Отрывок: ) P ( ) P ( ) P ( ) 1 y x y            r r r r r 32 33 34{2} 2 31 32 32 3211 3 3 2 31 32 32 32 35 362 2 2 32 32 32 362 32 32 38{2} 3722 3 3 31 32 1( ) = ( (2 ( ) ( ) ( 2 ( ))) ( ( ) ( )) ( ) 2 ( (2 ( ) ) 2( ( ) ) ( )) ( ) ( ) ), 4 1( ) = ( (2 2 x x x x x c c cF c xy z x l yp yp z x p y z x l l p p p p c cp y z x p y z x p y l Z l c y p p cF c xy p p      ...
Название : Optimal affine image normalization approach for optical character recognition
Авторы/Редакторы : Konovalenko, I.A.
Kokhan, V.V.
Nikolaev, D.P.
Ключевые слова : optical character recognition
image registration
image normalization
coordinate discrepancy
projective transformation
affine transformation
approximation
optimization
symbolic computation
Дата публикации : Фев-2021
Издательство : Самарский национальный исследовательский университет
Библиографическое описание : Konovalenko IA, Kokhan VV, Nikolaev DP. Optimal affine image normalization approach for optical character recognition. Computer Optics 2021; 45(1): 90-100. DOI: 10.18287/2412-6179-CO-759.
Серия/номер : 45;1
Аннотация : Optical character recognition (OCR) in images captured from arbitrary angles requires preliminary normalization, i.e. a geometric transformation resulting in an image as if it was captured at an angle suitable for OCR. In most cases, a surface containing characters can be considered flat, and a pinhole model can be adopted for a camera. Thus, in theory, the normalization should be projective. Usually, the camera optical axis is approximately perpendicular to the document surface, so the projective normalization can be replaced with an affine one without a significant loss of accuracy. An affine image transformation is performed significantly faster than a projective normalization, which is important for OCR on mobile devices. In this work, we propose a fast approach for image normalization. It utilizes an affine normalization instead of a projective one if there is no significant loss of accuracy. The approach is based on a proposed criterion for the normalization accuracy: root mean square (RMS) coordinate discrepancies over the region of interest (ROI). The problem of optimal affine normalization according to this criterion is considered. We have established that this unconstrained optimization is quadratic and can be reduced to a problem of fractional quadratic functions integration over the ROI. The latter was solved analytically in the case of OCR where the ROI consists of rectangles. The proposed approach is generalized for various cases when instead of the affine transform its special cases are used: scaling, translation, shearing, and their superposition, allowing the image normalization procedure to be further accelerated.
URI (Унифицированный идентификатор ресурса) : https://dx.doi.org/10.18287/2412-6179-CO-759
http://repo.ssau.ru/handle/Zhurnal-Komputernaya-optika/Optimal-affine-image-normalization-approach-for-optical-character-recognition-87756
Другие идентификаторы : Dspace\SGAU\20210228\87756
Располагается в коллекциях: Журнал "Компьютерная оптика"

Файлы этого ресурса:
Файл Описание Размер Формат  
450111.pdfОсновная статья1.06 MBAdobe PDFПросмотреть/Открыть



Все ресурсы в архиве электронных ресурсов защищены авторским правом, все права сохранены.