Yolov5-based defect detection method, apparatus and device, and storage medium

• Main Authors: TANG FENGYING, CAI CHUANBAO, CHEN GAO
• Format: Patent
• Language: Chinese; English
• Published: 03.10.2023
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS

The invention discloses a defect detection method and device based on Yolov5, equipment and a storage medium. The method comprises the steps that at least one defect picture is acquired; inputting the defect picture into a target defect detection network based on Yolov5, outputting a target frame corresponding to a defect in the defect picture, and/or outputting a target defect category corresponding to the defect in the defect picture; wherein the target frame is used for representing the position of a defect in the defect picture in the defect picture, the target defect category is one of a plurality of preset defect categories, and the target defect detection network is determined by performing channel pruning on Yolov5. By utilizing the method disclosed by the invention, computing resources required by defect detection can be effectively reduced. 本申请公开了一种基于Yolov5的缺陷检测方法、装置、设备及存储介质，所述方法包括：获取至少一张缺陷图片；将所述缺陷图片输入基于Yolov5的目标缺陷检测网络，输出所述缺陷图片中的缺陷对应的目标框，和/或输出所述缺陷图片中的缺陷对应的目标缺陷类别；其中，所述目标框用于表示所述缺陷图片中的缺陷在所述缺陷图片中的位置，所述目