Research On Block-based Adaptive Compressed Sensing Of Image

Wireless multimedia sensor networks is a new type of wireless sensor networks that can perceive and process audio, video, images and other multimedia information, it has broad application prospects in the monitoring field. However, computing power, storage space and battery life of wireless multimedia sensor networks are restricted, at the same time, as its main information carrier, the amount of data of the image signal is large. Therefore, before transmission, it is particularly important to reduce the amount of data transmission by image compression. However, the conventional image compression method based on Nyquist sampling theorem is not suitable for wireless multimedia sensor networks with limited resources.Based on the study and analysis of current image block compressed sensing algorithm, by using adaptive sampling rate allocation strategy, improved algorithms are proposed by improving BCS-SPL algorithm in this paper, the quality of the reconstructed image is improved and the progressive image transmission is achieved, to accommodate the wireless multimedia sensor networks with limited resources. The main research work is as follows:1、 An adaptive compressed sensing algorithm of image based on wavelet transform coefficients is proposed by improving BCS-SPL algorithm. Though adding adaptive sampling rate assignment policy based on wavelet transform coefficients, the quality of the overall reconstructed image is improved. Combined with adaptive sampling strategy, the fixed measurements and adaptive measurements are transmitted separately, achieve a progressive image transmission, and improve the transmission efficiency, to accommodate the low bandwidth and instability of wireless multimedia sensor networks. According to experimental results, the proposed algorithm can achieve a progressive image transmission, and at the same sampling rate, the quality of the reconstructed image is better than the original BCS-SPL algorithm.2、According to the framework of adaptive compressed sensing algorithm of image based on wavelet transform coefficients, standard deviation is adopted as adaptive sampling rate allocation criteria, an adaptive compressed sensing algorithm of image based on standard deviation. Experimental results show that, compared with the BCS-SPL algorithm is proposed, the proposed algorithm can improve the quality of image reconstruction. Compared with the adaptive compressed sensing algorithm of image based on wavelet transform coefficients, although the quality of the reconstructed image declines, but measurement time is less, the complexity of calculation is lower.In short, the two proposed adaptive compressed sensing algorithm of image can improve quality of the reconstructed image and achieve progressive image transmission, and two algorithms have their own advantages respectively in the quality of reconstructed images and computational complexity. The research of this paper can provide a reference for the study of wireless multimedia sensor network image compression algorithm.