Target detection method based on improved YOLOv8 in complex environment

CN119027643APending Publication Date: 2024-11-26JIANGSU UNIV OF SCI & TECH
View PDF 1 Cites 0 Cited by

Patent Information

Application Number
CN202410950712.3
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-16
Publication Date
2024-11-26

Smart Images

  • Figure CN119027643A_ABST
    Figure CN119027643A_ABST
Patent Text Reader

Abstract

The invention discloses a target detection method based on improved YOLOv8 in a complex environment, and the method comprises the following steps: (1) downloading an image through an open source website, and obtaining a multi-scene and multi-unit image data set through manual screening; (2) carrying out manual annotation on the obtained target image data in the complex environment by utilizing a labmeli tool; (3) introducing the CBAM attention mechanism, the AFPN progressive feature pyramid network and the InnerIoU loss function into a YOLOv8 model, and constructing an ASYOLOv8 model; (4) inputting the marked target image data set into an ASYOLOv8 model for training; (5) performing target detection on a target image to be detected by using the trained ASYOLOv8 model; according to the invention, the accuracy, reliability and generalization ability of small target detection in a complex environment are improved.
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Tunnel video monitoring protection equipment detection method under weak light condition

    CN117746011A