Thesis subject

Machine Learning for Evaluating Pet Health

Note: This is a cooperative project between INF group and Animal Nutrition group of Wageningen University & Research.

In this project, we aim to develop a solution to evaluate the quality of pet food in-home. One of the challenges is to obtain information from the pet owner. To determine the quality of pet food, the perspective of the pet owner is of high importance. We are exploring possibilities to score the health of animals from an objective perspective.

In this project, we focus on scoring the feces score and the body condition score of dog or cat. Specifically, we ask the pet owners to take pictures of the pets and to send them to the server. After that, a machine learning-based solution will score the pets' body condition by detecting the feces pictures.

background
The following tasks will be in this project:

  • Identify the key features that need to be recognized in the pictures.
  • Collect pictures of dog feces and link a score to these pictures.
  • Use machine learning to recognize and score new pictures.
  • Link the feces score to the body condition score of the animal.

The content above is an overall description of the project. The detailed research work of the project could be based on further discussion between supervisors and students.

    Literature

    Requirements (optional)

    • Pyhon programming
    • Machine learning

    Contact person

    • qingzhi.liu@wur.nl