Hey Kyuhee,
It is my understanding that you want to identify yellow region in your image and fit a line that best approximates it's location.
Identification of the yellow region can be done using the following steps:
- Extracting the RGB channels in the image (Assuming RGB file format):
channel_Red = image(:, :, 1);
channel_Green = image(:, :, 2);
channel_Blue = image(:, :, 3);
2. Thresholding the required channels. Herein we need to identify yellow, thus you could try thresholding the Red and Green channels. A simple 'and' statement between the 2 binarized outputs can be taken to identify regions of yellowish tones.
BW_Red = im2binarize(channel_Red, threshold_Red);
BW_Green = im2binarize(channel_Green, threshold_Green);
3. Identifying the ROI (Region of Interest) in the binarized ouput. This will generate a rectangular bounding box around the ROI.
regionprops(BW, 'BoundingBox');
Co-ordinates and dimensions of the bounding box can then be extracted and used to generate the corner points of the required line segment (These will be the 2 opposite corner points of the rectangle). To insert the line in the image, you can use the 'InsertShape' function.
Hope this helps!
