选票图像识别算法  

Algorithm on ballot image recognition

作　　者：韩里洋 崔喆[1,2] 

机构地区：[1]中国科学院成都计算机应用研究所,成都610041 [2]中国科学院大学,北京100049

出　　处：《计算机应用》2016年第A02期179-181,209,共4页journal of Computer Applications

基　　金：四川省科技支撑计划项目(2015GZ0088);"西部之光"联合学者项目

摘　　要：针对计算机选举系统中基于击中/击不中变换的选票倾斜度检测算法效率较低,难以检测包含内部噪声的选票定位块,且需要选票两侧同步道定位候选人信息块的问题,提出利用开运算和闭运算进行定位块及倾斜度检测,并利用候选人文本投影法定位候选人信息块的算法。实验结果表明,该算法比传统倾斜度检测算法运行时间减少了71%,且能保证候选人信息块在无同步道环境下定位的运行速度,这对于民主选举行为的高效进行具有实际的应用意义。Concerning the inefficient algorithm for inclination detection of ballot image based on hit/miss transformation, the difficulty to detect locating blocks which contain internal noise and the necessity for the synchronization tracks to locate information blocks in the computer electoral system, a ballot recognition method was proposed. Opening operation and closing operation were used to locate locating blocks and detect inclination while the candidate information blocks were located by candidate text projection. The experimental results show that more than 71% run-time would be saved with the speed of locating information blocks being guaranteed in the absence of synchronization tracks by the new method while compared with the traditional inclination detection algorithm. The new method has practical application significance to efficient democratic elections.

关 键 词：倾斜度检测 信息块 开运算 闭运算 候选人文本投影法 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]