Many countries have set up ecological coastal preserves for sustainable oceans and ecological conservation and prohibiting people from angling along the shoreline. Although there are regular patrols by law enforcement officers, there are still many anglers who do not comply with the regulations, and intrude into the ecological coastal preserve. This requires a lot of human resources and time for law enforcement officers. Therefore, we developed a fishing detection system by collecting images of anglers’ fishing behaviors. We used a deep learning model with object detection and posture recognition. When anglers engaged in fishing behaviors, the system immediately transmitted the anglers’ geographic locations to the back-end database and notified the law enforcement officers. The results showed that the system’s accuracy rate in determining angling postures was as high as 90%.
Fishing Posture Detection Model Based on Convolutional Neural Network
01.12.2023
858157 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Underwater posture adjusting device for fishing frame
Europäisches Patentamt | 2022
|Lateral distance detection model based on convolutional neural network
Wiley | 2019
|Convolutional Neural Network GNSS-R Sea Ice Detection Based on AlexNet Model
Springer Verlag | 2022
|