Computer Vision and Feeding Behavior Based Intelligent Feeding Controller for Fish in Aquaculture
In aquaculture, the feeding technology determined the feed conversion rate and cost. However, the intelligence of existing feeding devices is not very high. they can’t change the amount of feed according to the fish appetite automatically. In order to solve the above issues, in this paper, a feeding controller based on machine vision and feeding behavior was designed on the basis of the original feeder. The hardware platform was built on the I.MX6 microcontroller, and the software was designed via the embedded Linux OS. Moreover, the feeding behavior analysis and automatic feeding control method based on image processing were also studied. Firstly, the images of fish feeding process were collected and analyzed. Then the Delaunay Triangulation was used to extract the feeding behavior parameter FIFFB (flocking index of fish feeding behavior). Finally, the feeding decision was made according to the defined threshold. Compared with the traditional feeder, the controller designed in this paper is more intelligent and can reduce feed waste. Meanwhile, water pollution also can be reduced. The automatic feeding control was realized during feeding process.
KeywordsComputer vision Feeding behavior Intelligent control Aquaculture
The research in this paper was supported by the National Key Research and Development Program of China (2017YFD0701705) and the Beijing Excellent Talents Development Project (2017000057592G125).
- 1.Rola, W.R., Hasan, M.R.: Economics of aquaculture feeding practices: a synthesis of case studies undertaken in six Asian countries. FAO, Rome (2007)Google Scholar
- 3.Hasan, M.R., New, M.B.: On-farm feeding and feed management in aquaculture. Food & Agriculture Organization of the United Nations (2014)Google Scholar
- 14.Wyatt, T.: Some effects of food density on the growth and behaviour of plaice larvae. Mar. Biol. 14(3), 210–216 (1972)Google Scholar
- 20.Semiconductors, N. i.MX 6Dual/6Quad Applications Processors for Industrial Products (2017). http://cache.nxp.com/assets/documents/data/en/data-sheets/IMX6DQIEC.pdf?fsrch=1&sr=2&pageNum=1
- 25.Beucher, S., Lantuéjoul, C.: Use of watersheds in contour detection. In: International Workshop on Image Processing, Real-Time Edge and Motion Detection, Rennes,France, pp. 391–396 (1979)Google Scholar
- 26.Beucher, S.: Watersheds of functions and picture segmentation. In: IEEE International Conference on Acoustics, Speech, and Signal Processing, pp. 1928–1931. IEEE, Paris (1982)Google Scholar
- 29.Thiessen, A.H.: Precipitation averages for large areas. Mon. Weather Rev. 39(7), 1082–1084 (1911)Google Scholar