“…In recent years, researches on fish image detection segmentation based on deep learning have attracted great attention. Chen, Sun & Shang (2017) proposed an automatic fish classification system based on deep learning, Akgül, Çalik & Töreyın (2020) proposed fish detection in turbid underwater, Li, Tang & Gao (2017) proposed a deep and lightweight network for detecting fish, Wang et al (2020) proposed a detection of abnormal behaviors of underwater fish using artificial intelligence techniques, Alshdaifat, Talib & Osman (2020) proposed a deep learning framework for segmentation of fish with underwater videos, Knausgård et al (2022) proposed a method for detecting and classificating temperate fish, Parida (2019) proposed a hybrid transition region color image segmentation method based on dual transition region extraction for fish image segmentation applications. Lei, Ouyang & Xu (2018) proposed an image segmentation method based on equivalent 3D entropy and artificial fish population optimization algorithm, which is more efficient than the traditional 3D entropy method and the equivalent 3D entropy method.…”