عنوان انگلیسی مقاله:
Precise detection of Chinese characters in historical documents with deep reinforcement learning
ترجمه فارسی عنوان مقاله:
تشخیص دقیق حروف چینی در اسناد تاریخی با یادگیری تقویت عمیق
Sciencedirect - Elsevier - Pattern Recognition, 107 (2020) 107503. doi:10.1016/j.patcog.2020.107503
Wu Sihang 1 , Wang Jiapeng 1 , Ma Weihong, Jin Lianwen
The decision-making ability of deep reinforcement learning has been proved successfully in a variety of fields. Here, we use this method for precise character detection by making tight bounding boxes around the Chinese characters in historical documents. An agent is trained to learn the control policy of fine-tuning a bounding box step-by-step through a Markov Decision Process. We introduce a novel f ully c onvolutional n etwork with p osition-sensitive Region-of-Interest (RoI) pooling (FCPN). The network receives character patches as input without fixed size, and it can fuse position information into the fea- tures of actions. Besides, we propose a d ense r eward f unction (DRF) that provides excellent rewards ac- cording to different actions and environment states, improving the decision-making ability of the agent. Our approach is designed as a universal method that can be applied to the output of all character-level or word-level text detectors to obtain more precise detection results. Application to the Tripitaka Kore- ana in Han (TKH) and Multiple Tripitaka in Han (MTH) datasets confirm the very promising performance of this method. In particular, our approach yields a significant improvement under a large Intersection over Union (IoU) of 0.8. The robustness and generality are also proved by experiments on the scene text datasets ICDAR2013 and ICDAR2015.
Keywords: Deep reinforcement learning | Historical documents | Character detection | Reward function