Recognition of Zinc Stripping Residues in Zinc Electrolytic Cathode Plates Based on Improved YOLOv5 Algorithm.

The cathode plates in the industrial production process of zinc electrolysis sometimes remain residues after being stripped by zinc stripping machines, which need to be identified and removed in time to avoid affecting the subsequent processing of the cathode plates. At present, the identification and rejection from the process are mainly carried out by visual observation, which makes it difficult to ensure the accuracy and timeliness of visual identification, and may result in unqualified cathode plates being mistakenly introduced into the electrolysis process. A cathode plate residue machine vision recognition method based on the improved YOLOv5 algorithm is proposed, and the method has been used for residue recognition experiments and tests on zinc cathode plate image data samples collected from industry. The results show that the average recognition accuracy of unqualified plates based on the improved YOLOv5 algorithm reaches 95.2%, which is 6.8% higher than the accuracy of the original YOLOv5 algorithm, and its recognition accuracy can meet the requirements of real-time detection of zinc cathode plate residues in practical applications. [ABSTRACT FROM AUTHOR]