摘要
SF_(6)充气设备仪表短指针特点为非直线连接圆心的指针。首先,利用均值迁移、图像灰度化、霍夫圆环变换图像处理技术对SF_(6)仪表图像进行处理,并裁剪SF_(6)仪表表盘;然后使用高斯滤波、自适应阈值和轮廓查找对仪表图像进行处理,提取出仪表指针特征信息;最后,利用坐标变换和夹角信息计算得到SF_(6)压力仪表指针读数。实验结果表明,该方法能准确检测得到SF_(6)仪表指针,不仅具有较高的精度,而且还具有较强的实际应用价值。
The paper proposes a recognition method based on digital image processing technology for the reading and recognition of SF_(6) filling equipment instrument numbers.First,the SF_(6) instrument image is processed by means of mean valve migration,image grayscale,and Hough circle transformation image processing technology,and the SF_(6) instrument dial is cropped.Then,use Gaussian filtering,adaptive threshold and contour search to process the meter image to extract the meter pointer feature information.Finally,the SF_(6) pressure gauge pointer reading is calculated using coordinate transformation and pinch angle information.The experimental results show that the method is very effective for instrument image region identification,which not only has high accuracy,but also has strong practical application value.
作者
曾令诚
李新海
林悦德
周恒
林蔚
温云龙
ZENG Ling-cheng;LI Xin-hai;LIN Yue-de;ZHOU Heng;LIN Wei;WEN Yun-long(Zhongshan Power Supply Bureau of Guangdong Power Grid Co.Ltd.,Zhongshan 528401,China)
出处
《电气开关》
2021年第6期30-34,共5页
Electric Switchgear
基金
广东电网有限责任公司科技项目资助GDKJXM20190154(032000KK52190001)。
关键词
SF_(6)压力仪表识别
霍夫圆环变换
均值迁移
仪表读数识别
图像二值化
SF_(6)pressure gauge identification
Hough circle transformation
mean valve migration
meter reading recognition
image binarization