Image classification in video stream with contours

asked 2019-06-12 04:02:48 -0500

heflie gravatar image

Please I need your help with this problem, I want to create a program to differentiate between the two forms(2 images), with a camera in real time, here are the methods. I found but I’m not sure they’re going to work because I want the detection to be feasible if the object is inclined by 90 degrees or 180 degrees by example, I have to use machine learning in this problem but I am open to any proposition, also I do not have many images in the database.

Here are the methods I found but I'm not sure they will work;

1 - Apply Canny filter to extract contours.

2 - Use a features extractors such SIFT, Fourier Descriptors, Haralick's Features, Hough Transform to extract more details which could be summarised in a short vector.

3-Then train SVM or ANN with this vector.

image description

image description

Also the backgroud of the image can change.

And thank you.

edit retag flag offensive close merge delete


I want to create a program to differentiate between the two forms

can you explain it a bit more ? what exactly is the goal here ?

berak gravatar imageberak ( 2019-06-12 04:15:36 -0500 )edit

The goal is to detect two cases : Open or Close

heflie gravatar imageheflie ( 2019-06-12 04:18:10 -0500 )edit

Maybe contours are not the best way to solve this problem because the background changes a lot, I also tried Hough Transform to detect lines, but it's not working very well, I don't know what kind of methods I gonna use

heflie gravatar imageheflie ( 2019-06-12 04:57:50 -0500 )edit

You may want to look at template matching

It is sensitive to rotation and scale but if the camera is consistently positioned and you only care about 2 orientations, template matching may work

Chris gravatar imageChris ( 2019-06-12 10:43:25 -0500 )edit

Thank you Chris for your answer but I already tested it and it didn’t give me good results.

heflie gravatar imageheflie ( 2019-06-12 10:47:06 -0500 )edit