The Study of Computer Vision Algorithms for Underwater Fish Inspection

Updated in September, 2021

Computer vision is an interdisciplinary field that deals with how computers can be used to gain high-level understanding from digital images and videos. Fish detection and monitoring is an important topic in computer vision. The growth of high-powered computers, the evolution of high-quality video cameras with low cost and the growing need for automated video analysis have caused more interest in the development of underwater inspection.

The aim of the project is to develop algorithms  for remote observations in coastal and oceanic waters by combining new sensor technologies with commercially available sensors. The project is focused on the advancement of aquaculture technology systems that are economically sustainable and improve animal welfare. This includes the fish quantification for prevention of fish species as well as salmon, and even develop procedures for underwater monitoring and safe and efficient operation.

Quantifying human impact on fish biodiversity in order to propose solutions to preserve submarine ecosystems is an important line of research for marine ecology. The project aims to integrate computer vision techniques to enhance visibility in submarine environment. It aims also to design new automated methods for scene understanding of underwater images, in order to facilitate their interpretation. 

The system will have the potential to reduce human intervention to the minimum, revolutionising the affordability of a broad range of surveillance and data collection operations. Two different tests will be taken in consideration:

• Tests in artificial environment
• Tests in real environment

Postdoctoral research project

Project manager: Dr.sc.comp. Mohcine Boudhane (This email address is being protected from spambots. You need JavaScript enabled to view it.)

Supported by: ERDF

Project duration: 15.03.2019. - 14.03.2022.

Project number: 1.1.1.2/VIAA/2/18/348

Project partner: ENSET, Mohammed V University, Rabat, Morocco

 ​

Activities implemented:

1. Participation in the International Conference in the international Scientific Conference SOCIETY. TECHNOLOGY. SOLUTIONS, title: "Underwater optical observation, how to improve visibility" (24-25.04.2019)
2. Participation in  on new research methods, presentation on the topic "MACHINE LEARNING AND DATA SCIENCE" (15.05.2019)
3. Participation in FALLING THE WALL LAB, organized by the German Academic Exchange Service (DAAD) in cooperation with the university of Latvia, presentation title "BREAKING THE WALL OF UNDERWATER VISIBILITY" (05.06.2019)
4. Description of Activity 1.1.a "Visibility problems and underwater image enhancement methods" (28.06.2019)
5. Description of  Activity 1.1.b "Technologies on object detection  tracking in video sequences"  (28.06.2019)
6. Description of  Activity 1.1.c "Evaluative and comparative study of shape modelling methods" (28.06.2019)
7. Lecture by Dr. Thomas Blaschke "Object-Based Image Analysis: integrating GIS and remote sensing for environmental studies", Institute of Electronics and Computer Science, Riga (16.08.2019)
8. Participation in the International conference on Mathematics and computers in Science and Industry with presentation "A Method for Underwater Image Enhancement using Histogram Regularization", Corfu Island, Greece (24.-26.08.2019)
9. Participation in the International conference on Watermarking and Image Proecessing ICWIP with presentation "Underwater Exploration Issues, Deep Study on Optical Underwater Vision for an Effective Traditional Fishing", Marseille, France (18.-20.09.2019)
10. Project result dissemination in the European Researchers' Night 2019, Valmiera, Latvia (27.09.2019)
11. Publication Boudhane, M., & Balcers, O. (2019). Underwater Image Enhancement Method Using Color Channel Regularization and Histogram Distribution for Underwater Vehicles AUVs and ROVs. INTERNATIONAL JOURNAL OF CIRCUITS, SYSTEMS AND SIGNAL PROCESSING, 13, 570-578, ISSN: 1998-4464 (September 2019)
12. Project dissemination: presentation at Monthly Postdoctoral Seminar (18.12.2019)
13. Fish tank demonstration day (19.12.2019)
14. Database: https://www.dropbox.com/sh/n4oiyi6vvsn0ti6/AABXrTuiSE5RaqthGelTG72_a?dl=0 (March, 2020)
15. Presentation “Artificial Intelligence and Computer Vision, Applications in Oceanography” at the seminar Dabas un Tehnoloģiju Parks, Urda, Latvia, 12.03.2020. 

16. SCIENTIFIC PAPERS (WP2): 

Boudhane M., & Balcers, O. (2020). Underwater Optical Observation, How to Improve Visibility. Baltic Journal of Modern Computing, 8. DOI: 10.22364/bjmc.2020.8.1.09 (January 2020) (Web of Science)

Boudhane, M., & Balcers, O. (2019). Underwater Image Enhancement Method Using Color Channel Regularization and Histogram Distribution for Underwater Vehicles AUVs and ROVs. INTERNATIONAL JOURNAL OF CIRCUITS, SYSTEMS AND SIGNAL PROCESSING, 13, pp. 570-578, ISSN: 1998-4464 (December 2019) (Scopus)

17. Conference thesis:

Pratt, M., Boudhane, M., & Cakula, S. (2020). Predictive Data Analysis Model for Employee Satisfaction Using ML Algorithms. International Conference of Advanced Computing and Informatics, Casablanca, Morocco, 13th - 14th April, project synergy with the "Support for Effective Participation of Vidzeme University of Applied Sciences within the International Scientific Circles (ViA-lnt)" Nr. 1.1.1.5/18/I/005.

Boudhane, M., Azbeg, K., Toulni, H., Balcers, O. (2021). A Method-based Feature Extraction and Object Recognition for Underwater Imaging. International Scientific Conference SOCIETY. TECHNOLOGIES, SECURITY, March 25th-26th, VIDZEME UNIVERSITY OF APPLIED SCIENCES, VALMIERA.

18. Business trip and research at the Daugavpils University, Latvia (25.06-31.07.2020). Work at the lake Svēte, Tetele (Latvia). Presentation of the project results at the Daugavpils University, 01.07.2020 "Progress and the planed testing methodology".
19. Summer school Security of Things at the Univeristy of Rostock (24.08-24.09. 2020).
20. Reports of the research results: Technical Data Sheet, R4; Feature engineering and tracking of live fish, R6.
21. 3D design of the underwater robot prototype, WP2.1.
22. CAD underwater robot design (Face view, Side view, Top view, WP 2.1).
23. DAAD AI Networking (German Academic Exchange Service), March 24th-25th, 2021. WP 2.2, WP3.
24. Mobility to ENSAM university in Rabat, Morocco, August 2021-March 2022