Shape recognition with added noise
Afficher commentaires plus anciens
I am trying to write a program that will recognize shapes. The program works for the png "Stickers" that has no noise but I am having trouble removing the added "noise" in the png "StickersWithNoise." I am having trouble knowing where to begin.
My Shape_recognition.m does the following (you can also see my comments in MATLAB)
- Reads picture
- Converts the white background to black
- Converts the picture from rgb to grayscale
- Converts picture from grayscale to binary
- Find connected components
- Use regionprops to find area and perimeter
- Calculate ratio of perimeter/sqrt(area) as suggested by the pdf "The Perimeter-Area Relation"
- Determine shape based upon these values
Réponses (1)
Image Analyst
le 19 Fév 2022
Try getting rid of the purple and brown line like this
[labeledImage, numBlobs] = bwlabel(BW_pic);
% Get rid of the purple line
purpleLine = ismember(labeledImage, 1); % First find purple line. It will be blob #1.
BW_pic(purpleLine) = false; % Then erase it from the binary image.
Catégories
En savoir plus sur Semantic Segmentation dans Centre d'aide et File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!