A New Approach ToTemplate Matching And Image Recognition In Document Image
محل انتشار: اولین کنفرانس بازشناسی الگو و پردازش تصویر ایران
سال انتشار: 1391
نوع سند: مقاله کنفرانسی
زبان: انگلیسی
مشاهده: 992
فایل این مقاله در 5 صفحه با فرمت PDF قابل دریافت می باشد
- صدور گواهی نمایه سازی
- من نویسنده این مقاله هستم
استخراج به نرم افزارهای پژوهشی:
شناسه ملی سند علمی:
IPRIA01_122
تاریخ نمایه سازی: 11 مرداد 1393
چکیده مقاله:
The economic feasibility of creating a large database of document image has left a tremendous need for robust ways to access the information. Printed documentsare scanned for archiving or in an attempt to move towards a paperless office and are stored as images .In this paper a system for analysis and automatic indexing ofimaged documents for high-volume applicationsis described.The categorization of these documentimages isdone in terms of the presence of a certain graphical logodetected without segmentation[1].In this project we use the correlation for template matching and for measuringsimilarity.we present our technique of image processing for LOGO spotting and strategies in applying DTW for template detection with different poses and variations. Weelaborate through discussion and experiments’ results that how our image processing strategy and weighting schemeapplied to DTW algorithm effectively increase the performance and detection rate. In our technique, we haveused cross correlation for base of our work and improvedour results by DTW technique [5]. and also we tried other different techniques such as normalized centroid-radii model for achieving the best answer and comparing the results for finding the more efficient way[5].
نویسندگان
Omid Naderi
Computer Science Department University of Science And Research Birjand
Mehdi Shahedi
Computer Science Department University of Science And Research Birjand
Mehdi Razavi
Computer Science Department University of Science And Research Birjand
Javad Sardi
Computer Science Department University of Science And Research Birjand