Rectification of images for camera based OCR systems /
تصحيح صور آلات التصوير لتجهيزها لنظم التعرف على الحروف
Mohamed Fawzi Abdalrasoul Afifi ; Supervised Mohsen Abdelrazik Rashwan
- Cairo : Mohamed Fawzi Abdalrasoul Afifi , 2015
- 90 P. : charts , facsimiles ; 30cm
Thesis (M.Sc.) - Cairo University - Faculty of Engineering - Department of Electronics and Communications
Optical character recognition (OCR) works with scanners to convert printed characters into digital text. Current advance in cameras lens industry introduced the cameras as an alternative to scanner for document image capturing. Compared to scanners, cameras are considered to be more convenient way to capture document images. However, document images captured by camera suffer from problems such non uniform illumination, resolution and distortion which affect the OCR performance. We introduce new techniques to overcome the distortion combined with capturing Arabic document image by camera.