复杂背景下电子吊秤数字的定位识别方法  

A method of locating and identifying electronic hanging scale digits in complex background

作　　者：郭彩霞 GUO Caixia(Shaanxi Province Key Laboratory of Advanced Manufacturing and Evaluation of Robot Key Components Co-founded by Baoji and the Shaanxi Department of Science and Technology,Baoji University of Arts and Sciences,Baoji 721016,China)

机构地区：[1]宝鸡文理学院、陕西省机器人关键零部件先进制造与评估省市共建重点实验室,陕西宝鸡721016

出　　处：《现代电子技术》2023年第17期48-52,共5页Modern Electronics Technique

基　　金：陕西省2021年重点研发计划(2021GY⁃326)。

摘　　要：电子吊秤计量大多以人工操作为主,计量过程中重复性操作多、自动化程度低,针对该问题,提出一种基于计算机视觉的电子吊秤数字的自动识别方法。该方法通过颜色特征定位电子吊秤,利用透视变换对电子吊秤图像进行倾斜矫正,通过轮廓特征获得感兴趣数字区域,用电子吊秤单行字体样本训练光学字符识别引擎Tesseract⁃OCR,生成独立的数字库,并用该库识别电子吊秤的数字,提高了文字识别率。应用该方法对三种电子吊秤的图像进行处理和识别,结果表明该方法可有效实现电子吊秤数字的自动识别。The metrological work of the electronic hanging scale is operated mostly manually,with many repetitive operations and low automation in the process of metering.Therefore,an electronic hanging scale automatic digital recognition method based on computer vision is proposed.In this method,the color feature is used to locate the electronic hanging scale,the perspective transformation is used to correct the tilt of the electronic hanging scale image,the ROI(region of interest)is obtained by contour features,and the optical character recognition engine Tesseract⁃OCR(optical character recognition)is trained by the single text line font samples for electronic hanging scales,so as to generate an independent digital library.The library is used to recognize the digits of electronic hanging scale,and the character recognition rate is improved.This method is used to process and recognize the images of three kinds of electronic hanging scales.The results show that the method can effectively achieve automatic digital recognition of electronic hanging scales.

关 键 词：计算机视觉 文字识别 OPENCV Tesseract⁃OCR 感兴趣区域 透视变换 BGR HSV 

分 类 号：TN911.73-34[电子电信—通信与信息系统]