Steel coil end defect automatic detection method based on computer vision technology

• Main Authors: QIN YI, GAO JIARUI, ZENG LONGHUA, LIU WENSHENG
• Format: Patent
• Language: Chinese; English
• Published: 19.09.2023
• Subjects: CALCULATING; COMPUTING; COUNTING; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS

The invention relates to the technical field of steel coil automatic detection, and discloses a steel coil end defect automatic detection method based on a computer vision technology, which comprises the following steps: S1, adopting an OCRNet semantic segmentation algorithm, taking a collected, screened and labeled end detection defect image as a data basis, realizing end detection defect identification, and obtaining an end detection defect identification result; the CORNet algorithm explicitly converts a pixel classification problem into an object region classification problem, and explicitly enhances object information through the OCR method, so that finer defects are identified. The end defects are automatically recognized through the computer vision technology instead of manual confirmation, meanwhile, the image brightness enhancement technology means is added on machine vision, the end defects are easier to recognize, strip steel end defect recognition automation is achieved, manual recognition is repl