April 2, 2020 - Continuing New Color Segmentation Method

Lab Work

Today, I continued to try and find a new method in color segmentation to increase accuracy in individual egg measurements. Last time, I was able to combine multiple images together to form one larger image, which helped me to plot a color space graph that would help in the color segmentation, but was not able to combine all of the images that I wanted to. So, in order to find a way to combine all of the images, I used these sources:

  1. Python: Concatenate 2 irregular sized images by automatically detecting the sizes
  2. Python OpenCV: Concatenating images
  3. In python, how can I use regex to replace square bracket with parentheses

From these sources, I was able to produce this image (which consists of all of the egg images that I screenshotted):

final_image.jpg

and subsequently this graph:

final_image_color_plot.png


Next Steps

Analyze the graphs to find a better color threshold for segmentation

Written on April 2, 2020