基于局部自适应阈值与K近邻算法的空气滤芯漏粘识别  

Identification of Air Filter Element Leakage Based on Local Adaptive Threshold and K-Nearest Neighbor Algorithm

作　　者：高雅昆 高小红 胡永涛 吴超 郭华 Gao Ya-kun;Gao Xiao-hong;Hu Yong-tao;Wu Chao;Guo Hua(School of Electrical Engineering and Automation,Henan Institute of Technology,Henan Xinxiang 453003;Library,Henan Institute of Technology,Henan Xinxiang 453003)

机构地区：[1]河南工学院电气工程与自动化学院,河南新乡453003 [2]河南工学院图书馆,河南新乡453003

出　　处：《内燃机与配件》2023年第23期106-108,共3页Internal Combustion Engine & Parts

基　　金：河南省高等学校重点科研项目(21B413002);河南工学院高层次人才科研启动基金(KQ2001,KQ2109)。

摘　　要：空气滤芯产品生产时如果出现漏粘会导致空气直接进入发动机,过滤失效。针对该情况,设计了基于局部自适应阈值与K近邻算法的空气滤芯漏粘识别算法:首先,利用局部自适应阈值分割算法对滤芯图像分割,并通过区域标号算法定位到粘胶亮孔;然后,利用K近邻算法以每个粘胶亮孔为中心,通过该中心亮孔与周围近邻亮孔的灰度相似性,判断中心孔是否为漏粘孔。实验表明所提算法漏粘识别率达到99%,有效提高滤芯产品合格率。If there is leakage during the production of air filter element products,air will directly enter the engine and the filter will be failure.In view of this situation,an air filter element leakage identification algorithm based on local adaptive threshold and K-nearest neighbor algorithm was designed.Firstly,the filter element image was segmented by local adaptive threshold segmentation algorithm,and the viscose bright hole was located by region label algorithm.Then,the K-nearest neighbor algorithm is used to determine whether the central hole is a leaky hole through the gray similarity between the central bright hole and the neighboring bright hole.The experimental results show that the identification rate of leakage can reach 99%,and the qualified rate of filter element can be improved effectively.

关 键 词：空气滤芯 局部自适应阈值分割 K邻算法 漏粘识别 

分 类 号：TP13[自动化与计算机技术—控制理论与控制工程]