Improved YOLOv8-based land-based round pool dead fish detection method

CN119027977APending Publication Date: 2024-11-26GUANGXI UNIVERSITY OF TECHNOLOGY +2
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202410877563.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-02
Publication Date
2024-11-26

Smart Images

  • Figure CN119027977A_ABST
    Figure CN119027977A_ABST
Patent Text Reader

Abstract

The invention discloses a land-based round pond dead fish detection method based on improved YOLOv8, relates to the technical field of land-based round pond culture, and solves the problems that a traditional dead fish detection method mainly depends on manual observation and physical labor, the manual observation may cause missing detection or misjudgment, and especially in a large-scale farm, the manual labor may cause missing detection or misjudgment. The problems that a large amount of manpower and time investment is needed, efficiency is low, the method is not economical, due to periodicity of manual inspection, dead fish can be possibly detained in a culture pond for a long time, and the risk of water quality pollution and disease transmission is further aggravated are solved. A land-based round pond dead fish detection method based on improved YOLOv8 comprises the following steps that firstly, dead fish conditions are artificially manufactured and used for enriching a dead database of cultured fishes, and the selected fish variety is procypris merus; data comparison analysis and detection are performed on the dead fishes in the land-based round pool, and the dead fishes at the detected position are fished out of the land-based round pool, so that the dead fishes are prevented from polluting water quality and spreading diseases.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Cited By

  • Landslide detection method, system and equipment based on target detection model and medium

    CN119919805A

  • Automatic detection method for feeding behavior of fish shoal under complex background and feeding control system

    CN120953646A