Research On Compressed Sensing Technology And The Application In Image Fusion

Efficient fusion algorithm can improve the accuracy of image recognition and the richness of information.Compressive sensing technology is applied to image fusion,which can reduce the amount of data collected and speed up the efficiency of the algorithm.Through making a concrete analysis of the original method,a new fusion method is proposed to improve the image fusion efficiency and fusion effect.The main work of this paper is as follows:(1)The basic principle of common image fusion algorithm is studied.This paper studies the traditional algorithms of image fusion.We analyzed the development of image fusion methods.Then a new method of shear wave transformation is adopted.The basic idea of all kinds of reconstruction algorithms is analyzed.(2)A fusion algorithm based on NSST is proposed.The algorithm is used to multi-focus image fusion.Firstly,the multi-focus images are separated by NSST.The high-frequency sub-bands are measured using a measurement matrix to obtain the measured value.The low-frequency sub-band is used the regional energy matching rule to fuse.Finally,the fusion image is reconstructed by NSST inverse transform.The method can reduce the quality of the fusion image without affecting the visual effect,but the efficiency of the fusion algorithm is obvious.(3)This paper proposes a fusion algorithm using FFST.The algorithm is applicable to the infrared image type and visible image type.First,the images are divided into a low frequency sub-band and some high frequency sub-bands by FFST.For low-frequency sub-band,this paper uses the SML algorithm for fusion.The Laplacian energy value is used as the basis for selecting the pixels.For high-frequency sub-band,We use the taking maximum variance method for fusion to measured value of each blocked image.The method is more suitable for infrared and visible image.