Adaptive Block Compressed Sensing Algorithm based on Integral Imaging

In order to effectively compress and reconstruct the elemental image array in integral imaging, an improved block compressed sensing algorithm based on integral imaging is proposed. The amount of elemental image data is large and the redundancy is high, so the image is first sampled by interlaced rows and columns, and then the discrete cosine transform (DCT) is performed. The block classification is based on the discrete cosine transform coefficient difference between adjacent pixels in the image block, and is divided into four sub-blocks according to the characteristics of the image. Use different sampling rates to measure samples for different types of sub-blocks. In the reconstruction stage, the total variation algorithm is used to reconstruct each sub-block, the sub-blocks are recombined together to obtain the entire image, and then the image is restored by extracting samples, and finally a complete reconstructed image is obtained. Experimental results show that the use of this algorithm to compress and reconstruct integral imaging images has a good effect.