CRF based text detection for natural scene images using convolutional | |
Wang YN(王燕娜)1,2; Shi,Cunzhao1; Baihua Xiao1; Chunheng Wang1; Chengzuo Qi1,2 | |
发表期刊 | Neurocomputing |
2018 | |
期号 | 295页码:46-58 |
摘要 | This paper presents a novel scene text detection method based on conditional random field (CRF) framework. We estimate the confidence of Maximally Stable Extremal Region (MSER) being text by leveraging convolutional neural network (CNN) to define the unary cost item. In addition, we establish the neighboring interactions for MSERs using four different features including color, shape, stroke and spatial features to define the pairwise cost item. Considering the special layout of texts appearing in natural scene images, we employ context information to recover missing text MSER candidates. Furthermore, text MSERs are grouped into candidate text lines which are verified with shape-specific classifiers by integrating gray and binary features. Experimental results on four public benchmark datasets show that the proposed method achieves the comparable performance. |
关键词 | Scene Text Detection Mser Cnn Crf Context Information Shape-specific Classifiers |
文献类型 | 期刊论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/21036 |
专题 | 复杂系统管理与控制国家重点实验室_影像分析与机器视觉 |
作者单位 | 1.The State Key Laboratory of Management and Control for Complex Systems,Institute of Automation, Chinese Academy of Sciences 2.University of Chinese Academy of Sciences, Beijing, China |
第一作者单位 | 中国科学院自动化研究所 |
推荐引用方式 GB/T 7714 | Wang YN,Shi,Cunzhao,Baihua Xiao,et al. CRF based text detection for natural scene images using convolutional[J]. Neurocomputing,2018(295):46-58. |
APA | Wang YN,Shi,Cunzhao,Baihua Xiao,Chunheng Wang,&Chengzuo Qi.(2018).CRF based text detection for natural scene images using convolutional.Neurocomputing(295),46-58. |
MLA | Wang YN,et al."CRF based text detection for natural scene images using convolutional".Neurocomputing .295(2018):46-58. |
条目包含的文件 | ||||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
CRF based Text Detec(3736KB) | 期刊论文 | 作者接受稿 | 开放获取 | CC BY-NC-SA | 浏览 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论