基于OpenCV手机拍照快递单文字识别的研究

Study on recognition character in express list by mobile phone camera based on OpenCV

基于Open CV在Android手机上设计了快递单文字识别机制,识别内容主要分为电话和姓名识别.手机拍摄快递单后,在屏幕上画出两个矩形框,根据矩形框的周长面积特征,判定电话和姓名区域并提取这两部分图片;对这两个图片进行灰度化、二值化、形态学处理,再进行字符分割、归一化处理.为了提高数字分割速度提出一种基于轮廓检测的分割方法.根据数字和汉字的不同特征选取不同的识别方法进行字符识别.

In this paper , a system of recognition character in express list on the Android mo-bile was designed based on OpenCV .Identification content is mainly divided into telephone numbers recognition and names recognition .Mobile phone was used to take a photo of ex-press list and draw two rectangular boxes on the screen .According to the perimeter area of rectangular box , the phone numbers and the names area were determined , and the two parts of picture were extracted .The two pictures were grayed , binarized , morphology processed and then character segmented , normalization processed .In order to increase the speed of digital divide , a segmentation method was presented based on contour detection .Finally , ac-cording to the different characters of the numbers and the Chinese characters , different recog-nition methods were selected for the character recognition .