温湿度检定仪中的数字式仪表识别  

Digital Identification in Thermo-hygrometer Calibrator

作　　者：祝长生[1] 张成元[1] 徐亚超 ZHU Chang-sheng;ZHANG Cheng-yuan;XU Ya-chao(College of Intelligent Equipment,Shandong University of Science and Technology,Tai’an 271019,China)

机构地区：[1]山东科技大学智能装备学院,泰安271019

出　　处：《科学技术与工程》2020年第34期14120-14125,共6页Science Technology and Engineering

基　　金：山东省重点研发计划(2019GGX101005);泰安市科技计划(2018GX0039)。

摘　　要：为了提高温湿度计检定仪的智能化水平,提出了一种基于不变矩信息融合的温湿度计数字自动识别算法。首先,通过Otsu法对图像进行分割,截取仪表盘;其次,通过Hough变换对图像进行校正;再次,通过Two-Pass算法定位数字区域,并采用非端点位置像素置零法细化数字图像;最后,采用Hu矩与Zernike矩融合的方法,提取图像分类特征,并利用现场实际采集的仪表图像对该算法进行了验证。结果表明,基于不变矩信息融合的数字识别算法降低了算法的复杂度,提高了检测效率。In order to make the thermo-hygrometer calibrator more intelligent,an automatic digital recognition algorithm of the thermo-hygrometer based on the invariant moment information fusion was proposed.Firstly,the Otsu method was used to segment the image and intercept the instrument panel.Secondly,the Hough transform was used to correct the image.Thirdly,the Two-Pass algorithm was used to locate the digit region,and the non-endpoint pixel zeroing method was used to refine the digital images.Finally,both Hu moment and Zernike moment were fused to extract the image classification features,and the algorithm was validated by the actual instrument images collected on-site.The results show that the proposed algorithm reduces the complexity of the algorithm and improves the detection efficiency.

关 键 词：不变矩 信息融合 深度学习 HOUGH变换 数字式仪表 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]