Research On Cattle Mould Inspection System Based On Machine Vision

In the current production of high-grade car seats,the correct position of the molds for the leather forming process is essential.The current mold recognition is achieved through the human eye.But there are many drawbacks such as slow speed,poor real-time,low utilization of leather.Not only that,the labor intensity of the workers is high and the working environment is poor for workers and the employee cost is high for the enterprise.In order to solve the above problems,this paper designs a set of leather molds detection system based on machine vision to replace the artificial way and achieve the detection and identification in automated way.Leather molds detection and identification in the system consists of two parts:label positioning and character recognition.As the operating platform is too large,so there are two CMOS cameras in the system.The images of the molds captured by the camera are transmitted to the IPC(Industrial Personal Computer).Software system in the IPC completes the image processing,detection and identification.The main works of this study are:(1)The human computer interaction interface for camera calibration in real time;(2)Research on image stitching technology;(3)Design of the brightness adjustment algorithm based on radiation calibration in the case of great difference in exposure.The image brightness is optimized to facilitate the recognition;(4)Design of specific mold labels.Based on morphological image processing and character recognition technology,the tag is positioned and identified.Based on the above techniques,this study realizes the automatic identification of leather processing molds,the statistics and management of molds parameters.The leather mold visual inspection system is developed in the Windows system.The main compiler is Visio Studio 2013.The image processing module in the software system is based on the open source library OpenCV and the visual human-computer interaction interface is designed by Qt designer.