An Automated Fish Counting Algorithm in Aquaculture Based on Image Processing
Jiuyi Le, Lihong Xu
Available Online March 2017.
- https://doi.org/10.2991/ifmca-16.2017.56How to use a DOI?
- Fish Counting; Free-swimming; Overlap; Skeleton Extraction
- A new algorithm based on endpoints of skeleton is presented to efficiently get the number of fish in this paper. Considering the complexity of underwater environment like lack of light, this paper presents an improved adaptive thresholding method to segment the fish image better. In addition, the object of our research is free-swimming fish. The overlapped fish in the image makes the counting result inaccurate often. So after segmentation and morphological processing, this paper adopts image thinning method to extract the skeleton of fish. After that, we get the fish number according to the number of corresponding endpoints in the image. The experimental results show that the method can accurately count the fish population even under high overlapped degree.
- Open Access
- This is an open access article distributed under the CC BY-NC license.
Cite this article
TY - CONF AU - Jiuyi Le AU - Lihong Xu PY - 2017/03 DA - 2017/03 TI - An Automated Fish Counting Algorithm in Aquaculture Based on Image Processing BT - Proceedings of the 2016 International Forum on Mechanical, Control and Automation (IFMCA 2016) PB - Atlantis Press SP - 358 EP - 366 SN - 2352-5401 UR - https://doi.org/10.2991/ifmca-16.2017.56 DO - https://doi.org/10.2991/ifmca-16.2017.56 ID - Le2017/03 ER -