Teaching: 30 min
Exercises: 0 min
  • What are the questions?

  • What are the objectives?

In this episode, we will provide two different challenges for you to attempt, based on the skills you have acquired so far. One of the challenges will be related to the shape of objects in images (morphometrics), while the other will be related to colors of objects in images (colorimetrics). We will not provide solution code for either of the challenges, but your instructors should be able to give you some gentle hints if you need them.

Morphometrics: Bacteria Colony Counting

As mentioned in the workshop introduction, your morphometric challenge is to determine how many bacteria colonies are in each of these images. These images can be found in the Desktop/workshops/image-processing/10-challenges/morphometric directory.

Colony image 1

Colony image 2

Colony image 3

Write a Python program that uses OpenCV to count the number of bacteria colonies in each image, and for each, produce a new image that outlines the colonies, and displays the number of coloines. Your output should be similar to this image:

Sample morphometric output

Colorimetrics: titration color analysis

The video showing the titration process first mentioned in the workshop introduction episode can be found in the Desktop/workshops/image-processing/10-challenges/colorimetric directory. Write a Python program that uses OpenCV to analyze the video on a frame-by-frame basis. Your program should do the following:

  1. Sample a kernel from the same location on each frame, and determine the average red, green, and blue channel value.

  2. Display a graph plotting the average color channel values as a function of the frame number, similar to this image:

    Titration colors

  3. Save the graph as an image named titration.png.

  4. Output a CSV file named titration.csv, with each line containing the frame number, average red value, average green value, and average blue value

Key Points

  • What are the key points?