铝合金MIG焊熔池的脉冲耦合神经网络边缘检测

Welding pool edge detection of aluminum alloy MIG welding by using pulse coupled neural network

针对铝合金MIG焊熔池边缘检测问题,采用脉冲耦合神经网络检测铝合金MIG焊熔池边缘.介绍脉冲耦合神经网络算法的原理,利用MATLAB软件平台进行基于脉冲耦合神经网络的铝合金MIG焊熔池图像的边缘检测,并将结果与用Canny等算子提取的结果进行对比分析.结果表明,应用脉冲耦合神经网络算法进行熔池图像边缘检测是可行的,而且脉冲耦合神经网络算法提取的铝合金MIG焊熔池边缘图像清晰、连续,有效地克服熔池图像中的噪声影响,此算法的单帧处理时间为54ms.

Aimed at the problem of detection of welding pool edge of aluminum alloy MIG welding,a method of pulse coupled neural network algorithm was adopted for edge detection of welding pool.The principle of pulse coupled neural network algorithm was introduced first.Then,on the platform of software MATLAB,the edge detection of welding pool image was conducted with pulse coupled neural network.Finally,the detection result obtained with the algorithm presented in this paper and that with canny operator were compared and analyzed.The result showed that it was feasible to use the pulse coupled neural network algorithm for detecting the edge of welding pool image.The extracted edge was distinct and continuous and the influence of noise on welding pool image was effectively eliminated.By using this algorithm,the processing time of single frame of image was 54 ms.