CSC3014 Computer Vision assignment
The purpose of this script is to process a birds-eye view image of a tray of eggs. The script produces a new image with red rectangles indicating high certainty of cracks and yellow rectangles indicating moderate certainty of cracks.
I have added a notebook (cracked-egg-detection.ipynb) for some visualization.
No modifications were made to the script — it is still the same code that was submitted for my assignment. I just added this for fun.
To test the system, create a folder named images in the cloned repo folder. Add these 3 images into the images folder. Ensure that the files are properly named test_image1, test_image2 and test_image3.
There are now 3 test images located in the images folder.
Change the cv2.imread() function parameter to the image you wish to use in the system. By default, test_image1 is read.
Simply run the code to use the system.
Comments marked with a # PLOT indicates that the code snippet can be commented out to plot a figure of some code done in that section.
To use the adaptive threshold method for abnormality detection, change the second parameter of the abnormal function call in Section 4 to 2 instead of the current 1.
Also change the tuple unpacking in Section 5 to the adaptive_thres variable instead from the current canny_thres.