Research Of Spatial Domain Multi-focus Image Fusion Method Based On Image Block

Multifocus image fusion is an important research direction in the field of image fusion,which can effectively overcome the limitation of optical lens focusing range,and present all scenes in different distances in an image clearly.This can not only effectively improve the utilization of image information,enhance the reliability of system recognition and target detection,but also provide a good foundation for further image processing,such as image segmentation,target recognition,feature extraction and so on.At present,multifocus image fusion has been widely used in target recognition,machine vision and other fields.In the multifocus image fusion,the fusion method based on the spatial domain directly fuses on the pixel space of the image,which can retain more original pixel information in the image,and has the advantages of good pixel consistency.In recent years,it is the focus of domestic and foreign researchers in this field.Based on the existing research at home and abroad,this paper focuses on the spatial multifocus image fusion based on image block.The main research contents are as follows:(1)In order to improve the accuracy of the measurement of the focus degree of the source image in the multifocus image fusion based on the spatial domain,an image fusion method based on weighted multiscale morphological gradient and uniform segmentation is proposed.In the process of obtaining the measurement map of the focus degree of the source image,firstly,the multiscale morphological gradient is obtained by multiscale morphological operation,and then the multiscale morphological gradient is obtained by scale weighting and distance weighting respectively.In the process of image segmentation,in order to reduce the situation that the image block contains both focus area pixels and defocus area pixels,the image is evenly divided into 2?2 size image blocks,and the value of 6?6 size range composed of image block and domain pixels is used as the focus degree value of the image block.Experimental results show that this method can improve the accuracy of the measurement of the focus degree of the smooth region in the image.(2)In order to solve the problem that the size of image block is not easy to determine in image segmentation,an image fusion method based on dense SIFT descriptor and adaptive segmentation is proposed.In the process of obtaining the focus measurement map of the source image,the dense SIFT descriptor graph of the source image is obtained first,and then it is used as the focus measurement map.In the process of image segmentation,the quadtree image segmentation method is used to determine the segmentation of image blocks according to the image focus distribution and realize the adaptive selection of image block size.Experiments show that this method can reduce the "block effect" in the fusion process,and is comparable to the existing multifocus image fusion methods in terms of subjective vision and objective evaluation indexes.