how to inpaint a region of interest recursively

Question

David Levi 2022-1-28

0
链接

此问题的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1638200-how-to-inpaint-a-region-of-interest-recursively

评论： Manolis about 6 hours 前

采纳的回答： Image Analyst

Hi everyone!

I have an image called 'lion.jpg' .

I did segmentation to the lions in the center using the segmentImage.m function , and got this image:

I need to inpaint the black region that was created , and I need a solution to do that recursively from outside to inside:

I tried to use the inpaintCoherent function but the result was not acually what i needed , I got this:

You can see that the inpaintCoherent function filled the area of the mountains with colored lines , and this is not good for me. I want to fille the region smoothly.

I tried to blure the region but I really dont want to do that.

I shared my code : try1.m with you, and you can see it here too:

clc;

clear;

close all;

I = imread('lion.jpg');

figure; imshow(I,[]);

[BW,mask] = segmentImage(I);

figure; imshow(BW);

bg = I - mask;

figure; imshow(bg)

figure; montage({I,BW});

%% inpaint

title(['Image to Be Inpainted',' | ','Mask for Inpainting']);

J = inpaintCoherent(I,BW,'Radius',7);

figure; imshow(J);

figure;

montage({I,J});

title(['Image to Be Inpainted',' | ','Inpainted Image']);

%% try to blur the image:

redChannel = J(:, :, 1);

greenChannel = J(:, :, 2);

blueChannel = J(:, :, 3);

% H = fspecial('disk',20);

H = fspecial('average',33);

redChannel = roifilt2(H,redChannel,BW);

greenChannel = roifilt2(H,greenChannel,BW);

blueChannel = roifilt2(H,blueChannel,BW);

rgbImage = cat(3, redChannel,greenChannel,blueChannel);

figure;

subplot(1,2,1); imshow(J); title('Inpainted Image');

subplot(1,2,2); imshow(rgbImage); title('image after blure');

0 个评论
显示 -2更早的评论隐藏 -2更早的评论

请先登录，再进行评论。

请先登录，再回答此问题。

Answer 1

Image Analyst 2022-1-28

0
链接

此回答的直接链接

https://ww2.mathworks.cn/matlabcentral/answers/1638200-how-to-inpaint-a-region-of-interest-recursively#answer_883810

The function you want is called regionfill().

Description

example

J = regionfill(I,mask) fills the regions in image I specified by mask. Non-zero pixels in mask designate the pixels of image I to fill. You can use regionfill to remove objects in an image or to replace invalid pixel values using their neighbors.

example