Journal of the Korea Convergence Society (한국융합학회논문지)

Korea Convergence Society (한국융합학회)
권호 표지
DOI QR코드

DOI QR Code

A Study on Extraction of text region using shape analysis of text in natural scene image

자연영상에서 문자의 형태 분석을 이용한 문자영역 추출에 관한 연구

  • Yang, Jae-Ho (Dept of Plasma Bio Display, KwangWoon University) ;
  • Han, Hyun-Ho (Dept of Plasma Bio Display, KwangWoon University) ;
  • Kim, Ki-Bong (Department of computer information, Daejeon health institute of technology) ;
  • Lee, Sang-Hun (Ingenium college of liberal arts, Kwangwoon university)
  • 양재호 (광운대학교 플라즈마 바이오 디스플레이학과) ;
  • 한현호 (광운대학교 플라즈마 바이오 디스플레이학과) ;
  • 김기봉 (대전보건대학 컴퓨터정보학과) ;
  • 이상훈 (광운대학교 인제니움학부대학)
  • Received : 2017.11.13
  • Accepted : 2018.01.20
  • Published : 2018.11.28
Download PDF
⟨ Previous Next ⟩

Abstract

In this paper, we propose a method of character detection by analyzing image enhancement and character type to detect characters in natural images that can be acquired in everyday life. The proposed method emphasizes the boundaries of the object part using the unsharp mask in order to improve the detection rate of the area to be recognized as a character in a natural image. By using the boundary of the enhanced object, the character candidate region of the image is detected using Maximal Stable Extermal Regions (MSER). In order to detect the region to be judged as a real character in the detected character candidate region, the shape of each region is analyzed and the non-character region other than the region having the character characteristic is removed to increase the detection rate of the actual character region. In order to compare the objective test of this paper, we compare the detection rate and the accuracy of the character region with the existing methods. Experimental results show that the proposed method improves the detection rate and accuracy of the character region over the existing character detection method.

본 논문에서는 일상에서 획득할 수 있는 자연 영상에서 문자를 검출하기 위해 영상 개선 및 문자의 형태를 분석하여 문자를 검출하는 방법을 제안한다. 제안하는 방법은 자연 영상에서 문자로 인식될 영역의 검출률을 향상시키기 위해 객체부분의 경계를 언샤프 마스크를 사용하여 강조하였다. 향상된 객체의 경계 부분을 이용하여 영상의 문자 후보영역을 MSER(Maximally Stable Extermal Regions)을 이용하여 검출하였다. 검출된 문자 후보영역에서 실제 문자로 판단될 영역을 검출하기 위해 각 영역들의 형태를 분석하여 글자의 특성을 갖는 영역외의 비 문자영역을 제거하여 실제 문자영역 검출률을 높였다. 본 논문의 정량적 평가를 위해 문자 영역의 검출률과 정확도를 이용하여 기존의 방법들과 비교하였다. 실험결과 기존의 문자 검출 방법보다 제안하는 방법이 비교적 높은 문자영역의 검출률 및 정확도를 보였다.

Keywords

OHHGBW_2018_v9n11_61_f0001.png 이미지

Fig. 1. Texts in the image. (a) Scene Text, (b) Caption Text.

OHHGBW_2018_v9n11_61_f0002.png 이미지

Fig. 2. SWT Method.

OHHGBW_2018_v9n11_61_f0003.png 이미지

Fig. 3. MSER Method.

OHHGBW_2018_v9n11_61_f0004.png 이미지

Fig. 4. Flow chart

OHHGBW_2018_v9n11_61_f0005.png 이미지

Fig. 5. Graph of Unsharp mask. (a) Boundary of Input image, (b) Boundary of Blur Image, (c) Different in Boundary part, (d) Result of Unsharp mask

OHHGBW_2018_v9n11_61_f0006.png 이미지

Fig. 6. Boundary contrast enhancement results using unsharp mask. (a) Original Image (b) Blur image (c) Detected Boundary (d) Contrast enhancement of object boundaries.

OHHGBW_2018_v9n11_61_f0007.png 이미지

Fig. 7. Flowchart of text detection in image (a) Original Image (b) Result of MSER (c) Result of Contrast Enhancement MSER.

OHHGBW_2018_v9n11_61_f0008.png 이미지

Fig. 8. Process of text area detection using text pattern analysis.

OHHGBW_2018_v9n11_61_f0009.png 이미지

Fig. 9. Results of MSER region labeling (a) Original Image (b) Result of Contour Labeling (c) Result of rectangle Labeling

OHHGBW_2018_v9n11_61_f0010.png 이미지

Fig. 10. Incorrect removal non-text results using aspect ratio. (a) Original Image (b) Comparison area (c) Result of Labeling.

OHHGBW_2018_v9n11_61_f0011.png 이미지

Fig. 11. Result of non-text area removal using aspect ratio. (a) Original Image (b) Result of Contour Labeling (c) Non-text area (d) Result of removing the non-text area.

OHHGBW_2018_v9n11_61_f0012.png 이미지

Fig. 12. Result of non-text area removal using Extent. (a) Result of aspect ratio, (b) Result of Aspect ratio, (c) Non-text area, (d) Result of removing the non-text area.

OHHGBW_2018_v9n11_61_f0013.png 이미지

Fig. 13. Test set Images

OHHGBW_2018_v9n11_61_f0014.png 이미지

Fig. 14. Training set Images.

Table 1. Performance comparison between proposed method and existing method

OHHGBW_2018_v9n11_61_t0001.png 이미지

References

  1. Khatib, T., Karajeh, H., Mohammad, H., & Rajab, L. (2015). A hybrid multilevel text extraction algorithm in scene images. Scientific Research and Essays, 10(3), 105-113. https://doi.org/10.5897/SRE2014.6146
  2. Raza, A., Siddiqi, I., Djeddi, C., & Ennaji, A. (2013). Multilingual artificial text detection using a cascade of transforms. In Document Analysis and Recognition (ICDAR), 2013 12th International Conference on, 309-313
  3. Seeri, S. V., Pujari, J. D., & Hiremath, P. S. (2016). Text Localization and Character Extraction in Natural Scene Images using Contourlet Transform and SVM Classifier. International Journal of Image, Graphics and Signal Processing, 8(5), 36. https://doi.org/10.5815/ijigsp.2016.05.02
  4. Koo, H. I., & Kim, D. H. (2013). Scene text detection via connected component clustering and nontext filtering. IEEE transactions on image processing, 22(6), 2296-2305. https://doi.org/10.1109/TIP.2013.2249082
  5. Feng, Y., Song, Y., & Zhang, Y. (2016, December). Scene text detection based on multi-scale SWT and edge filtering. In Pattern Recognition (ICPR), 2016 23rd International Conference on. 645-650.
  6. Sujatha, P., & Sudha, K. K. (2015). Performance analysis of different edge detection techniques for image segmentation. Indian Journal of Science and Technology, 8(14).
  7. Kang, L., Li, Y., & Doermann, D. (2014). Orientation robust text line detection in natural images. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 4034-4041.
  8. Gupta, N., & Banga, V. K. (2012). Localization of text in complex images using haar wavelet transform. International Journal of Innovative Technology and Exploring Engineering (IJITEE) ISSN, 2278-3075.
  9. Epshtein, B., Ofek, E., & Wexler, Y. (2010, June). Detecting text in natural scenes with stroke width transform. In Computer Vision and Pattern Recognition (CVPR), 2010 IEEE Conference on. 2963-2970.
  10. Cao, X., Ren, W., Zuo, W., Guo, X., & Foroosh, H. (2015). Scene text deblurring using text-specific multiscale dictionaries. IEEE Transactions on Image Processing, 24(4), 1302-1314. https://doi.org/10.1109/TIP.2015.2400217
  11. Yu, Y. J., (2014). Document Image Binarization Technique using MSER, Journal of the Korea Institute of Information and Communication Engineering, 18(8), 1941-1947. https://doi.org/10.6109/jkiice.2014.18.8.1941
  12. Jeong, J. W & Jun K. K, Noise Robust Document Image Binarization using Text Region Detection and Down Sampling, Journal of Korea Multimedia Society, 18(7), 843-852, 2015. https://doi.org/10.9717/kmms.2015.18.7.843
  13. Lira, J., & Rodriguez, A. (2014). Edge enhancement in multispectral satellite images by means of vector operators. Geofisica internacional, 53(3), 289-308. https://doi.org/10.1016/S0016-7169(14)71506-5
  14. Park, J. H., Lee, G. S., Kim, J. S., Ryu, S. H., & Lee, S. H. (2016). Independent Object Tracking from Video using the Contour Information in HSV Color Space. Indian Journal of Science and Technology, 9(S1).
  15. Yao, C., Bai, X., Liu, W., Ma, Y., & Tu, Z. (2012). Detecting texts of arbitrary orientations in natural images. In 2012 IEEE Conference on Computer Vision and Pattern Recognition. 1083-1090.