一种新的彩色图象文字提取算法

Automatic Text Extraction from Color Image

文字信息在描述图象内容时起着重要的作用,因此文字提取及识别是基于内容视频检索的关键技术。提出了一个从彩色图象背景中提取文字的快速而有效的算法。由于文本字符串的对比度较高,首先用一个改进的sobel算子将彩色图象变换为二值的边缘图象,再对该边缘图象进行涂抹处理,然后基于候选文本区的特征从不同复杂度的彩色图象中提取文本信息,最后将提取出的文本输入到文字识别(OCR)引擎,识别结果证明了此方法的有效性。

Text extraction is a key technique in content-based video retrieval as textual information plays an important role in describing the content of image.An effective and fast method to extract text from color background has been proposed in this paper.Under the assumption that the text strings with high contrast are usually important,a modified sobel operator is utilized to transform the color image into an edge binary image,and an edge-based smear processing is used to speed up the color image process.Combining the feature-based text candidates identification with the moment preserving region segmentation,we can effectively extract text from scanned true-color image varying in text font,size and color complexity.The extracted texts are finally inputted to the OCR engine.Experimental results demonstrate the effectivity and feasibility of this method.