Sistem kamera-berbilang telah digunakan secara meluas dalam banyak aplikasi seperti kenderaan kamera-berbilang, telefon pintar dengan dwi kamera dan sistem Penjejakan Sasaran-Berbilang (MTMCT). Dalam projek ini, tujuan utama melaksanakan sistem kamera-berbilang adalah untuk melaksanakan kebezajelasan super. Kebezajelasan super terdiri daripada memproses imej atau satu set imej untuk meningkatkan resolusi urutan video atau satu bingkai. Antara kaedah-kaedah kebezajelasan super, teknik kebezajelasan super gabungan dianggap sebagai kaedah yang paling sesuai untuk pelaksanaan langsung. Dengan menggunakan teknik ini, algoritma kebezajelasan super dibina dengan menggabungkan beberapa imej yang mempunyai kebezajelasan rendah Dalam projek ini, algoritma kebezajelasan super dicadangkan berdasarkan model pengimejan yang merangkumi fungsi tindak balas kamera, masa pendedahan dan ralat pengkuantuman sebagai tambahan kepada kekaburan ruang dan persampelan.
_______________________________________________________________________________________________________
Multi-camera system has been widely used in many applications such as multi-camera vehicle, smart phone with dual camera and Multi-Target Multi-Camera Tracking (MTMCT) system. In this project, the main purpose of developing multi-camera system is to perform super-resolution. Super-resolution consists of processing an image or a set of images in order to enhance the resolution of a video sequence or a single frame. There are several methods to apply super-resolution, from which fusion super-resolution techniques are considered to be the most adequate for real-time implementation. In fusion, super-resolution and high-resolution images are constructed from several observed low-resolution images. In this project, a super-resolution algorithm is proposed based on an imaging model that includes camera response function and distortion error in addition to spatial blurring and sampling.
Multi-camera system on a single raspberry pi / Tay Ying Sheng