Journal of Applied Sciences

• Articles • Previous Articles     Next Articles

Improving Line-Based Light Balancing for Text Images

Chin-chen CHANG 1,2, Chin-hao CHEN 2, Pei-yu LIN 1   

  1. 1. Department of Computer Science and Information Engineering, National Chung Cheng University; 
    2. Department of Information Engineering and Computer Science, Feng Chia University

  • Received:2008-05-09 Revised:2008-07-10 Online:2008-09-27 Published:2008-09-27
  • Contact: Chin-chen CHANG

Abstract:

With digital cameras, an inappropriate light source often leads to uneven illumination of the picture. The same problem also occurs when capturing an image via digital scanners; irregular illumination seriously distorts the image quality. Light balancing can eliminate such uneven illumination in multimedia. In this paper, we provide an improved version of this technique for text images. The new version adopts a contrast weight to amplify distinction between the text and the background. This can equalize the histogram and reduce uneven contrast of the disproportionate light distribution. The simulator shows that the new version can effectively improve the distorted documents with satisfactory quality and readability.

Key words: light balancing, illumination, digitization, contrast enhancement

CLC Number: