| 引用本文: | 张鹏伟,张伟伟.用卷积神经网络分类最大稳定极值区域实现汉字区域定位.[J].国防科技大学学报,2017,39(3):91-96.[点击复制] |
| ZHANG Pengwei,ZHANG Weiwei.Scene Chinese text localization by convolutional neural network classifying maximum stable extremal regions[J].Journal of National University of Defense Technology,2017,39(3):91-96[点击复制] |
|
| |
|
|
| 本文已被：浏览 8043次 下载 7551次 |
| 用卷积神经网络分类最大稳定极值区域实现汉字区域定位 |
| 张鹏伟, 张伟伟 |
|
(信息工程大学 密码工程学院， 河南 郑州 450001)
|
| 摘要: |
| 获取对应笔画级连通区的最大稳定极值区域，实施形态学闭操作融合相距较近的最大稳定极值区域，融合后最大稳定极值区域对应的单个汉字区域；利用灰度共生矩阵描述最大稳定极值矩形区域的纹理信息，将其作为卷积神经网络的输入，卷积神经网络对最大稳定极值区域进行分类，过滤非汉字部分；利用最大稳定极值区域颜色直方图的Bhattacharyya距离等特征对最大稳定极值区域进行聚类，同一类最大稳定极值区域组合得到汉字文本候选区域；再次利用卷积神经网络对候选文本区域进行分类，过滤非文本部分，剩余的就是定位到的汉字文本区域。实验结果表明，该算法对于汉字区域定位具有良好的效果。 |
| 关键词: 汉字区域定位 最大稳定极值区域 卷积神经网络 深度学习 灰度共生矩阵 |
| DOI：10.11887/j.cn.201703015 |
| 投稿日期：2016-01-07 |
| 基金项目:国家863计划资助项目(20157011012) |
|
| Scene Chinese text localization by convolutional neural network classifying maximum stable extremal regions |
| ZHANG Pengwei, ZHANG Weiwei |
|
(School of Cryptography Engineering, Information Engineering University, Zhengzhou 450001, China)
|
| Abstract: |
| Firstly, the MSERs (maximum stable extremal regions) which corresponded to Chinese strokes was extracted. The morphological close operation was used to connect the nearby MSERs. The fused MSER corresponded to Chinese characters. Gray level co-occurrence matric was used to describe the textural characteristics of the fused MSER rectangle. They were the input of CNN (convolutional neural network). The MSER rectangles were classified by CNN in order to filter none Chinese character rectangle. Then, Chinese text candidates were constructed by clustering MSER rectangles based on the features such as the color histogram Bhattacharyya distance of MSER rectangles. CNN was reused to classify Chinese text candidates to filter none Chinese text clusters. Finally, the rectangle of the remaining clusters was the Chinese text regions of natural scene image. Experiment shows that the proposed algorithm is desirable in localizing the Chinese text in natural scene images. |
| Keywords: Chinese text localization maximum stable extremal region convolutional neural network deep learning gray level co-occurrence matric |
|
|
