基于CMOS图像传感器的压缩感知成像算法

A Compressive Sensing Imaging Algorithm Based on CMOS Image Sensor

近年来提出的压缩感知理论将信号采样和压缩同时进行,突破了奈奎斯特采样定理的限制,为低采样高分辨率成像提供了可能.为此,提出了一种基于CMOS图像传感器的压缩感知成像算法,采用并行处理策略对CMOS图像传感器A/D转换前的模拟像素矩阵进行压缩采样,减轻了A/D转换模块的负担,大大降低了CMOS图像传感器的功耗,并且该算法实现电路简单.仿真结果表明,所提算法能快速有效地进行测量值的获取,利用TVAL3算法重构的图像主客观质量较好.

Compressive sensing theory proposed in recent years integrates signal acquisition and compression steps,which breaks through the limit of Nyquist sampling theorem and provides possibility for high resolution imaging from low sampling data.A compressive sensing imaging algorithm based on CMOS image sensor was proposed in this paper,which adopts parallel processing strategies to compress and sample the analog pixel matrix prior to A/D conversion of CMOS image sensor.The proposed algorithm can alleviate the burden of A/D conversion module and reduce the power of CMOS image sensor,and moreover,it has a simple implementation circuit.Simulation results show that measurement values can be obtained rapidly and effectively with the proposed algorithm,and the image reconstructed using TVAL3 arithmetic has better subjective and objective quality.