Deblurred-improved PCNN Model Image Fusi On Algorithm Based On Compressed Sensing Theory

Compressed sensing as the emerging data sampling theory, on the condition that a signal is compressible or sparse in a transform domain, it can reconstruct the high-dimensional signal from a small amount of low-dimensional measurements with high robability. The sparsity has a great effect in the reconstructed image quality. Apply the compressed sensing in image reconstruction. Therefore, the idea of image fusion based on compressed sensing is worth studying. This paper image put forward a new image fusion nethod with PCNN model in compressed sensing from signal structure and information ntegrity, which is given full consideration to system physical characteristics restrictions ind measurement environmental impact.a new method of image fusion based on improved PCNN is proposed in compressive ensing domain. Original images are compressed by compressed sensing to obtain neasurements. Improved PCNN model is built based on physical significance of neasurements. Namely link coefficient, weighting matrix and threshold amplification coefficient are set adaptively. Measurements are selected as input neuron to obtain gnition-map as fusion operator. Fusion image is obtained according reconstructing ilgorithm. The algorithm overcomes traditional compressed sensing image fusion defects, he subjective and objective evaluation indexes show that this method is superior to compressed sensing image fusion methods in existence.Considering that there is a deviation between the measurement image and the real mage because of the measurement system physical characteristics restrictions and neasurement environmental impact. In order to achieve image transfer, repair, and super esolution imaging without increasing the system complexity, a improved image fusion )ased on PCNN model in compressed sensing domain is designed based on the theory of compressed sensing and image degenerate model. This system modifies the measurement natrix by using the down-sampling operator and blurring operator of the image degenerate model as well as employ the low redundant image multiscale transform based on a new anti-aliasing direction filtering, then combine with the iterative hard threshold algorithm to reconstruct images. The noise limit is regarded as the iteration stopping criteria, which can remove the measurement noise, and a Poisson-singular-integral operator is introduced for deblurring as well as suppressing the amplification of the noise, Experimental results demonstrate the effectiveness of our system in the quality of the reconstructed images and computing efficiency.