Document image orientation based on both text and image

This paper investigated the problem of orientation detection for document images with Chinese characters. These images may be in four orientations: right side up, up-side down, 90° and 270° rotated counterclockwise. First, we presented the structure of text-recognition-based orientation detection algorithm. Text line verification and orientation judgment methods were mainly discussed, afterwards multiple experiments were carried. Distance-difference based text line verification and confidence based text line verification were proposed and compared with methods without text line verification. Then, a picture-based orientation detection framework was adopted for the situation where no text line was detected. This high-level classification problem was solved by relatively low-level vision features including Color Moments (CM) and Edge Direction Histogram (EDH), with distant-based classification scheme. Finally, confidencebased classifier combination strategy was employed in order to make full use of the complementarity between different features and classifiers. Experiments showed that both text line verification methods were able to improve the accuracy of orientation detection, and picture-based orientation detection had a good performance for no-text image set.

[1]  Robert S. Caprari.  Algorithm for text page up/down orientation determination , 2000, Pattern Recognit. Lett..

[2]  Shijian Lu,et al.  Automatic Detection of Document Script and Orientation , 2007 .

[3]  Anil K. Jain,et al.  Automatic image orientation detection , 2002, IEEE Trans. Image Process..

[4]  Rui Zhang,et al.  Adaptive confidence transform based classifier combination for Chinese character recognition , 1998, Pattern Recognit. Lett..

[5]  Changsong Liu,et al.  Automatic picture orientation detection based on classifier combination , 2011, Electronic Imaging.

[6]  Chew Lim Tan,et al.  Fast and Accurate Detection of Document Skew and Orientation , 2007, Ninth International Conference on Document Analysis and Recognition (ICDAR 2007).

[7]  Dan S. Bloomberg,et al.  Measuring document image skew and orientation , 1995, Electronic Imaging.

[8]  Hrishikesh B. Aradhye.  A generic method for determining up/down orientation of text in roman and non-roman scripts , 2005, Pattern Recognit..

[9]  Youbin Chen,et al.  Evaluation and Application of Recognition Confidence in OCR , 1998, ACCV.

[10]  Norihiro Hagita,et al.  Automated entry system for printed documents , 1990, Pattern Recognit..

[11]  Costas S. Xydeas,et al.  Detecting the skew angle in document images , 1994, Signal Process. Image Commun..

[12]  Harry Wechsler,et al.  Automated page orientation and skew angle detection for binary document images , 1994, Pattern Recognit..

[13]  Ming Chen,et al.  Analysis, understanding and representation of Chinese newspaper with complex layout , 2000, Proceedings 2000 International Conference on Image Processing (Cat. No.00CH37101).