Degraded gray-scale document recognition using pseudo two- dimensional hidden Markov models and N-best hypotheses

摘要

The present invention provides a method for recognizing connected and degraded text embedded in a gray-scale image. In accordance with the invention, pseudo two-dimensional hidden Markov models (PHMMs) are used to represent characters. Observation vectors for the gray-scale image are produced from pixel maps obtained by gray-scale optical scanning. Three components are employed to characterize a pixel: a convoluted, quantized gray-level component, a pixel relative position component, and a pixel major stroke direction component. These components are organized as an observation vector, which is continuous in nature, invariant in different font sizes, and flexible for use in various quantization processes. In this matter, information loss or distortion due to binarization processes is eliminated; moreover, in cases where documents are binary in nature (e. g., faxed documents), the bi-level image may be compressed by subsampling into multi(gray)-level without losing information, thereby enabling recognition of the compressed images in a much shorter time. Furthermore, documents in gray-level may be scanned and processed with much lower resolution than in binary without sacrificing the performance. This can also significantly increase the processing speed.