Enhancing quality of degraded images is important preliminary step for automatic extraction of the Sundanese digitized ancient manuscript known as lontar. The degraded lontar images have circular, inhomogeneity illumination characteristics which are represented by more brightness on inner center of images and darker in the outer ellipse-like area of the images. The purpose of enhanced image processing is to distribute the pixel's intensity so the images can have similar characteristics. The proposed method used in this paper is primarily based on background subtraction technique which is implemented in three color spaces. Background images are estimated using morphological mathematics using specific type and size of structuring element. This estimation process has to be performed due to unavailability of exact background image captured during image acquisition step. Theoretical foundations of the proposed method along with experimental results are reported in this paper. The proposed method has been tested on 12 Sundanese lontar images which denotes this approach different and unique from most of the existing methods in this field of study. Conducted experiment demonstrated the successful results of the method which is represented by displaying profile characteristics of each images on row 128.
[1]
Eric Dubois.
The Structure and Properties of Color Spaces and the Representation of Color Images
,
2009,
The Structure and Properties of Color Spaces and the Representation of Color Images.
[2]
Asep K. Supriatna,et al.
Sundanese ancient manuscripts search engine using probability approach
,
2017
.
[3]
Angelika Garz,et al.
Multi-scale texture-based text recognition in ancient manuscripts
,
2010,
2010 16th International Conference on Virtual Systems and Multimedia.
[4]
Claes Lundström,et al.
Technical report: Measuring digital image quality
,
2006
.
[5]
Sitti Rachmawati Yahya,et al.
Review on Image Enhancement Methods of Old Manuscript with Damaged Background
,
2010
.
[6]
Iping Supriana Suwardi,et al.
DEWA: A Multiaspect Approach for Multiple Face Detection in Complex Scene Digital Image
,
2007
.
[7]
Alejandro Héctor Toselli,et al.
Handwritten Text Recognition for Ancient Documents
,
2010,
WAPA.