Abstract
Scanned documents commonly suffer from skew and redundant edges when a paper document is scanned and saved as an image file. In this paper, we propose an improved algorithm that can automatically deskew and crop scanned documents. To improve the accuracy of edge detection, the image edge can be made salient based on different edge types. The estimation of the deskew and the cropping value can benefit from the salient image edge. This paper also adopts the improved region growing method to automatically obtain the cropping value to crop the scanned image. The proposed method mainly includes image preprocessing, image classification, skew angle estimation, deskewing and cropping; estimation of the cropping values is based on different image types. Compared with the previous algorithms, the proposed algorithm not only has good anti-interference ability, but can also accurately estimate the cropping value and skew angle. Since the scanned images in the database from DISEC’2013 do not have redundant edges, another experiment with redundant edges must be performed with our database. The experimental results illustrate that the proposed method performs better than other methods.
Similar content being viewed by others
References
Postl, W.: Detection of linear oblique structures and skew scan in digitized documents. In: Conference on Pattern Recognition, pp. 687–689 (1986)
Alireza, A., et al.: A painting based technique for skew estimation of scanned documents. In: Conference on Document Analysis and Recognition, pp. 299–303 (2011)
Aradhya, V.N.M., Kumar, G.H.: An accurate and efficient skew estimation technique for South Indian documents: a new boundary growing and nearest neighbor clustering based approach. Int. J. Robot. Autom. 22(4), 272–280 (2007)
Fabrizio, J.: A Precise Skew Estimation Algorithm for Document Images Using KNN Clustering and Fourier Transform. ICIP, Paris (2014)
Stahlberg, F., Vogel, S.: Document Skew Detection Based on Hough Space Derivations. ICDAR, Nancy (2015)
Singh, C., Bhatia, N., Kaur, A.: Hough transform based fast skew detection and accurate skew correction methods. Pattern Recogn. 41(12), 3528–3546 (2008)
Yildirim, B.: Projection profile analysis for skew angle estimation of woven fabric images. J. Text. Inst. 105(6), 654–660 (2015)
Papandreou, A., et al.: Efficient skew detection of printed document images based on novel combination of enhanced profiles. Int. J. Doc. Anal. Recogn. 17(4), 433–454 (2014)
Li, S., Shen, Q., Sun, J.: Skew detection using wavelet decomposition and projection profile analysis. Pattern Recognit. Lett. 28(5), 555–562 (2007)
Papandreou, A., Gatos, B.: A novel skew detection technique based on vertical projections. In: International Conference on Document Analysis and Recognition, pp. 384–388 (2011)
Yi, R; Wu, MH. “Digital Compensation for Timing Mismatches in Interleaved ADCs” ATS, Yilan, Taiwan (2013)
Zhao, D., DuA, F.: Novel approach for scale and rotation adaptive estimation based on time series alignment. Vis. Comput. 35, 175–189 (2020)
Srihari, N., Govindaraju, V.: Analysis of textual images using the Hough transform. Mach. Vis. Appl. 2(3), 141–153 (1989)
Kleber, F., Diem, M.: Robust skew estimation of handwritten and printed documents based on gray value images. In: ICPR, Sweden (2014)
Zhang, F., Zhang, Y.F.: Scanned document images skew correction based on shearlet transform. In: MIWAI, Artificial Intelligence, vol. 9426, pp. 226–232. Fuzhou Univ, Fuzhou (2015)
Brodic, D., Milivojevic, Z.N.: Log-polar transformation as a tool for text skew estimation. Elektron. IR Elektrotech. 19(2), 61–64 (2013)
Saba, T., Sulong, G.: Document image analysis: issues, comparison of methods and remaining problems (Retracted article. See vol. 42, pg. 1067, 2014). Artif. Intell. Rev. 35(2), 101–118 (2011)
Ouwayed, N., Belaid, A., Auger, F.: Skew angle estimation of scanned handwritten Arabic documents using a time-frequency analysis of the projection histograms. Traitement DU Signal 26(4), 307–319 (2009)
Shivakumara, P., Hemantha, K.G., Manjunath, A.V.N.: Character skew estimation: a new and simple edge based model. In: International Conference on Advanced Computing and Communications, Mangalore (2006)
Brodic, D., Maluckov, C.A., Peng, L.: Estimation of the text skew in the old printed documents. Int. J. Comput. Commun. Control 8, 673–680 (2013)
Papandreou, A., Gatos, B., Louloudis, G., Stamatopoulos, N.: DISEC 2013—document image skew estimation contest. In: International Conference on Document Analysis and Recognition, pp. 1476–1480 (2013)
Epshtein, B.: Determining document skew using inter-line spaces. In: Proceedings of the International Conference on Document Analysis and Recognition, pp. 27–31 (2011)
Diem, M., Kleber, F., Sablatnig, R.: Skew estimation of sparsely inscribed document fragments. In: Proceedings of the 10th IAPR International Workshop on Document Analysis Systems, pp. 292–296 (2012)
Jiang, X., Bunke, H., Widmer-Kljajo, D.: Skew detection of document images by focused nearest-neighbor clustering. In: Proceedings of the 5th International Conference on Document Analysis and Recognition, pp. 629–632 (1999)
Dobai, L., Teletin, M.: A document detection technique using convolutional neural networks for optical character recognition systems. In: 27th European symposium on artificial neural networks, pp. 547–552 (2019)
Wenguan, W., Jianbing, S., Haibin, L.: A deep network solution for attention and aesthetics aware photo cropping. IEEE Trans. Pattern Anal. Mach. Intell. 41, 1531–1544 (2019)
Zhao, J.X., Cao, Y., Fan, D.P.: Contrast prior and fluid pyramid integration for RGBD salient object detection. In: CVPR (2019)
Fu, K., Zhao, Q., et al.: Refinet: a deep segmentation assisted refinement network for salient object detection. IEEE Trans. Multimed. 21, 457–469 (2019)
Keren, F., Zhao, Q., et al.: Deepside: a general deep framework for salient object detection. Neurocomputing 356, 69–82 (2019)
Fan, D.P., Cheng, M.M., et al.: Salient objects in clutter: bringing salient object detection to the foreground. In: ECCV (2018)
Zhao, L., Zhao, Q., Liu, H., Lv, P., Gu, D.: Structural sparse representation-based semi-supervised learning and edge detection proposal for visual tracking. Vis. Comput. 33, 1169–1184 (2017)
Bissacco, A., et al.: PhotoOCR: reading text in uncontrolled conditions. In: IEEE International Conference on Computer Vision (ICCV), pp. 785–792 (2013)
Jiao, X., Wu, T.: A visual consistent adaptive image thresholding method. Imaging Sci. J. 64(1), 34–39 (2016)
Diem, M., Hollaus, F., Sablatnig, R.: MSIO: MultiSpectral Document Image Binarization. DAS, Greece (2016)
Han, X.W., Gao, Y., Cao, Y., Lu, Z., Niu, D.: Video moving target binary image processing method based on OTSU. AER Adv. Eng. Res. 1–4, 12 (2015)
Gao, Y.F., Zhang, H.T., Ji, J.: Image segmentation based on maximum relationship principle of conditional distribution under the assumption of Poisson distribution-art. no.66250 K. In: International Conference on Photoelectronic Detection and Imaging, vol. 6625, pp. 6250–6250 (2008)
Himeur, Y., Boukabou, A.: Robust image transmission over powerline channel with impulse noise. Multimed. Tools Appl. 76(2), 2813–2835 (2017)
Fan, K.C., Wang, Y.K., Lay, T.R.: Marginal noise removal of document images. Pattern Recogn. 35(11), 2593–2611 (2002)
Ding, J.H., Lin, Z.J., Yu, L.Y.: A correction algorithm for document images based on edge contour. In: ITMS, Tianjin, pp. 105–108 (2015)
Chen, Y.K., Wang, J.F.: Locating the destination address block on images of complex mail pieces. J. Chin. Inst. Eng. 24(6), 761–770 (2001)
Zhang, Z., Zhang, C., Shen, W., Yao, C., Liu, W., Bai, X.: Multioriented text detection with fully convolutional networks. In: Proceedings of the IEEE Conference on Computer Vision Pattern Recognition (2016)
Yao, C., Bai, X., Sang, N., Zhou, X., Zhou, S., Cao, Z.: Scene text detection via holistic, multi-channel prediction. arXiv preprint arXiv:1606.09002 (2016)
Ma, J., Shao, W., Ye, H., Wang, L., et al.: Arbitrary-oriented scene text detection via rotation proposals. arXiv preprint arXiv:1703.01086 (2018)
Acknowledgements
This work was supported by the National Natural Science Foundation of China (Nos. 61673129, 51674109).
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Cai, C., Meng, H. & Qiao, R. Adaptive cropping and deskewing of scanned documents based on high accuracy estimation of skew angle and cropping value. Vis Comput 37, 1917–1930 (2021). https://doi.org/10.1007/s00371-020-01952-z
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00371-020-01952-z