摘要
课表图片区别于人脸及其他常见的图像识别对象,而对其信息的提取可使无课表的制作更加便捷。针对课表图像的视角校正、图像分割、文字信息提取等问题,本文基于Hough变换和透视变换等常用的图像处理方法,设计课表图片的信息自动提取算法,并结合实例,对算法的可行性和实用性进行验证测试。结果表明,该算法能够很好地提取出实例中课表的课程信息。
Schedule pictures are different from faces and other common image recognition objects, and the extraction of information can make the production of no-schedule more convenient. Aiming at the problems of visual angle correction, image segmentation and text information extraction of timetableimages,mationextractionalgorithmfortimetableimagesbasedonthecommonlyusedimageprocessingmethodssuchasHough transformandperspectivetransform,sultsshowedthatthealgorithmcouldextractthecourseinformation of the timetable in an example very well.
引文
[1]李牧,闫继红,李戈,等.自适应Canny算子边缘检测技术[J].哈尔滨工程大学学报,2007(9):1003-1007.
[2]CANNY J F. A computational approach to edge detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence,1986(6):679-698.
[3]张少伟.基于机器视觉的边缘检测算法应用与研究[D].上海:上海交通大学,2013.
[4]王化楠. Hough变换在视觉检测系统中的应用研究[D].大连:大连理工大学,2006.
[5] Kiryati N,Eldar Y,Bruckstein A M. A probabilistic Hough transform[J]. Pattern Recognition,1991(4):303-316.
[6]张振杰,郝向阳,刘松林,等.基于Hough一维变换的直线检测算法[J].光学学报,2016(4):166-173.