一种基于边缘检测和线条特征的视频字符检测算法

叠加在视频帧中的字符串为多媒体数据的高层语义分析提供了重要信息．该文提出了一种基于边缘的视频字符检测算法．利用字符区域内边缘的分布和强度的特点，对Canny边缘检测算子进行改进，使其根据图像内容自动确定分割阈值，确保获得高质量的字符边缘；然后，提取字符线条的关键特征，过滤非字符线条并用神经网络分类器判断最终的字符区域．实验结果表明该文算法在保持高速度和高查全率的同时，有效地降低了虚警，克服了基于边缘的字符检测算法的普遍缺点．

A Video Text Detecting Method Based on Edge Detection and Line Features

Text superimposed on video frames provides synoptic or supplemental information on video semantics. This paper proposes a novel method to detect superimposed t ext effectively. First, edges are detected by an improved Canny edge detector. Then, a line-feature vector graph is generated based on the edge map and the st roke information is extracted. Finally text regions are generated and filtered a ccording to line features. Experimental results show that, without much increasi ng the computational cost, the proposed method could suppress the false alarms notably, which outperforms other existing edge-based text detection methods.