Image Reconstruction System On Compressed Sensing

Image reconstruction has gradually become an essential research branch of image processing field.And more importantly,image reconstruction is regarded as one of the core content of innovative compressed sensing technology.It can be reconstructed precisely by using efficient algorithm when image signals meet the needs of sparsity.The compressed sensing is applied into image reconstruction system to achieve compression,transmission and reconstruction of information data on transmitting terminal and receiving terminal only by sampling a small amount of data.As a good result,the bandwidth resource and hardware equipment cost can be reduced apparently in the process of transmission and storage.Therefore,the research of image compressed sensing plays an important role in image communication,which has aroused scholars’ widely concern.In order to tackle unknown sparsity of current greedy algorithms in image compression sampling,a Sparsity Adaptive Compressive Sampling Matching Pursuit algorithm is proposed on basis of Compressive Sampling Matching Pursuit algorithm.The optimized algorithm implements adaptive estimation of sparsity by utilizing new estimated method and iteration.And then,signal reconstruction can be finished by Compressive Sampling Matching Pursuit algorithm.The simulation results indicate that the proposed algorithm not only overcome the influence of unknown sparsity,but also has the advantages of strong anti-noise ability,strong adaptive ability,high accuracy and good image reconstruction effect.Image reconstruction system on compressed sensing is designed,which includes image capture,sparse representation,measurement matrix construction,image data transmission,image reconstruction.And meanwhile,the improved algorithm is applied into image reconstruction system.Based on this,the test platform can be established,which means that the transmission and reconstruction between the transmitter and receiver can be achieved by camera capturing real-time image and compressed sensing processing.The test results show that the designed system can greatly satisfy performance requirements.And meanwhile,the reconstructed images of the system have greatly improved in terms of compression ratio,peak signal to noise ratio and reconstruction error.