Book ChapterDOI
Epigraphic Document Image Enhancement Using Retinex Method
H. T. Chandrakala,G. Thippeswamy +1 more
- pp 178-184
Reads0
Chats0
TLDR
A new approach for enhancement of Epigraphic Document images using Retinex method is presented in this paper, which enhances the visual clarity of the degraded images by highlighting the foreground text and suppressing the background noise.Abstract:
Epigraphic Documents are the ancient handwritten text documents inscribed on stone, metals, wood and shell. They are the most authentic, solitary and unique documented evidences available for the study of ancient history. In the recent years, Archeological Departments worldwide have taken up the massive initiative of converting their repository of ancient Epigraphic Documents into digital libraries for the perennial purpose of their preservation and easy dissemination. The visual quality of the digitized Epigraphic Document images is poor as they are captured from sources that would have suffered from various kinds of degradations like aging, depositions and risky handling. Enhancement of these images is an essential prerequisite to make them suitable for automatic character recognition and machine translation. A new approach for enhancement of Epigraphic Document images using Retinex method is presented in this paper. This method enhances the visual clarity of the degraded images by highlighting the foreground text and suppressing the background noise. The method has been tested on digitized estampages of ancient stone inscriptions of 11th century written in old Kannada language. The results achieved are efficient in terms of root mean square contrast and standard deviation.read more
Citations
More filters
Journal ArticleDOI
Adaptive shock-diffusion model for restoration of degraded document images
Jiebin Guo,Chuanjiang He +1 more
TL;DR: Numerical experiments show that the proposed adaptive enhancement model is very effective for restoration of degraded document images with blur, noise and bleed-through, and has averagely the best performance on the DIBCO (Document Image Binarization Competition) series datasets, compared to five PDE (partial differential equation)-based models.
Book ChapterDOI
Deep Convolutional Neural Networks for Recognition of Historical Handwritten Kannada Characters
H. T. Chandrakala,G. Thippeswamy +1 more
TL;DR: A deep convolutional neural networks approach for character recognition of handwritten historical Kannada manuscripts is presented, a model that unifies feature extraction and classification and promising results are observed.
Journal ArticleDOI
Impact of Total Variation Regularization on Character Segmentation from Historical Stone Inscriptions
TL;DR: In this article, an automatic segmentation scheme for accurate segmentation of characters from Historical Handwritten Kannada Stone Inscription images is presented in which a framework of TVR enhancement and Connected Component Labeling segmentation is implemented and evaluated on the dataset of digitized Estampages of historical Handwritten Karnada stone Inscriptions (EHHKSI).
Book ChapterDOI
Edge Enhancement Method for Detection of Text From Handwritten Documents
H. T. Chandrakala,G. Thippeswamy +1 more
TL;DR: Sobel edge detection method, which efficiently enhances the image contrast and detects the character edges, is proposed and experimentation of this approach yielded high F-measure and global contrast factor values.
References
More filters
Journal ArticleDOI
Lightness and Retinex Theory
Edwin H Land,John J. McCann +1 more
TL;DR: The mathematics of a lightness scheme that generates lightness numbers, the biologic correlate of reflectance, independent of the flux from objects is described.
Journal ArticleDOI
Recent Advances in Retinex Theory
TL;DR: It is shown that the paradox of colour constancy does not really exist because it is not true that the colour of a point on an object is determined by the composition of the light coming from the object.
Proceedings Article
Retinex in Matlab.
TL;DR: This work provides concise MATLABTM implementations of two of the spatial techniques of making pixel comparisons using Retinex methods of lightness computation, along with test results on several images and a discussion of the results.
Proceedings ArticleDOI
Contrast Enhancement by Multi-scale Adaptive Histogram Equalization
TL;DR: In this paper, an approach for contrast enhancement utilizing multi-scale analysis is introduced, where the sub-band coefficients were modified by the method of adaptive histogram equalization to achieve optimal contrast enhancement, the sizes of subregions were chosen with consideration to the support of the analysis filters.