摘要
为精确计算铜铝过渡线夹钎焊结合面的缺陷面积,根据颜色空间开发了面积计算软件。软件基于HSV颜色空间实现缺陷识别,根据图像尺寸的输入计算缺陷面积大小。测试结果表明,图像识别软件能精准识别相控阵超声成像检测仪获得的钎焊结合面缺陷图像,并根据图像比例计算准确获得缺陷面积,计算误差在1.2%以内。
To accurately calculate the defect area of the soldering joint in the copper-aluminum transition clamp,software for area calculation was developed based on the color space.The software uses the HSV color space to identify defects and calculates the size of the defective area based on the dimensions of the input image.Test results indicate that the image recognition software can accurately identify defective images of the soldering joint captured by the phased array ultrasonic imaging detector.Furthermore,the defect area is accurately calculated based on the image proportions,with a calculation error of less than 1.2%.
作者
余超
王利昭
马延会
王一波
季昌国
王智春
张鹏鲲
YU Chao;WANG Lizhao;MA Yanhui;WANG Yibo;JI Changguo;WANG Zhichun;ZHANG Pengkun(North China Electric Power Research Institute Co.,Ltd.,Beijing 100045,China)
出处
《无损探伤》
2024年第3期17-20,共4页
Nondestructive Testing Technology
关键词
图像识别
面积计算
颜色空间
铜铝钎焊结合面
Image recognition
Area calculation
Color space
Soldering joint surface of copper-aluminum