Image Preprocessing Techniques

Forfattere

  • Ravi Kant Yadav

Nøgleord:

Preprocessing, offset, character recognition

Resumé

 Preprocessing method is an important step, specifically for low-quality picture images. This method is robust to lighting variations and produces images with very little noise and consistent stroke width. The process computes a “surface” of background intensities at every point in the image that performs adaptive thresholding based on this result. The surface is estimated by identifying regions of low resolution text and interpolating neighboring background intensities into these regions. The final threshold is a combination of this surface and a global offset. Most image-processing techniques involve treating the image as a two-dimensional signal and applying standard signal-processing techniques to it. The term applies both to mental processes used by humans when reading text, and to artificial processes implemented in computers, which are the subject of optical character recognition. Keywords: Preprocessing, offset, character recognition 

Forfatterbiografi

  • Ravi Kant Yadav
    Dept. of Computer Science, School of Management Sciences, Varanasi, India

Publiceret

2012-08-09

Nummer

Sektion

Research Articles