Texture-based approach for text detection in images using support vector machines and continuously adaptive mean shift algorithm
Texture-based approach for text detection in images using support vector machines and continuously adaptive mean shift algorithm
复制标题
DOI：
10.1109/tpami.2003.1251157
复制
发表时间：
2003-12-01
影响因子：
23.6
通讯作者：
Kim, JH
中科院分区：
文献类型：
--
作者：
Kim, KI;Jung, K;Kim, JH
提出了一种新的基于纹理的图像文本检测方法。利用支持向量机对文本的文本特征进行分析。不使用外部纹理特征提取模块；相反，组成纹理模式的原始像素的强度被直接馈送到支持向量机，即使在高维空间也能很好地工作。接下来，通过将连续自适应Mean Shift算法(CAMSHIFT)应用于纹理分析结果来识别文本区域。CAMSHIFT和支持向量机的组合产生了稳健和高效的文本检测，因为对相关性较低的像素进行耗时的纹理分析受到限制，只留下一小部分输入图像进行纹理分析。
The current paper presents a novel texture-based method for detecting texts in images. A support vector machine (SVM) is used to analyze the textural properties of texts. No external texture feature extraction module is used; rather, the intensities of the raw pixels that make up the textural pattern are fed directly to the SVM, which works well even in high-dimensional spaces. Next, text regions are identified by applying a continuously adaptive mean shift algorithm (CAMSHIFT) to the results of the texture analysis. The combination of CAMSHIFT and SVMs produces both robust and efficient text detection, as time-consuming texture analyses for less relevant pixels are restricted, leaving only a small part of the input image to be texture-analyzed.