[image detection] Research on cumulative weighted edge detection method based on gray image, with matlab code

1 Content introduction

The problem of edge detection based on gray image is studied , The traditional edge detection method is very sensitive to noise , An accumulation weighted edge detection method based on gray image is proposed . This method adopts the method of cumulative weighting , It not only effectively filters out random noise , And the image details are well preserved . Theoretical analysis and simulation results show that , This method has good noise resistance , Isotropy , Good real-time , Good edge retention . Compared with the traditional template method , The edge detection effect of this method is better .

In image segmentation , Edge detection method is the most studied method , It tries to solve the problem of image segmentation by detecting the edges of different regions . Most of the main information of the image exists in the edge of the image , It is mainly manifested in the discontinuity of local features of the image , It is the place where the gray level changes violently in the image , That is, the place where the signal changes strangely . The gray level of the singular signal along the edge changes violently , Generally, the edges are divided into two types: step shape and roof shape . The gray values on both sides of the edge in the step edge have obvious changes ; The middle edge of the roof edge is located at the junction of gray increase and decrease . Mathematically, the derivative of gray scale can be used to describe the change of edge points , To the step edge 、 Find the first order of the roof edge 、 Second derivative . so , For the step edge point, the first derivative of the gray change curve reaches the maximum at the edge point , The second derivative crosses zero at the edge . To the roof like edge point , The first derivative of its gray change curve crosses zero at the edge point , The second derivative reaches its extreme value at the edge . Classic edge detection methods , It is to construct an edge detection operator for a small neighborhood of pixels in the original image . First, the noise in the image is filtered by smoothing , Then perform first-order differential or second-order differential operation , Find the maximum value of the gradient or the zero crossing of the second derivative , Finally, select the appropriate threshold to detect the boundary . For various reasons , Images are often disturbed by random noise . Classical edge detection methods introduce various forms of differential operations , This will inevitably cause extreme sensitivity to noise , Plus sex [1]. Aiming at the problem of low noise resistance of traditional edge detection methods , In this paper, a cumulative weighted edge detection method is proposed . This method has good noise suppression ability , At the same time, it can effectively maintain the edge characteristics of the image . The classical edge detection method is introduced , The cumulative weighted edge detection method and its implementation steps are proposed , The cumulative weighted edge detection method is simulated , And with the traditional Ｌａｐｌａｃｅ and Ｓｏｂｅｌ Methods are compared , Finally, the advantages of cumulative weighted edge detection method are discussed . The result of performing edge detection is often to detect the noise as an edge , The real edge is not detected due to noise interference . So for noisy images , A good edge detection method should have good noise suppression ability , At the same time, it has complete edge preserving characteristics

​

2 Simulation code

function Final=img_fusion(LH_out,HL_out,lphaar,hphaar,lphaar2,hphaar2,it2)​  %LH_output hereit2=1;LL=[];LH=[];HL=[];HH=[];       app_img = LH_out;    % Initializing the Approximation Image.        for i = 1:it2        [LL{i}, LH{i} ,HL{i} , HH{i}] = img_decomp(app_img, lphaar, hphaar);        app_img = LL{i};    end            LH_final = LL{it2};    for i = it2:-1:1        LH_final = [LH_final,LH{i};HL{i},HH{i}];    end   LH_final=(LH_final-min(min(LH_final)))/(max(max(LH_final)) - min(min(LH_final)));        %HL_output here     app_img = HL_out;    % Initializing the Approximation Image.        for i = 1:it2        [LL{i}, LH{i} ,HL{i} , HH{i}] = img_decomp(app_img, lphaar, hphaar);        app_img = LL{i};    end        HL_final = LL{it2};    for i = it2:-1:1        HL_final = [HL_final,LH{i};HL{i},HH{i}];    end HL_final=(HL_final-min(min(HL_final)))/(max(max(HL_final)) - min(min(HL_final)));​    Final = img_fuse(LH_final, HL_final);                   Final = img_recomp(Final, it2, lphaar2, hphaar2);Final=(Final-min(min(Final)))/(max(max(Final)) - min(min(Final)));end

3 Running results

4 reference

[1] Jiao Jingfeng , Xiaohuaitie , Fu Qiang . Accumulation weighted edge detection method based on gray image [J]. Electronic Engineer , 2005, 31(10):4.

About bloggers ： Good at intelligent optimization algorithms 、 Neural networks predict 、 signal processing 、 Cellular automata 、 The image processing 、 Path planning 、 UAV and other fields Matlab Simulation , relevant matlab Code problems can be exchanged by private letter .

Some theories cite network literature , If there is infringement, contact the blogger to delete .