A Novel Method for Character Segmentation in Natural Scenes

• 作 者: XU Li QI Fei-hu JIANG Ren-jie
• 机 构: Dept. of Computer Science and Eng.
• 出 处: 上海交通大学学报（英文版）
• 发表日期: 2004年期
• 关键词: character segmentation feature extraction model-based hierarchical merging (MHM)
• 基金项目: OMRON Corporation

A new approach to extract and segment characters in natural scenes was proposed in this paper. First, a set of intrinsic features were calculated based on connected components (CCs) extracted by a non-linear Nilblack algorithm. Then, feature propagation was conducted for feature enhancement, under the constraint of the layout relations. Next, candidate CCs were fed into classifiers with the enhanced feature vector. At last, a model-based hierarchical merging (MHM) procedure was presented to obtain understandable characters. The proposed merging algorithm utilized the constraint of text lines for specific languages and dynamically merges CCs into characters.The whole algorithm was evaluated at both pixel level and character level, experimental results showed that the proposed method is effective in detecting scene characters with significant geometric variations, uneven illumination, extremely low contrast and cluttered background.