Research On Compressed Sensing Technology Based On Integral Imaging

Three-dimensional integral imaging technology has the advantages of viewing without visual fatigue and without the need for auxiliary equipment.It is an emerging naked-eye 3D display technology.In the research of stereo image,integral imaging technology is one of the research directions of scientific researchers.Through integral stereo imaging technology,people can watch scenes with a sense of real depth.But at the same time to strengthen this visual reality,the transmission and storage of integral imaging images has also become an urgent problem.The use of traditional image compression methods not only increase storage pressure,but also have high requirements on hardware devices,thus leading to unable to efficiently compress the integral imaging image;Compressed sensing technology is a new type of image compression technology,which has the advantages of convenient transmission and reduced waste of resources.Therefore,this paper applies compressed sensing technology into integral imaging image compression.Based on a careful study of traditional methods,several compressed sensing algorithms for integral imaging images have been proposed.In order to effectively compress and reconstruct the integral imaging image,a compressive sensing algorithm based on integral imaging image has been proposed.The block compression algorithm is used to compress the image.The integral imaging image has the characteristics of large amount of data and high redundancy;so the sample is sampled first,then the image is adaptively partitioned,and the image is subjected to discrete cosine transform.According to the discreteness between adjacent pixels in the image block.The cosine transform coefficient difference is used for block classification,and different types of image blocks are measured and sampled using different sampling rates.In the reconstruction stage,a full variational algorithm is used to reconstruct each image block,the image blocks are recombined together to obtain the entire image,and then the image is sampled and finally restored to obtain a complete reconstructed image.In order to further improve the accuracy of image reconstruction,a sparsity adaptive orthogonal matching pursuit algorithm has been proposed.This algorithm can adaptively estimate the sparsity K value of signal.The matching pursuit algorithm is combined to improve the orthogonal matching pursuit algorithm,so that the algorithm has better recovery quality and efficiency.Finally,the two proposed algorithms are combined to improve both the sparse step and the reconstruction step of compressed sensing,and to observe and reconstruct the integral imaging image.The experimental results show that the algorithm proposed in this paper can effectively compress and reconstruct the integral image,and the whole restored image and the details are very clear,which not only improves the reconstruction accuracy of the integral image,but also reduces the running time,with obvious advantages.