| The explosion-proof plate of lithium battery is the core part of the battery poles.It not only plays the role of conductive electrode,but also has the function of battery closure and safety valve.Its irregular assembly mode directly affects the service life of the battery and causes hidden trouble to the user’s personal safety.At present,the production line is full of manual inspection and assembly of explosion-proof pieces.Low inspection efficiency is a key factor restricting the production efficiency of the industry.In order to improve production efficiency,automatic intelligent detection on front and back of explosion-proof pieces has become the focus of manufacturers.In recent years,the rapidly developing machine vision technology is more and more widely used in product testing to improve the testing speed.Taking the circular aluminum explosion-proof pieces with a diameter of 17 mm of a battery manufacturer as the test object,according to the test requirements put forward by the customer,a kind of online detection system for front and back of explosion-proof pieces based on machine vision was developed.The system hardware is composed of CMOS gigabit network camera,FL-BC3518-9M lens,ring dome blue LED,optical fiber sensor,solenoid valve and computer.The on-line detection algorithm uses hough gradient method to locate the target.According to the circular features of the detected object,the ring at a specific distance is extracted and the gray value features of the area are obtained.The positive and negative threshold is determined by k-means clustering method,so as to achieve the purpose of positive and negative detection of the explosion-proof plate.The on-line testing equipment has been applied to a battery manufacturer and the equipment runs well.Compared with the traditional manual detection method,the online detection system can greatly improve the detection speed,reduce the detection error,improve the production efficiency and reduce the labor cost. |
