Research On Detection Method Of Circumferential Surface Defect Of Cylindrical Lithium Battery

18650 cylindrical lithium battery is a kind of lithium battery widely used in portable devices,mobile power supply,reserve energy and other aspects.It has the characteristics of mature technology,high-quality cells,high automation of production process and stable product quality.It occupies a very high share in the market and is widely used.During the production process,due to the manufacturing process and inevitable bumps,the surface of lithium battery will have defects such as film breaking,pits and scratches,which will not only affect the appearance of lithium battery,but also cause potential safety hazards.At present,the detection methods of lithium battery mainly rely on manual detection,which has low efficiency and high rate of missed detection,and can not meet the requirements of modern manufacturing industry.In this paper,according to the characteristics of lithium battery circumferential surface,a circumferential defect detection system is designed by machine vision technology.This paper introduces the causes of various defects on the peripheral surface of lithium battery,and analyzes the imaging characteristics.The circumference of lithium battery is divided into two areas: bright area and dark area.At the same time,the key points and difficulties of detection are analyzed,and the overall detection scheme is designed.For the location of lithium battery,the contour of lithium battery is extracted for location.Contour extraction is mainly based on the idea of gray spatial extremum,and the interference signal of the circumference of lithium battery is reduced by one-dimensional mean filtering,and then the threshold is set according to the amplitude frequency distribution histogram.Two marking schemes were designed for the light and dark area of lithium battery.The contour was continuous by fixed template,and finally the contour of lithium battery was obtained.Compared with the traditional Edge detection Canny algorithm,the algorithm in this paper could extract more details in the dark area.For lithium battery circular rupture surface defects,the threshold value of the amplitude of adjacent concave convex segments is set according to the difference of the gray distribution curve between the film breaking defect and the oil stain to extract the defect area,so as to achieve the purpose of distinguishing the oil stain and the defect.The defect area is determined by morphological features.This method avoids the false detection caused by oil pollution to a certain extent.For lithium battery circular pits surface defects,pit defect are analyzed and the gray distribution curve of oil pollution,through the method of least square fitting of gray distribution curve on the surface of the lithium battery and compared with the original grayscale distribution curve,according to the two curves in a certain area of the sum of the gray scale difference characteristics to determine whether there is a pit defect.This method can effectively detect pit defects and reduce the error detection rate.