本文提出一种基于小波压缩域的图像/视频字符自动检测定位方法。对于基于小波变换的压缩格式图像/视频,仅部分解压即可实现具有复杂背景的字符区域定位。与其他定位方法相比,明显降低计算复杂度,对字符尺寸的鲁棒性,检测精度显著提高。初步实验结果验证了该方法的有效性和快速性。
The paper describes a novel automatic character location scheme which operates directly on wavelet compressed domain. For compressed format images/videos based on wavelet transform, the character regions can be detected and located automatically from complex background without full-decompressing. Compared with other traditional character location methods, the proposed scheme has the advantages of low computational cost, robust to size and font of characters and high accuracy. Preliminary experiment results show that the proposed scheme is efficient and effective.