Graphics Recognition Methods and Applications

Volume 1072 of the series Lecture Notes in Computer Science pp 35-48


Detection of horizontal lines in noisy run length encoded images: The FAST method

  • Atul K. ChhabraAffiliated withNYNEX Science & Technology
  • , Vishal MisraAffiliated withNYNEX Science & TechnologyDepartment of Electrical and Computer Engineering, University of Massachusetts at Amherst
  • , Juan AriasAffiliated withNYNEX Science & Technology

* Final gross prices may vary according to local VAT.

Get Access


We present a fast method for finding horizontal lines in run length encoded images. The method was motivated by the need for quick and reliable detection of horizontal lines in an interactive drawing conversion system for telephone company drawings. At the core of the algorithm are the processes of filtering run lengths, assembling filtered run lengths, generating top silhouette, and thresholding the gradient of the top silhouette to extract one horizontal line at a time. The method is robust in the presence of distortion; it can tolerate significant skew and warping, both local and global, and can bridge significant breaks in lines without too many false positive lines.