How to detect AAL

Question

Brian Pravin 2021-1-6

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/710128-how-to-detect-aal

评论： Image Analyst 2021-1-14

How do I detect

Ive tried the coding below, but the segmentation is awful.

I=imread('black.jpg');
I2=rgb2gray(I);
J = imnoise(I2,'salt & pepper',0.02);
figure
imshow(J)
Kaverage = filter2(fspecial('average',3),J)/255;
figure
imshow(Kaverage)
Kmedian = medfilt2(J);
imshowpair(Kaverage,Kmedian,'montage')
BW1 = edge(Kmedian,'Canny');
BW2 = edge(Kmedian,'Prewitt');

3 个评论
显示 1更早的评论隐藏 1更早的评论

Brian Pravin 2021-1-7

Ive updated the question and coding.

Image Analyst 2021-1-14

Yeah, and removed the image and some words, and made the explanation incomprehensible, which essentially makes this question worthless.

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Image Analyst 2021-1-7

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/710128-how-to-detect-aal#answer_593163

编辑：Image Analyst 2021-1-7

There is no justification for doing edge detection on that image. Just because you can see edges does not mean you should do edge detection. If another novice told you to do that, then they are wrong.

What you should do it to threshold for non-black. That will find the foil. Then, if you don't have aligned foils, you should align the image. You can do that by scanning the middle half of the image to find the left edge and then getting the slope and calling imrotate. Then threshold again and call crop. Then you should have a mask prepared with locations of the circles. Pass that mask and the rotated, cropped image into regionprops to get the MeanIntensity and PixelValues of each circle. compare those to the values you'd get from a perfect/filled blister pack to determine which circles are filled and which are empty/broken.

So those are the steps. Are you willing to try them on your own first?