基于视觉传感的铝合金TIG焊熔池特征提取  

Feature extraction of molten pool in TIG welding for aluminum alloy based on vision sensing

作　　者：单俊[1] 熊震宇[1] 傅建安[1] 薛诚[1] 

机构地区：[1]南昌航空大学航空制造工程学院,江西南昌330063

出　　处：《电焊机》2015年第11期10-13,共4页Electric Welding Machine

基　　金：江西省教育厅科技项目(GJJ13502)

摘　　要：根据铝合金TIG焊特点,在充分考虑电弧光谱特征的基础上,选择合适的滤光组件,确定了单个面阵CCD摄像机安装结构参数,成功构建了面向铝合金TIG焊的被动视觉传感系统,获得了清晰的熔池图像。针对拍摄到的熔池图像,提出一套包括中值滤波、拉普拉斯锐化以及改进蚁群算法等的图像处理方法,对传统蚁群算法进行了选择策略和信息更新策略的改进,形成一种自适应检测方法,有效地提高了检测精度和检测时间,较好地提取了熔池边缘,为实现焊接质量在线监测与控制奠定了基础。According to the characteristics of aluminum alloy TIG welding, and on the basis of analyzing the characteristics of are spectrum, the visual sensing system is constructed successfully by choosing suitable light components and confirming the installation structural parameters of CCD camera. By means of the developed visual sensing system, the clear images of molten pool are obtained. Aiming at the captured images,a method of image processing is proposed,including median filtering,Laplace sharpening and improved ant colony algorithm. The traditional ant colony algorithm is improved with selection strategy and information update strategy, and forms a kind of self-adaptive detection method which can effectively improve detection precision and time. Consequently, the edge of molten pool is extracted ,and this would be helpful to the on-line monitoring and control of welding quality in future.

关 键 词：铝合金 钨极氩弧焊 视觉传感 图像处理 

分 类 号：TG444.72[金属学及工艺—焊接]