Font Size: a A A

Research And Application Of Answer Card Mobile Phone Intelligent Recognition

Posted on:2019-04-20Degree:MasterType:Thesis
Country:ChinaCandidate:Q ZhangFull Text:PDF
GTID:2348330545499472Subject:Computer technology
Abstract/Summary:PDF Full Text Request
In recent years,the automatic answer sheet recognition system based on digital image processing has attracted attention because of its low hardware cost,easy operation,and ease of popularization.If we use a mobile phone that can be photographed everywhere to get answer card images,combined with digital image processing technology,we can quickly identify answer cards without special equipment.While ensuring the efficiency of answer sheet recognition and reducing costs,it also increases the mobility of answer sheet recognition work.This paper aims at the school's usual test scene and studies the intelligent recognition system based on the objective answer card.Through the processing of the answer card image collected by the mobile phone,the candidates filled in with the answer and handwritten character information are identified,and compared with the standard answer,and the corresponding scores are calculated and entered into the student database.The main research content is as follows:(1)For the problem of image shadow caused by uneven illumination during acquisition,the method of averaging and background difference is used to deal with the problem.For the problem of image tilt,the edge image is detected by Hough transform,and then the edge straight line is filtered to obtain the image tilt angle.This method reduces the Hough detection computation and improves the processing efficiency.(2)Design the objective problem recognition criteria and achieve the objective problem identification;Combining the template matching algorithm with the data structuring feature,it can recognize the handwritten characters of the answer sheet and ensure the recognition accuracy.(3)The answer card smart phone recognition system was implemented.The recognition speed under different operating systems(PC,Android,and IOS)is within 2s,the answer card recognition rate under the standard fill-in is 100%,and it can meet the application requirements.
Keywords/Search Tags:image processing, pattern recognition, slant correction, answer sheet recognition
PDF Full Text Request
Related items