Research On Multi-focus Image Fusion Performance Improvement Of Differential Image Driving

Due to the limitations of the optical lens depth of field,most of the lens imaging devices can not make all objects clear when acquiring multi-focus images.However,in some cases,people often need to get a clear image of all objects in a scene.In the above context,multi-focus image fusion technology has been widely applied in many field.With the rapid development of science and technology,multi-focus image fusion has played an important role in a series of important tasks such as medical imaging,computer vision,pattern recognition technology,military image and so on.Multi-focus image fusion algorithm can be divided into two methods:fusion domain-based fusion method and spatial domain fusion method.The spatial domain based method is roughly divided into block-based,region-based and focus-based detection methods.The focus area detection based method is most concerned by the scholar,but because of the complexity of the image,there is a clear boundary between the focus area and the defocused area because many images have no explicit texture and clear boundary.In the transform domain-based approach,the most representative method is the multi-scale fusion method.It is found that the fusion rules of different subbands are a key factor influencing the image fusion effect in the process of multi-scale transformation to multi-focus image fusion.However,in the new fusion rules to integrate the image,the results of the fusion results did not significantly improve.This article solves the problem as follows(1)Two kinds of fusion rules based on multi-scale and multi-directional domain distance(MMND)are proposed,which improve the performance of multi-focus image fusion.The theoretical basis of this method is that the difference between low quality fusion image and source image is more obvious than that of high quality fusion image and source image.(2)Based on the above theoretical basis,in the process of optimizing the fusion image,the pixels on the source image are divided into three kinds,the pixel of the focus area,the pixel of the focus area,and the transition area pixel.Based on the three pixel points of the source image,the results of the fusion image can be optimized by the decision graph constructed by the difference graph based on the MMND transform domain fusion rule and the MMND space domain fusion rule.Based on a large number of experimental results,this paper proves that the two fusion rules proposed in this paper are more excellent than the advanced algorithms proposed in recent years.