Smoothing edges change black color - OpenCV Q&A Forum

I am getting foreground image from grabCut algorithm and then i apply below algorithm to smooth its edges. It is working fine and smoothing edges but it is also changing black color in the image. You can see for this image it is working fine for this image but for this image it is changing black color please tell me how to fix this. This is my code for edge smoothing and i got it from Image edge smoothing with opencv #include <iostream> #include <vector> #include <string> #include <fstream> #include <opencv2/opencv.hpp> using namespace cv; using namespace std; int main(int argc, char **argv) { namedWindow("originalImage"); namedWindow("grabCutImage"); namedWindow("smoothImage"); Mat originalImage = imread("../data/original_image_1.jpg"); Mat grabCutImage = imread("../data/grab_cut_image_1.png"); // got from grabcut algorithm originalImage.convertTo(originalImage, CV_32FC3, 1.0 / 255.0); resize(originalImage, originalImage, grabCutImage.size()); grabCutImage.convertTo(grabCutImage, CV_32FC3, 1.0 / 255.0); Mat bg = Mat(grabCutImage.size(), CV_32FC3); bg = Scalar(1.0, 1.0, 1.0); Mat mask; Mat img_gray; cvtColor(grabCutImage, img_gray, COLOR_BGR2GRAY); img_gray.convertTo(mask, CV_32FC1); threshold(1.0 - mask, mask, 0.9, 1.0, THRESH_BINARY_INV); GaussianBlur(mask, mask, Size(21, 21), 11.0); Mat smooth; vector<Mat> ch_img(3); vector<Mat> ch_bg(3); split(originalImage, ch_img); split(bg, ch_bg); ch_img[0] = ch_img[0].mul(mask) + ch_bg[0].mul(1.0 - mask); ch_img[1] = ch_img[1].mul(mask) + ch_bg[1].mul(1.0 - mask); ch_img[2] = ch_img[2].mul(mask) + ch_bg[2].mul(1.0 - mask); merge(ch_img, smooth); merge(ch_bg, bg); imshow("originalImage", originalImage); imshow("grabCutImage", grabCutImage); imshow("smoothImage", smooth); waitKey(0); destroyAllWindows(); } here are images i am using

What is OpenCV? The Complete Guide (2024)

Opencv Tutorials

AMT - A comparative evaluation of snowflake particle shape estimation techniques used by the Precipitation Imaging Package (PIP), Multi-Angle Snowflake Camera (MASC), and Two-Dimensional Video Disdrometer (2DVD)

How to make a one object FOMO classification, have fewer false positives? - Help - Edge Impulse

Guide To OpenCV Functions For Image Processing

Advance your knowledge in tech - Packt+

Remote Sensing, Free Full-Text

What are blobs in image processing? - Quora

Good foreground segmentation methods apart from Canny - OpenCV Q&A Forum

Edge-preserving smoothing filter using fast M-estimation method with an automatic determination algorithm for basic width

12 Best Python Image Processing Libraries for Data Scientists

Automation of measurements inside an specific region using color threshold - Image Analysis - Image.sc Forum

Quantified emotion analysis based on design principles of color feature recognition in pictures

2011 · opencv/opencv Wiki · GitHub