Advertisement

DeepITQA: Deep Based Image Text Quality Assessment

  • Hongyu Li
  • Fan Zhu
  • Junhua Qiu
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 11306)

Abstract

To predict the OCR accuracy of document images, text related image quality assessment is necessary and of great value, especially in online business processes. Such quality assessment is more interested in text and aims to compute the quality score of an image through predicting the degree of degradation at textual regions. In this paper, we propose a deep based framework to achieve image text quality assessment, which is composed of three stages: text detection, text quality prediction, and weighted pooling. Text detection is used to find potential text lines and the quality is solely estimated on detected text lines. To predict text line quality, we train a deep neural network model with our synthetic samples. The overall text quality of an image can be computed through pooling the quality of all detected text lines by way of weighted averaging. The proposed method has been tested on two benchmarks and our collected pictures. Experimental results show that the proposed method is feasible and promising in image text quality assessment.

Keywords

Image quality assessment Text detection Text quality Deep neural network 

References

  1. 1.
    Bong, D.B.L., Khoo, B.E.: Blind image blur assessment by using valid reblur range and histogram shape difference. Signal Process. Image Commun. 29(6), 699–710 (2014)CrossRefGoogle Scholar
  2. 2.
    Bong, D.B.L., Khoo, B.E.: Objective blur assessment based on contraction errors of local contrast maps. Multimedia Tools Appl. 74(17), 7355–7378 (2015). https://doi.org/10.1007/s11042-014-1983-5CrossRefGoogle Scholar
  3. 3.
    Bosse, S., Maniry, D., Wiegand, T., Samek, W.: A deep neural network for image quality assessment. In: 2016 IEEE International Conference on Image Processing (ICIP), September 2016, pp. 3773–3777 (2016)Google Scholar
  4. 4.
    Buta, M., Neumann, L., Matas, J.: Fastext: Efficient unconstrained scene text detector. In: IEEE International Conference on Computer Vision, pp. 1206–1214 (2015)Google Scholar
  5. 5.
    De, K., Masilamani, V.: Discrete orthogonal moments based framework for assessing blurriness of camera captured document images. In: Vijayakumar, V., Neelanarayanan, V. (eds.) Proceedings of the 3rd International Symposium on Big Data and Cloud Computing Challenges (ISBCC – 16’). SIST, vol. 49, pp. 227–236. Springer, Cham (2016).  https://doi.org/10.1007/978-3-319-30348-2_18CrossRefGoogle Scholar
  6. 6.
    He, K., Zhang, X., Ren, S., Sun, J.: Deep residual learning for image recognition. In: 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 770–778, June 2016Google Scholar
  7. 7.
    Hou, W., Gao, X., Tao, D., Li, X.: Blind image quality assessment via deep learning. IEEE Trans. Neural Netw. Learn. Syst. 26(6), 1275–1286 (2015)MathSciNetCrossRefGoogle Scholar
  8. 8.
    Kang, L., Ye, P., Li, Y., Doermann, D.: A deep learning approach to document image quality assessment. In: IEEE International Conference on Image Processing, pp. 2570–2574 (2014)Google Scholar
  9. 9.
    Kumar, J., Chen, F., Doermann, D.: Sharpness estimation for document and scene images. In: International Conference on Pattern Recognition, pp. 3292–3295 (2013)Google Scholar
  10. 10.
    Nayef, N.: Metric-based no-reference quality assessment of heterogeneous document images. In: SPIE Electronic Imaging, pp. 94020L–94020L-12 (2015)Google Scholar
  11. 11.
    Nayef, N., Luqman, M.M., Prum, S., Eskenazi, S., Chazalon, J., Ogier, J.M.: SmartDoc-QA: a dataset for quality assessment of smartphone captured document images - single and multiple distortions. In: International Conference on Document Analysis and Recognition, pp. 1231–1235 (2015)Google Scholar
  12. 12.
    Peng, X., Cao, H., Natarajan, P.: Document image quality assessment using discriminative sparse representation. In: Document Analysis Systems, pp. 227–232 (2016)Google Scholar
  13. 13.
    Rusinol, M., Chazalon, J., Ogier, J.M.: Combining focus measure operators to predict OCR accuracy in mobile-captured document images. In: IAPR International Workshop on Document Analysis Systems, pp. 181–185 (2014)Google Scholar
  14. 14.
    Tian, Z., Huang, W., He, T., He, P., Qiao, Y.: Detecting text in natural image with connectionist text proposal network. In: European Conference on Computer Vision, pp. 56–72 (2016)CrossRefGoogle Scholar
  15. 15.
    Ye, P., Doermann, D.: Document image quality assessment: a brief survey. In: International Conference on Document Analysis and Recognition, pp. 723–727 (2013)Google Scholar
  16. 16.
    Ye, Q., Doermann, D.: Text detection and recognition in imagery: a survey. IEEE Trans. Pattern Anal. Mach. Intell. 37(7), 1480–1500 (2015)CrossRefGoogle Scholar
  17. 17.
    Zhang, Z., Zhang, C., Shen, W., Yao, C., Liu, W., Bai, X.: Multi-oriented text detection with fully convolutional networks, pp. 4159–4167 (2016)Google Scholar
  18. 18.
    Zhou, X., et al.: EAST: an efficient and accurate scene text detector. In: 2017 IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017, Honolulu, HI, USA, 21–26 July 2017, pp. 2642–2651 (2017)Google Scholar
  19. 19.
    Zhu, Y., Yao, C., Bai, X.: Scene text detection and recognition: recent advances and future trends. Front. Comput. Sci. 10(1), 19–36 (2016)CrossRefGoogle Scholar

Copyright information

© Springer Nature Switzerland AG 2018

Authors and Affiliations

  1. 1.AI LabZhongAn Information Technology Service Co., Ltd.ShanghaiChina

Personalised recommendations