A Comparison in Colored Text Enhancement
Fatin E. M. Al-Obaidi, Ali Jassim Mohamed Ali · 2013
Text data present in images contain useful information for automatic explanation, indexing, and structuring for images. Extracting information involves detection, localization, tracking, extraction, enhancement, and recognition of the text from a given image. However variations of text due to differences in color as an example make the problem of automatic text enhancement extremely challenging in the computer vision research area especially when we deal with image's color component. A comparison between seven different filters has been executed upon colored text image by processing each image's color component individually. MSE, SNR, and PSNR for each filter's application have been analyzed to determine the success and limitations of each approach. Among the seven different filters that have been adopted and in spite of its changeable effect in luminance, one can use the New-Equalization filter that has been suggested in this research as a good enhanced filter for such type of text.