Automatic Character Recognition in Complex Images

AbstractCharacter recognition system has received considerable attention in recent years due to the tremendous need for digitization of printed documents. The textual representations of images convey information relating to what is actually depicted in the image as well as what the image is about. Manual assignment of text data from images is time consuming and costly. Hence automation of text extraction from images is a challenging area in image processing due to its potential applications. Since text contained in images might be of different fonts, different languages, having surface deformation, poor quality, or even blurred, the proper text extraction becomes very difficult. If all these difficulties are overcome, the text extraction can be beneficial for different applications. Some of these applications include character recognitions in surveillance, object identification, banking sector, market survey and all.. The recognition rate drops drastically in some methods, when the font style or size of the character changes. This survey addresses a comparative study of different techniques available for extraction of characters from complex image