Kodu (<code>app.py<\/code>):<\/strong><\/p>\n<p>\n            Basit bir Flask uygulamas\u0131 olu\u015ftural\u0131m. Bu uygulama, ana sayfada \"Merhaba Luckfox!\" yazacak ve bir de kamera durumunu g\u00f6sterecek bir sayfa i\u00e7erecek.\n        <\/p>\n<pre><code class=\"language-python\">\nfrom flask import Flask, render_template_string, Response\nimport cv2\nimport time\n\napp = Flask(__name__)\n\n# Basit bir HTML \u015fablonu\nHTML_TEMPLATE = '''\n<!DOCTYPE html><body>\n    <div class=\"container\">\n        <h2>Luckfox Lyra Zero W Ak\u0131ll\u0131 Kamera Aray\u00fcz\u00fc<\/h2>\n        <p>Cihaz\u0131n\u0131z\u0131n durumu ve canl\u0131 kamera ak\u0131\u015f\u0131 a\u015fa\u011f\u0131dad\u0131r.<\/p>\n        <img decoding=\"async\" src=\"{{ url_for('video_feed') }}\" class=\"video-feed\" alt=\"Canl\u0131 Kamera Ak\u0131\u015f\u0131\">\n        <button class=\"button\" onclick=\"alert('Ayarlar yak\u0131nda eklenecek!')\">Ayarlar\u0131 De\u011fi\u015ftir<\/button>\n        <p>Son g\u00fcncelleme: {{ current_time }}<\/p>\n    <\/div>\n<\/body>\n<\/html>\n'''\n\n# Kamera ak\u0131\u015f\u0131 i\u00e7in generator fonksiyonu\ndef generate_frames():\n    camera = cv2.VideoCapture(0)  # Kamera indeksi Luckfox'ta farkl\u0131 olabilir (\u00f6rn: 0 veya 1)\n    if not camera.isOpened():\n        print(\"Web aray\u00fcz\u00fc i\u00e7in kamera a\u00e7\u0131lamad\u0131!\")\n        return\n\n    while True:\n        success, frame = camera.read()\n        if not success:\n            break\n        else:\n            ret, buffer = cv2.imencode('.jpg', frame)\n            frame = buffer.tobytes()\n            yield (b'--frame\\r\\n'\n                   b'Content-Type: image\/jpeg\\r\\n\\r\\n' + frame + b'\\r\\n')\n    camera.release()\n\n@app.route('\/')\ndef index():\n    return render_template_string(HTML_TEMPLATE, current_time=time.strftime('%H:%M:%S'))\n\n@app.route('\/video_feed')\ndef video_feed():\n    # Canl\u0131 video ak\u0131\u015f\u0131 i\u00e7in \u00f6zel bir endpoint\n    return Response(generate_frames(), mimetype='multipart\/x-mixed-replace; boundary=frame')\n\nif __name__ == '__main__':\n    # T\u00fcm a\u011f aray\u00fczlerinden eri\u015filebilir hale getir\n    app.run(host='0.0.0.0', port=5000)\n        <\/pre>\n<p><\/code>\n    <\/li>\n<li>\n        <strong>Web Sunucusunu \u00c7al\u0131\u015ft\u0131r\u0131n:<\/strong><\/p>\n<p>\n            Kodu Luckfox Lyra Zero W'ye aktar\u0131n ve \u00e7al\u0131\u015ft\u0131r\u0131n:\n        <\/p>\n<pre><code class=\"language-bash\">\n# scp app.py root@<Luckfox_IP_Adresi>:\/home\/\n# ssh root@<Luckfox_IP_Adresi>\n# python3 \/home\/app.py\n        <\/pre>\n<p><\/code><\/p>\n<p>\n            Sunucu \u00e7al\u0131\u015fmaya ba\u015flad\u0131ktan sonra, a\u011f\u0131n\u0131zdaki herhangi bir cihazdan (telefon, tablet, bilgisayar) web taray\u0131c\u0131n\u0131z\u0131 a\u00e7\u0131n ve Luckfox Lyra Zero W'nin IP adresini ve port numaras\u0131n\u0131 (varsay\u0131lan olarak <code>5000<\/code>) girin: <code>http:\/\/<Luckfox_IP_Adresi>:5000<\/code>.\n        <\/p>\n<\/li>\n<\/ol>\n<p>\n    Art\u0131k, ak\u0131ll\u0131 kameran\u0131z\u0131n canl\u0131 g\u00f6r\u00fcnt\u00fcs\u00fcn\u00fc mobil cihaz\u0131n\u0131zdan veya bilgisayar\u0131n\u0131zdan izleyebilirsiniz. CSS i\u00e7indeki <code>@media (max-width: 600px)<\/code> kural\u0131, taray\u0131c\u0131 penceresinin geni\u015fli\u011fi 600 pikselin alt\u0131na d\u00fc\u015ft\u00fc\u011f\u00fcnde belirli stil de\u011fi\u015fikliklerinin (\u00f6rne\u011fin, d\u00fc\u011fmelerin tam geni\u015fli\u011fi kaplamas\u0131) uygulanmas\u0131n\u0131 sa\u011flayarak aray\u00fcz\u00fcn mobil cihazlarda daha iyi g\u00f6r\u00fcnmesini sa\u011flar. Bu, responsive (duyarl\u0131) tasar\u0131m\u0131n basit bir \u00f6rne\u011fidir. Flask uygulaman\u0131z\u0131 daha da geli\u015ftirerek, kamera ayarlar\u0131n\u0131 de\u011fi\u015ftirme, alg\u0131lama e\u015fiklerini ayarlama veya ge\u00e7mi\u015f olaylar\u0131 g\u00f6r\u00fcnt\u00fcleme gibi \u00f6zellikler ekleyebilirsiniz. Bu, Luckfox Lyra Zero W tabanl\u0131 IoT projelerinize kullan\u0131c\u0131 dostu bir aray\u00fcz eklemenin etkili bir yoludur.\n<\/p>\n<h2>Geli\u015fmi\u015f Optimizasyonlar ve \u0130pu\u00e7lar\u0131: Performans\u0131 Nas\u0131l Art\u0131r\u0131rs\u0131n\u0131z?<\/h2>\n<p>\n    Luckfox Lyra Zero W gibi s\u0131n\u0131rl\u0131 kaynaklara sahip g\u00f6m\u00fcl\u00fc sistemlerde \u00e7al\u0131\u015f\u0131rken, performans\u0131 optimize etmek ve enerji verimlili\u011fini sa\u011flamak kritik \u00f6neme sahiptir. Projeleriniz daha karma\u015f\u0131k hale geldik\u00e7e veya daha uzun s\u00fcre kesintisiz \u00e7al\u0131\u015fmas\u0131 gerekti\u011finde, bu optimizasyon teknikleri size \u00f6nemli avantajlar sa\u011flayacakt\u0131r. \u0130\u015fte deneyimli kullan\u0131c\u0131lar i\u00e7in baz\u0131 ipu\u00e7lar\u0131 ve p\u00fcf noktalar\u0131:\n<\/p>\n<h3>1. NPU Kullan\u0131m\u0131n\u0131 Maksimuma \u00c7\u0131karma<\/h3>\n<p>\n    NPU, Luckfox Lyra Zero W'nin y\u0131ld\u0131z \u00f6zelli\u011fidir. Yapay zeka modellerinizi NPU \u00fczerinde \u00e7al\u0131\u015ft\u0131rmak, CPU'nun y\u00fck\u00fcn\u00fc azalt\u0131r ve hem h\u0131z hem de enerji verimlili\u011fi a\u00e7\u0131s\u0131ndan \u00f6nemli kazan\u00e7lar sa\u011flar.\n<\/p>\n<ul>\n<li>\n        <strong>Model Kuantizasyonu (Quantization):<\/strong> Modellerinizi INT8 gibi daha d\u00fc\u015f\u00fck hassasiyetli formatlara kuantize etmek, NPU \u00fczerinde daha h\u0131zl\u0131 \u00e7\u0131kar\u0131m yapmalar\u0131n\u0131 sa\u011flar ve bellek ayak izini k\u00fc\u00e7\u00fclt\u00fcr. Rockchip'in RKNN Toolkit'i genellikle bu \u00f6zelli\u011fi destekler. Kuantizasyon s\u0131ras\u0131nda do\u011fruluk kayb\u0131n\u0131 minimize etmek i\u00e7in bir kalibrasyon veri seti kullanmak \u00f6nemlidir.\n    <\/li>\n<li>\n        <strong>Model Se\u00e7imi:<\/strong> Daha k\u00fc\u00e7\u00fck ve daha verimli AI modellerini (\u00f6rn. MobileNet, EfficientNet'in daha k\u00fc\u00e7\u00fck varyantlar\u0131, YOLO-nano) tercih edin. B\u00fcy\u00fck modeller, NPU'nun performans s\u0131n\u0131rlar\u0131n\u0131 zorlayabilir.\n    <\/li>\n<li>\n        <strong>\u00d6n \u0130\u015fleme ve Son \u0130\u015fleme:<\/strong> M\u00fcmk\u00fcnse, modelinize beslemeden \u00f6nce ve NPU'dan \u00e7\u0131kt\u0131 ald\u0131ktan sonra yap\u0131lan \u00f6n i\u015fleme ve son i\u015fleme ad\u0131mlar\u0131n\u0131 CPU yerine NPU'ya veya GPU'ya (e\u011fer varsa ve destekleniyorsa) devretmeye \u00e7al\u0131\u015f\u0131n. Ancak Luckfox Lyra Zero W'de genellikle bu ad\u0131mlar CPU \u00fczerinde kal\u0131r, bu y\u00fczden bu k\u0131s\u0131mlar\u0131n Python\/OpenCV optimizasyonlar\u0131yla h\u0131zl\u0131 olmas\u0131n\u0131 sa\u011flay\u0131n.\n    <\/li>\n<\/ul>\n<h3>2. Bellek Y\u00f6netimi ve S\u00fcre\u00e7 Optimizasyonu<\/h3>\n<p>\n    G\u00f6m\u00fcl\u00fc sistemlerde RAM miktar\u0131 s\u0131n\u0131rl\u0131d\u0131r. Belle\u011fi verimli kullanmak, uygulaman\u0131z\u0131n stabil \u00e7al\u0131\u015fmas\u0131n\u0131 sa\u011flar.\n<\/p>\n<ul>\n<li>\n        <strong>Python Bellek Ayak \u0130zi:<\/strong> Python uygulamalar\u0131 bazen beklendi\u011finden daha fazla bellek t\u00fcketebilir. Gerekli olmayan k\u00fct\u00fcphaneleri dahil etmeyin. B\u00fcy\u00fck veri yap\u0131lar\u0131 veya resim tamponlar\u0131 \u00fczerinde \u00e7al\u0131\u015f\u0131rken, referanslar\u0131 do\u011fru \u015fekilde y\u00f6neterek bellek s\u0131z\u0131nt\u0131lar\u0131n\u0131 \u00f6nleyin. Python'\u0131n <code>gc<\/code> mod\u00fcl\u00fc ile \u00e7\u00f6p toplama s\u00fcrecini manuel olarak tetikleyebilir veya bellek profilleyici ara\u00e7lar\u0131 kullanarak s\u0131z\u0131nt\u0131lar\u0131 tespit edebilirsiniz.\n    <\/li>\n<li>\n        <strong>Arka Plan S\u00fcre\u00e7leri:<\/strong> \u00c7al\u0131\u015fan gereksiz arka plan servislerini ve s\u00fcre\u00e7lerini devre d\u0131\u015f\u0131 b\u0131rak\u0131n. Linux sistemlerde <code>systemctl disable <servis_ad\u0131><\/code> komutuyla bunu yapabilirsiniz.\n    <\/li>\n<li>\n        <strong>Takas Alan\u0131 (Swap Space):<\/strong> \u00c7ok k\u0131s\u0131tl\u0131 bellek ko\u015fullar\u0131nda, bir miktar takas alan\u0131 (swap space) yap\u0131land\u0131rmak sistemin \u00e7\u00f6kmesini \u00f6nleyebilir. Ancak, microSD kart\u0131n \u00f6mr\u00fcn\u00fc k\u0131saltabilece\u011fi i\u00e7in uzun vadede \u00e7\u00f6z\u00fcm de\u011fildir; ana odak noktas\u0131 bellek kullan\u0131m\u0131n\u0131 azaltmak olmal\u0131d\u0131r.\n    <\/li>\n<\/ul>\n<h3>3. G\u00fc\u00e7 T\u00fcketimi Y\u00f6netimi<\/h3>\n<p>\n    \u00d6zellikle batarya ile \u00e7al\u0131\u015fan uygulamalar i\u00e7in g\u00fc\u00e7 t\u00fcketimi \u00e7ok \u00f6nemlidir.\n<\/p>\n<ul>\n<li>\n        <strong>Saat H\u0131z\u0131n\u0131 D\u00fc\u015f\u00fcrme:<\/strong> RV1103 SoC'nin CPU ve NPU saat h\u0131zlar\u0131n\u0131 dinamik olarak ayarlayabilirsiniz. \u0130\u015f y\u00fck\u00fc azald\u0131\u011f\u0131nda saat h\u0131z\u0131n\u0131 d\u00fc\u015f\u00fcrmek, g\u00fc\u00e7 t\u00fcketimini \u00f6nemli \u00f6l\u00e7\u00fcde azalt\u0131r. Linux'ta CPU frekans\u0131 genellikle <code>cpufreq<\/code> servisleri \u00fczerinden y\u00f6netilebilir.\n    <\/li>\n<li>\n        <strong>Periferalleri Kapatma:<\/strong> Kullan\u0131lmayan Wi-Fi, Bluetooth veya MIPI CSI gibi \u00e7evre birimlerini devre d\u0131\u015f\u0131 b\u0131rakmak g\u00fc\u00e7 tasarrufu sa\u011flar. \u00d6rne\u011fin, Wi-Fi'\u0131 s\u00fcrekli a\u00e7\u0131k tutmak yerine, sadece veri g\u00f6nderirken aktifle\u015ftirebilirsiniz.\n    <\/li>\n<li>\n        <strong>Uyku Modlar\u0131:<\/strong> Cihaz\u0131n uzun s\u00fcre bo\u015fta kalaca\u011f\u0131 durumlarda, daha derin uyku modlar\u0131na ge\u00e7i\u015f yapmak g\u00fc\u00e7 t\u00fcketimini minimuma indirebilir. Ancak bu, genellikle daha karma\u015f\u0131k yaz\u0131l\u0131m entegrasyonu gerektirir.\n    <\/li>\n<\/ul>\n<h3>4. Uzaktan Hata Ay\u0131klama ve Geli\u015ftirme<\/h3>\n<p>\n    G\u00f6m\u00fcl\u00fc sistemlerde g\u00f6rsel bir aray\u00fcz olmadan geli\u015ftirme yapmak zorlay\u0131c\u0131 olabilir.\n<\/p>\n<ul>\n<li>\n        <strong>SSH ve Terminal Multiplexer:<\/strong> SSH ba\u011flant\u0131s\u0131 temeldir. <code>tmux<\/code> veya <code>screen<\/code> gibi terminal multiplexer'lar, tek bir SSH oturumu \u00fczerinden birden fazla terminal oturumunu y\u00f6netmenizi sa\u011flar, bu da uzun s\u00fcreli i\u015flemler veya arka plan g\u00f6revleri i\u00e7in idealdir.\n    <\/li>\n<li>\n        <strong>Remote Debugging:<\/strong> VS Code gibi IDE'ler, uzaktan hata ay\u0131klama (remote debugging) eklentileri sunar. Bu, Python kodunuzu do\u011frudan Luckfox Lyra Zero W \u00fczerinde ad\u0131m ad\u0131m \u00e7al\u0131\u015ft\u0131rman\u0131za, de\u011fi\u015fkenleri incelemenize ve hatalar\u0131 daha h\u0131zl\u0131 bulman\u0131za olanak tan\u0131r.\n    <\/li>\n<li>\n        <strong>G\u00fcnl\u00fck Kay\u0131tlar\u0131 (Logging):<\/strong> Uygulamalar\u0131n\u0131zda kapsaml\u0131 g\u00fcnl\u00fck kayd\u0131 mekanizmalar\u0131 kullan\u0131n. Loglar\u0131 bir dosyaya yazmak veya uzak bir sunucuya g\u00f6ndermek, cihaz \u00fczerinde do\u011frudan g\u00f6zlem yapmadan sorunlar\u0131 te\u015fhis etmenize yard\u0131mc\u0131 olur.\n    <\/li>\n<\/ul>\n<h3>5. \u00c7apraz Derleme (Cross-Compilation)<\/h3>\n<p>\n    Karma\u015f\u0131k C\/C++ projeleriniz varsa, bunlar\u0131 do\u011frudan Luckfox Lyra Zero W \u00fczerinde derlemek \u00e7ok zaman alabilir ve kart\u0131n kaynaklar\u0131n\u0131 zorlayabilir.\n<\/p>\n<ul>\n<li>\n        <strong>Toolchain Kullan\u0131m\u0131:<\/strong> Geli\u015ftirme ortam\u0131n\u0131zda (\u00f6rn. masa\u00fcst\u00fc bilgisayar\u0131n\u0131zda) Luckfox Lyra Zero W'nin ARM Cortex-A7 i\u015flemcisi i\u00e7in bir \u00e7apraz derleme toolchain'i kurarak kodunuzu daha h\u0131zl\u0131 derleyebilir ve ard\u0131ndan kart\u0131n\u0131za aktarabilirsiniz. Bu, \u00f6zellikle b\u00fcy\u00fck C\/C++ tabanl\u0131 k\u00fct\u00fcphaneler veya uygulamalar i\u00e7in zaman kazand\u0131r\u0131r.\n    <\/li>\n<\/ul>\n<blockquote>\n<p><strong>Uzman \u0130pucu:<\/strong> A\u011f performans\u0131 kritikse, Wi-Fi yerine USB-Ethernet adapt\u00f6r\u00fc kullanarak kablolu ba\u011flant\u0131y\u0131 tercih edebilirsiniz. Bu, hem daha stabil bir ba\u011flant\u0131 sa\u011flar hem de Wi-Fi'\u0131n s\u00fcrekli tarama ve iletim y\u00fck\u00fcnden kaynaklanan enerji t\u00fcketimini azalt\u0131r. Ayr\u0131ca, NPU'nun tam performans\u0131n\u0131 g\u00f6rebilmek i\u00e7in modelinizi farkl\u0131 kuantizasyon seviyelerinde (FP32, FP16, INT8) test ederek en iyi dengeyi bulmaya \u00e7al\u0131\u015f\u0131n.<\/p>\n<\/blockquote>\n<p>\n    Bu geli\u015fmi\u015f optimizasyon ve hata ay\u0131klama teknikleri, Luckfox Lyra Zero W ile geli\u015ftirdi\u011finiz projelerin sadece \u00e7al\u0131\u015fmas\u0131n\u0131 de\u011fil, ayn\u0131 zamanda verimli, g\u00fcvenilir ve uzun \u00f6m\u00fcrl\u00fc olmas\u0131n\u0131 sa\u011flayacakt\u0131r. G\u00f6m\u00fcl\u00fc sistemlerde her bayt ve her watt \u00f6nemlidir, bu y\u00fczden bu detaylara dikkat etmek uzun vadede size b\u00fcy\u00fck faydalar sa\u011flayacakt\u0131r.\n<\/p>\n<h2>Sonu\u00e7 ve Gelecek Perspektifleri<\/h2>\n<p>\n    Luckfox Lyra Zero W ile \u00e7\u0131kt\u0131\u011f\u0131m\u0131z bu teknik yolculukta, k\u00fc\u00e7\u00fck boyutuna ra\u011fmen ne kadar b\u00fcy\u00fck potansiyeller ta\u015f\u0131d\u0131\u011f\u0131n\u0131 hep birlikte ke\u015ffettik. Temel kurulum ad\u0131mlar\u0131ndan g\u00f6m\u00fcl\u00fc yapay zeka ile nesne alg\u0131lamaya, ak\u0131ll\u0131 bir kamera sistemi kurmaktan mobil uyumlu web aray\u00fczleri geli\u015ftirmeye ve performans\u0131 optimize etmeye kadar bir\u00e7ok konuyu ele ald\u0131k. Bu kart, \u00f6zellikle entegre NPU'su sayesinde, geleneksel mikrodenetleyicilerin \u00f6tesine ge\u00e7erek yapay zeka yeteneklerini kenar bili\u015fime ta\u015f\u0131yor ve bu da onu AIoT projeleri i\u00e7in e\u015fsiz bir se\u00e7enek haline getiriyor.\n<\/p>\n<p>\n    Luckfox Lyra Zero W, uygun maliyeti ve zengin \u00f6zellikleriyle hem hobiler hem de profesyoneller i\u00e7in eri\u015filebilir bir platform sunuyor. Ger\u00e7ek zamanl\u0131 veri analizi, d\u00fc\u015f\u00fck gecikmeli tepki s\u00fcreleri ve art\u0131r\u0131lm\u0131\u015f gizlilik gibi avantajlar sayesinde, ak\u0131ll\u0131 ev otomasyonundan end\u00fcstriyel otomasyona, ak\u0131ll\u0131 \u015fehir uygulamalar\u0131ndan giyilebilir teknolojilere kadar bir\u00e7ok alanda inovatif \u00e7\u00f6z\u00fcmler geli\u015ftirmek m\u00fcmk\u00fcn. \u00d6rne\u011fin, g\u00fcr\u00fclt\u00fc azaltma \u00f6zellikli ak\u0131ll\u0131 mikrofonlar, y\u00fcz tan\u0131ma \u00f6zellikli eri\u015fim kontrol sistemleri, \u00fcretim hatlar\u0131nda kalite kontrol\u00fc yapan g\u00f6rsel denetim sistemleri gibi projeler bu kart\u0131n yetenekleri dahilindedir.\n<\/p>\n<p>\n    G\u00f6m\u00fcl\u00fc sistemler ve yapay zeka d\u00fcnyas\u0131 h\u0131zla geli\u015fiyor. Luckfox Lyra Zero W gibi kartlar, bu teknolojik ilerlemelerin \u00f6n saflar\u0131nda yer alarak, fikirleri h\u0131zl\u0131ca prototipleme ve ger\u00e7ek \u00fcr\u00fcnlere d\u00f6n\u00fc\u015ft\u00fcrme konusunda geli\u015ftiricilere g\u00fc\u00e7 veriyor. Bu rehber, ba\u015flang\u0131\u00e7 i\u00e7in sa\u011flam bir temel olu\u015ftururken, as\u0131l potansiyelin sizin yarat\u0131c\u0131l\u0131\u011f\u0131n\u0131zda yatt\u0131\u011f\u0131n\u0131 unutmay\u0131n. Edindi\u011finiz bu bilgilerle, kendi \u00f6zg\u00fcn projelerinizi hayata ge\u00e7irmek ve gelece\u011fin ak\u0131ll\u0131 cihazlar\u0131n\u0131 tasarlamak i\u00e7in gerekli donan\u0131ma ve bilgiye sahipsiniz. Ke\u015ffetmeye, denemeye ve \u00f6\u011frenmeye devam edin!\n<\/p>\n<h3>S\u0131k\u00e7a Sorulan Sorular (SSS)<\/h3>\n<dl>\n<dt>Luckfox Lyra Zero W ne kadar g\u00fc\u00e7 t\u00fcketir?<\/dt>\n<dd>\u00c7al\u0131\u015fma y\u00fck\u00fcne ba\u011fl\u0131 olarak g\u00fc\u00e7 t\u00fcketimi de\u011fi\u015fmekle birlikte, genellikle 5V ile beslenir ve tam y\u00fck alt\u0131nda ortalama 0.5A - 1.5A aras\u0131nda bir ak\u0131m \u00e7ekebilir. D\u00fc\u015f\u00fck g\u00fc\u00e7 modlar\u0131nda bu de\u011ferler \u00e7ok daha a\u015fa\u011f\u0131lara inebilir.<\/dd>\n<dt>Hangi i\u015fletim sistemlerini destekler?<\/dt>\n<dd>Luckfox Lyra Zero W, genellikle Linux tabanl\u0131 (\u00f6rn. Buildroot, Yocto veya \u00f6zelle\u015ftirilmi\u015f Debian\/Ubuntu varyantlar\u0131) i\u015fletim sistemlerini destekler. Rockchip'in RV1103 SoC'si i\u00e7in \u00f6zel olarak optimize edilmi\u015f imajlar mevcuttur.<\/dd>\n<dt>AI modellerini kendim e\u011fitip kullanabilir miyim?<\/dt>\n<dd>Evet, AI modellerini (TensorFlow, PyTorch gibi pop\u00fcler \u00e7er\u00e7evelerde) kendi bilgisayar\u0131n\u0131zda e\u011fitebilirsiniz. Ard\u0131ndan, Rockchip'in RKNN Toolkit gibi ara\u00e7lar\u0131 kullanarak e\u011fitilmi\u015f modelleri NPU \u00fczerinde \u00e7al\u0131\u015ft\u0131r\u0131labilir RKNN format\u0131na d\u00f6n\u00fc\u015ft\u00fcrmeniz gerekir.<\/dd>\n<dt>G\u00f6m\u00fcl\u00fc sistemler i\u00e7in hangi programlama dilleri \u00f6nerilir?<\/dt>\n<dd>Python, basitlik ve zengin k\u00fct\u00fcphane ekosistemi (OpenCV, NumPy, Flask) nedeniyle ba\u015flang\u0131\u00e7 ve prototipleme i\u00e7in \u00e7ok pop\u00fclerdir. Performans\u0131n kritik oldu\u011fu yerlerde C\/C++ dilleri tercih edilir ve \u00e7apraz derleme ile Luckfox Lyra Zero W \u00fczerinde \u00e7al\u0131\u015ft\u0131r\u0131labilir.<\/dd>\n<dt>Luckfox Lyra Zero W'nin Raspberry Pi Zero W'den fark\u0131 nedir?<\/dt>\n<dd>En temel fark, Luckfox Lyra Zero W'nin entegre bir NPU'ya (Yapay Zeka H\u0131zland\u0131r\u0131c\u0131s\u0131) sahip olmas\u0131d\u0131r. Bu, AI \u00e7\u0131kar\u0131m g\u00f6revlerinde (nesne alg\u0131lama gibi) Raspberry Pi Zero W'ye g\u00f6re \u00e7ok daha y\u00fcksek performans ve enerji verimlili\u011fi sunar. Raspberry Pi Zero W daha genel ama\u00e7l\u0131 bir mikro bilgisayarken, Luckfox Lyra Zero W AIoT ve g\u00f6m\u00fcl\u00fc vizyon uygulamalar\u0131 i\u00e7in daha optimize edilmi\u015ftir.<\/dd>\n<\/dl>\n<p><\/body><\/p>\n","protected":false},"excerpt":{"rendered":"IoT ve g\u00f6m\u00fcl\u00fc yapay zeka projeleriniz i\u00e7in uygun maliyetli, g\u00fc\u00e7l\u00fc ve k\u00fc\u00e7\u00fck bir \u00e7\u00f6z\u00fcm m\u00fc ar\u0131yorsunuz? Luckfox Lyra&hellip;","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"csco_page_header_type":"","csco_page_load_nextpost":"","csco_page_subscribe_form":"","csco_page_contact_form":"","footnotes":""},"categories":[1],"tags":[],"class_list":{"0":"post-31037","1":"post","2":"type-post","3":"status-publish","4":"format-standard","6":"category-genel","7":"cs-entry","8":"cs-video-wrap"},"yoast_head":"<!-- This site is optimized with the Yoast SEO Premium plugin v20.5 (Yoast SEO v25.3.1) - https:\/\/yoast.com\/wordpress\/plugins\/seo\/ -->\n<title>Luckfox Lyra Zero W ile Geli\u015fime Ba\u015flang\u0131\u00e7 Rehberi<\/title>\n<meta name=\"description\" content=\"IoT ve g\u00f6m\u00fcl\u00fc yapay zeka projeleriniz i\u00e7in uygun maliyetli, g\u00fc\u00e7l\u00fc ve k\u00fc\u00e7\u00fck bir \u00e7\u00f6z\u00fcm m\u00fc ar\u0131yorsunuz? Luckfox Lyra Zero W, inovatif fikirlerinizi ger\u00e7e\u011fe d\u00f6n\u00fc\u015ft\u00fcrmek i\u00e7in size harika bir ba\u015flang\u0131\u00e7 noktas\u0131 sunuyor. Bu rehberle, s\u0131f\u0131rdan ba\u015flayarak Luckfox Lyra Zero W d\u00fcnyas\u0131na ad\u0131m atacak ve kendi ak\u0131ll\u0131 cihazlar\u0131n\u0131z\u0131 in\u015fa etmenin keyfini ya\u015fayacaks\u0131n\u0131z.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/fatihsoysal.com\/blog\/luckfox-lyra-zero-w-ile-gelisime-baslangic-rehberi\/\" \/>\n<meta property=\"og:locale\" content=\"tr_TR\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Luckfox Lyra Zero W ile Geli\u015fime Ba\u015flang\u0131\u00e7 Rehberi\" \/>\n<meta property=\"og:description\" content=\"IoT ve g\u00f6m\u00fcl\u00fc yapay zeka projeleriniz i\u00e7in uygun maliyetli, g\u00fc\u00e7l\u00fc ve k\u00fc\u00e7\u00fck bir \u00e7\u00f6z\u00fcm m\u00fc ar\u0131yorsunuz? Luckfox Lyra Zero W, inovatif fikirlerinizi ger\u00e7e\u011fe d\u00f6n\u00fc\u015ft\u00fcrmek i\u00e7in size harika bir ba\u015flang\u0131\u00e7 noktas\u0131 sunuyor. Bu rehberle, s\u0131f\u0131rdan ba\u015flayarak Luckfox Lyra Zero W d\u00fcnyas\u0131na ad\u0131m atacak ve kendi ak\u0131ll\u0131 cihazlar\u0131n\u0131z\u0131 in\u015fa etmenin keyfini ya\u015fayacaks\u0131n\u0131z.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/fatihsoysal.com\/blog\/luckfox-lyra-zero-w-ile-gelisime-baslangic-rehberi\/\" \/>\n<meta property=\"og:site_name\" content=\"Kodlar\u0131n Gizemli D\u00fcnyas\u0131\" \/>\n<meta property=\"article:published_time\" content=\"2025-10-04T22:02:00+00:00\" \/>\n<meta name=\"author\" content=\"Fatih Soysal\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:label1\" content=\"Yazan:\" \/>\n\t<meta name=\"twitter:data1\" content=\"Fatih Soysal\" \/>\n\t<meta name=\"twitter:label2\" content=\"Tahmini okuma s\u00fcresi\" \/>\n\t<meta name=\"twitter:data2\" content=\"33 dakika\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\/\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\/\/fatihsoysal.com\/blog\/luckfox-lyra-zero-w-ile-gelisime-baslangic-rehberi\/#article\",\"isPartOf\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/luckfox-lyra-zero-w-ile-gelisime-baslangic-rehberi\/\"},\"author\":{\"name\":\"Fatih Soysal\",\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1\"},\"headline\":\"Luckfox Lyra Zero W ile Geli\u015fime Ba\u015flang\u0131\u00e7 Rehberi\",\"datePublished\":\"2025-10-04T22:02:00+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/luckfox-lyra-zero-w-ile-gelisime-baslangic-rehberi\/\"},\"wordCount\":5033,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1\"},\"inLanguage\":\"tr\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\/\/fatihsoysal.com\/blog\/luckfox-lyra-zero-w-ile-gelisime-baslangic-rehberi\/#respond\"]}],\"copyrightYear\":\"2025\",\"copyrightHolder\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#organization\"}},{\"@type\":\"WebPage\",\"@id\":\"https:\/\/fatihsoysal.com\/blog\/luckfox-lyra-zero-w-ile-gelisime-baslangic-rehberi\/\",\"url\":\"https:\/\/fatihsoysal.com\/blog\/luckfox-lyra-zero-w-ile-gelisime-baslangic-rehberi\/\",\"name\":\"Luckfox Lyra Zero W ile Geli\u015fime Ba\u015flang\u0131\u00e7 Rehberi\",\"isPartOf\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#website\"},\"datePublished\":\"2025-10-04T22:02:00+00:00\",\"description\":\"IoT ve g\u00f6m\u00fcl\u00fc yapay zeka projeleriniz i\u00e7in uygun maliyetli, g\u00fc\u00e7l\u00fc ve k\u00fc\u00e7\u00fck bir \u00e7\u00f6z\u00fcm m\u00fc ar\u0131yorsunuz? Luckfox Lyra Zero W, inovatif fikirlerinizi ger\u00e7e\u011fe d\u00f6n\u00fc\u015ft\u00fcrmek i\u00e7in size harika bir ba\u015flang\u0131\u00e7 noktas\u0131 sunuyor. Bu rehberle, s\u0131f\u0131rdan ba\u015flayarak Luckfox Lyra Zero W d\u00fcnyas\u0131na ad\u0131m atacak ve kendi ak\u0131ll\u0131 cihazlar\u0131n\u0131z\u0131 in\u015fa etmenin keyfini ya\u015fayacaks\u0131n\u0131z.\",\"breadcrumb\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/luckfox-lyra-zero-w-ile-gelisime-baslangic-rehberi\/#breadcrumb\"},\"inLanguage\":\"tr\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\/\/fatihsoysal.com\/blog\/luckfox-lyra-zero-w-ile-gelisime-baslangic-rehberi\/\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\/\/fatihsoysal.com\/blog\/luckfox-lyra-zero-w-ile-gelisime-baslangic-rehberi\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Anasayfa\",\"item\":\"https:\/\/fatihsoysal.com\/blog\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Luckfox Lyra Zero W ile Geli\u015fime Ba\u015flang\u0131\u00e7 Rehberi\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#website\",\"url\":\"https:\/\/fatihsoysal.com\/blog\/\",\"name\":\"Fatihsoysal.com\",\"description\":\"Blog - Yaz\u0131l\u0131m D\u00fcnyas\u0131 Tecr\u00fcbelerim\",\"publisher\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\/\/fatihsoysal.com\/blog\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"tr\"},{\"@type\":[\"Person\",\"Organization\"],\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1\",\"name\":\"Fatih Soysal\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"tr\",\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/image\/\",\"url\":\"https:\/\/fatihsoysal.com\/blog\/wp-content\/uploads\/2024\/04\/cropped-replicate-prediction-3kgg1hgjn5rgp0cf0p5tr0jw7w-1.png\",\"contentUrl\":\"https:\/\/fatihsoysal.com\/blog\/wp-content\/uploads\/2024\/04\/cropped-replicate-prediction-3kgg1hgjn5rgp0cf0p5tr0jw7w-1.png\",\"width\":512,\"height\":512,\"caption\":\"Fatih Soysal\"},\"logo\":{\"@id\":\"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/image\/\"},\"description\":\"Kullan\u0131m ve kodlama m\u00fckemmeliyetini odak alan uygulamalar olu\u015fturma deneyimine sahip, profesyonel olarak 15+ y\u0131l \u00fczeri deneyime sahip bir yaz\u0131l\u0131m m\u00fchendisi.\",\"url\":\"https:\/\/fatihsoysal.com\/blog\/author\/fatihsoysal\/\"}]}<\/script>\n<!-- \/ Yoast SEO Premium plugin. -->","yoast_head_json":{"title":"Luckfox Lyra Zero W ile Geli\u015fime Ba\u015flang\u0131\u00e7 Rehberi","description":"IoT ve g\u00f6m\u00fcl\u00fc yapay zeka projeleriniz i\u00e7in uygun maliyetli, g\u00fc\u00e7l\u00fc ve k\u00fc\u00e7\u00fck bir \u00e7\u00f6z\u00fcm m\u00fc ar\u0131yorsunuz? Luckfox Lyra Zero W, inovatif fikirlerinizi ger\u00e7e\u011fe d\u00f6n\u00fc\u015ft\u00fcrmek i\u00e7in size harika bir ba\u015flang\u0131\u00e7 noktas\u0131 sunuyor. Bu rehberle, s\u0131f\u0131rdan ba\u015flayarak Luckfox Lyra Zero W d\u00fcnyas\u0131na ad\u0131m atacak ve kendi ak\u0131ll\u0131 cihazlar\u0131n\u0131z\u0131 in\u015fa etmenin keyfini ya\u015fayacaks\u0131n\u0131z.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/fatihsoysal.com\/blog\/luckfox-lyra-zero-w-ile-gelisime-baslangic-rehberi\/","og_locale":"tr_TR","og_type":"article","og_title":"Luckfox Lyra Zero W ile Geli\u015fime Ba\u015flang\u0131\u00e7 Rehberi","og_description":"IoT ve g\u00f6m\u00fcl\u00fc yapay zeka projeleriniz i\u00e7in uygun maliyetli, g\u00fc\u00e7l\u00fc ve k\u00fc\u00e7\u00fck bir \u00e7\u00f6z\u00fcm m\u00fc ar\u0131yorsunuz? Luckfox Lyra Zero W, inovatif fikirlerinizi ger\u00e7e\u011fe d\u00f6n\u00fc\u015ft\u00fcrmek i\u00e7in size harika bir ba\u015flang\u0131\u00e7 noktas\u0131 sunuyor. Bu rehberle, s\u0131f\u0131rdan ba\u015flayarak Luckfox Lyra Zero W d\u00fcnyas\u0131na ad\u0131m atacak ve kendi ak\u0131ll\u0131 cihazlar\u0131n\u0131z\u0131 in\u015fa etmenin keyfini ya\u015fayacaks\u0131n\u0131z.","og_url":"https:\/\/fatihsoysal.com\/blog\/luckfox-lyra-zero-w-ile-gelisime-baslangic-rehberi\/","og_site_name":"Kodlar\u0131n Gizemli D\u00fcnyas\u0131","article_published_time":"2025-10-04T22:02:00+00:00","author":"Fatih Soysal","twitter_card":"summary_large_image","twitter_misc":{"Yazan:":"Fatih Soysal","Tahmini okuma s\u00fcresi":"33 dakika"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/fatihsoysal.com\/blog\/luckfox-lyra-zero-w-ile-gelisime-baslangic-rehberi\/#article","isPartOf":{"@id":"https:\/\/fatihsoysal.com\/blog\/luckfox-lyra-zero-w-ile-gelisime-baslangic-rehberi\/"},"author":{"name":"Fatih Soysal","@id":"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1"},"headline":"Luckfox Lyra Zero W ile Geli\u015fime Ba\u015flang\u0131\u00e7 Rehberi","datePublished":"2025-10-04T22:02:00+00:00","mainEntityOfPage":{"@id":"https:\/\/fatihsoysal.com\/blog\/luckfox-lyra-zero-w-ile-gelisime-baslangic-rehberi\/"},"wordCount":5033,"commentCount":0,"publisher":{"@id":"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1"},"inLanguage":"tr","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/fatihsoysal.com\/blog\/luckfox-lyra-zero-w-ile-gelisime-baslangic-rehberi\/#respond"]}],"copyrightYear":"2025","copyrightHolder":{"@id":"https:\/\/fatihsoysal.com\/blog\/#organization"}},{"@type":"WebPage","@id":"https:\/\/fatihsoysal.com\/blog\/luckfox-lyra-zero-w-ile-gelisime-baslangic-rehberi\/","url":"https:\/\/fatihsoysal.com\/blog\/luckfox-lyra-zero-w-ile-gelisime-baslangic-rehberi\/","name":"Luckfox Lyra Zero W ile Geli\u015fime Ba\u015flang\u0131\u00e7 Rehberi","isPartOf":{"@id":"https:\/\/fatihsoysal.com\/blog\/#website"},"datePublished":"2025-10-04T22:02:00+00:00","description":"IoT ve g\u00f6m\u00fcl\u00fc yapay zeka projeleriniz i\u00e7in uygun maliyetli, g\u00fc\u00e7l\u00fc ve k\u00fc\u00e7\u00fck bir \u00e7\u00f6z\u00fcm m\u00fc ar\u0131yorsunuz? Luckfox Lyra Zero W, inovatif fikirlerinizi ger\u00e7e\u011fe d\u00f6n\u00fc\u015ft\u00fcrmek i\u00e7in size harika bir ba\u015flang\u0131\u00e7 noktas\u0131 sunuyor. Bu rehberle, s\u0131f\u0131rdan ba\u015flayarak Luckfox Lyra Zero W d\u00fcnyas\u0131na ad\u0131m atacak ve kendi ak\u0131ll\u0131 cihazlar\u0131n\u0131z\u0131 in\u015fa etmenin keyfini ya\u015fayacaks\u0131n\u0131z.","breadcrumb":{"@id":"https:\/\/fatihsoysal.com\/blog\/luckfox-lyra-zero-w-ile-gelisime-baslangic-rehberi\/#breadcrumb"},"inLanguage":"tr","potentialAction":[{"@type":"ReadAction","target":["https:\/\/fatihsoysal.com\/blog\/luckfox-lyra-zero-w-ile-gelisime-baslangic-rehberi\/"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/fatihsoysal.com\/blog\/luckfox-lyra-zero-w-ile-gelisime-baslangic-rehberi\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Anasayfa","item":"https:\/\/fatihsoysal.com\/blog\/"},{"@type":"ListItem","position":2,"name":"Luckfox Lyra Zero W ile Geli\u015fime Ba\u015flang\u0131\u00e7 Rehberi"}]},{"@type":"WebSite","@id":"https:\/\/fatihsoysal.com\/blog\/#website","url":"https:\/\/fatihsoysal.com\/blog\/","name":"Fatihsoysal.com","description":"Blog - Yaz\u0131l\u0131m D\u00fcnyas\u0131 Tecr\u00fcbelerim","publisher":{"@id":"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/fatihsoysal.com\/blog\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"tr"},{"@type":["Person","Organization"],"@id":"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/002a254750921dcfd568a99e48240dd1","name":"Fatih Soysal","image":{"@type":"ImageObject","inLanguage":"tr","@id":"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/image\/","url":"https:\/\/fatihsoysal.com\/blog\/wp-content\/uploads\/2024\/04\/cropped-replicate-prediction-3kgg1hgjn5rgp0cf0p5tr0jw7w-1.png","contentUrl":"https:\/\/fatihsoysal.com\/blog\/wp-content\/uploads\/2024\/04\/cropped-replicate-prediction-3kgg1hgjn5rgp0cf0p5tr0jw7w-1.png","width":512,"height":512,"caption":"Fatih Soysal"},"logo":{"@id":"https:\/\/fatihsoysal.com\/blog\/#\/schema\/person\/image\/"},"description":"Kullan\u0131m ve kodlama m\u00fckemmeliyetini odak alan uygulamalar olu\u015fturma deneyimine sahip, profesyonel olarak 15+ y\u0131l \u00fczeri deneyime sahip bir yaz\u0131l\u0131m m\u00fchendisi.","url":"https:\/\/fatihsoysal.com\/blog\/author\/fatihsoysal\/"}]}},"yoast_meta":{"yoast_wpseo_title":"","yoast_wpseo_metadesc":"","yoast_wpseo_canonical":""},"amp_enabled":true,"_links":{"self":[{"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/posts\/31037","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/comments?post=31037"}],"version-history":[{"count":0,"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/posts\/31037\/revisions"}],"wp:attachment":[{"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/media?parent=31037"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/categories?post=31037"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/fatihsoysal.com\/blog\/wp-json\/wp\/v2\/tags?post=31037"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}