| 摘要 |
<p>A method and system for recognizing text in computer images comprising distorted text provides an adaptive iterative process wherein recognition rules are adapted, added or omitted based on the present state of the recognition process. When the first pass through the recognition and adaptation is completed, the remaining unrecognized words (15) are passed through the recognition system 1 using the modified set of recognition rules stored in (18) and the process is repeated. In most cases the recognition system 1 will identify further reliable recognized words, which iteratively can be used to improve the recognition rules until the true text comprised in image (10) is recognized throughout the whole text. The steps of the method according to the present invention are thus repeated until convergence.</p> |
