Advertisement

An efficient character recognition method using enhanced HOG for spam image detection

  • Fatemeh Naiemi
  • Vahid GhodsEmail author
  • Hassan Khalesi
Methodologies and Application
  • 20 Downloads

Abstract

Generally, a spam image is an unsolicited message electronically sent to a wide group of arbitrary addresses. Due to attractiveness and more difficult detection, spam images are the most complicated type of spam. One of the ways to encounter the spam images is an optical character recognition, OCR, method. In this paper, the proposed enhanced HOG feature extraction method has been used so that the optical character recognition system of spam has been enhanced by using the HOG feature extraction method in such a way to be both resistant against the character variations on scale and translation and to be computationally cost-effective. For these purposes, two steps of the cropped image and input image size normalization have been added to pre-processing stages. Support vector machine, SVM, was employed for classification. Two heuristic modifications including thickening of the thin characters in the pre-processing stage and non-discrimination in detecting the uppercase and lowercase letters with the same shapes in the classification stage have been also proposed to increase the system recognition accuracy. In the first heuristic modification, when all pixels of the output image are empty (the character is eliminated), the original image was made thicker by one layer. In the second modification, when recognizing the letters, no differentiation was considered between the uppercase and lowercase letters with the same shapes. An average recognition accuracy of the modified HOG method with two heuristic modifications equals 91.61% on Char74K database. Then, an optimum threshold for classification was investigated by ROC curve. The optimal cutoff point was 0.736 with the highest average accuracy, 94.20%, and AUC, area under curve, for ROC and precision–recall, PR, curves were 0.96 and 0.73, respectively. The proposed method was also examined on ICDAR2003 database, and the average accuracy and its optimum using ROC curve were 82.73% and 86.01%, respectively. These results of recognition accuracy and AUC for ROC and PR curve showed an outstanding enhancement in comparison with the best recognition rate of the previous methods.

Keywords

Spam detection OCR Histogram of oriented gradients Enhanced HOG SVM Social media Security ROC curve 

Notes

Compliance with ethical standards

Conflict of interest

The authors declare that they have no conflict of interest.

Ethical approval

This article does not contain any studies with human participants performed by any of the authors.

Informed consent

Informed consent was obtained from all individual participants included in the study.

References

  1. Alghamdi B, Watson J, Xu Y (2016) Toward detecting malicious links in online social networks through user behavior. In: IEEE/WIC/ACM international conference on web intelligence workshops (WIW). IEEE, pp 5–8Google Scholar
  2. Attar A, Rad RM, Atani RE (2013) A survey of image spamming and filtering techniques. Artif Intell Rev 40(1):71–105CrossRefGoogle Scholar
  3. Bhowmick A, Hazarika SM (2016) Machine learning for e-mail spam filtering: review, techniques and trends. arXiv preprint arXiv:1606.01042
  4. Bowling JR, Hope P, Liszka KJ (2008) Spam image identification using an artificial neural network. The University of Akron Akron, Ohio, pp 44003–44325Google Scholar
  5. Brodić D, Milivojević ZN, Maluckov ČA (2015) An approach to the script discrimination in the Slavic documents. Soft Comput 19(9):2655–2665CrossRefGoogle Scholar
  6. Camastra F (2007) A SVM-based cursive character recognizer. Pattern Recognit 40(12):3721–3727CrossRefzbMATHGoogle Scholar
  7. Chen C, Wang Y, Zhang J, Xiang Y, Zhou W, Min G (2017a) Statistical features-based real-time detection of drifted Twitter spam. IEEE Trans Inf Forensics Secur 12(4):914–925CrossRefGoogle Scholar
  8. Chen J, Zhao H, Yang J, Zhang J, Li T, Wang K (2017b) An intelligent character recognition method to filter spam images on cloud. Soft Comput 21(3):753–763CrossRefGoogle Scholar
  9. Constine J (2016) Facebook climbs to 1.59 billion users and crushes Q4 estimates with $5.8B revenue. http://techcrunch.com/2016/01/27/facebook-earnings-q4-2015/. Accessed 21 July 2017
  10. Cumming JG (2010) The spammer’s compendium. http://www.jgc.org/tsc.html. Accessed Mar 2018
  11. Dalal N, Triggs B (2005) Histograms of oriented gradients for human detection. In: IEEE Computer Society conference on computer vision and pattern recognition, CVPR 2005. IEEEGoogle Scholar
  12. Das M, Prasad V (2014) Analysis of an image spam in email based on content analysis. Int J Nat Lang Comput (IJNLC) 3(3):129–140CrossRefGoogle Scholar
  13. De Campos TE, Babu BR, Varma M (2009) Character recognition in natural images. In: Proceedings of the Int’l conference on computer vision theory and applicationGoogle Scholar
  14. Derrac J, García S, Molina D, Herrera F (2011) A practical tutorial on the use of nonparametric statistical tests as a methodology for comparing evolutionary and swarm intelligence algorithms. Swarm Evol Comput 1(1):3–18CrossRefGoogle Scholar
  15. Dhanaraj S, Karthikeyani V (2013) A study on e-mail image spam filtering techniques. In: International conference on pattern recognition, informatics and mobile engineering (PRIME). IEEEGoogle Scholar
  16. Fan G-F, Peng L-L, Hong W-C (2018) Short term load forecasting based on phase space reconstruction algorithm and bi-square kernel regression model. Appl Energy 224:13–33CrossRefGoogle Scholar
  17. Friedman M (1940) A comparison of alternative tests of significance for the problem of m rankings. Ann Math Stat 11(1):86–92MathSciNetCrossRefzbMATHGoogle Scholar
  18. Fumera G, Pillai I, Roli F (2006) Spam filtering based on the analysis of text information embedded into images. J Mach Learn Res 7:2699–2720Google Scholar
  19. Galdi P, Tagliaferri R (2019) Data mining: accuracy and error measures for classification and prediction. In: Shoba R (ed) Reference module in life sciences, Encyclopedia of Bioinformatics and Computational Biology, vol 1. Elsevier, Amsterdam, pp 1–14Google Scholar
  20. Gao Y, Choudhary A, Hua G (2010) A nonnegative sparsity induced similarity measure with application to cluster analysis of spam images. In: IEEE international conference on acoustics speech and signal processing (ICASSP). IEEEGoogle Scholar
  21. Jithesh K, Sulochana K, Kumar RR (2003) Optical character recognition (OCR) system for Malayalam language. In: National workshop on application of language technology in Indian languagesGoogle Scholar
  22. Kaur R, Singh S, Kumar H (2018) Rise of spam and compromised accounts in online social networks: a state-of-the-art review of different combating approaches. J Netw Comput Appl 112:53–88CrossRefGoogle Scholar
  23. Keys R (1981) Cubic convolution interpolation for digital image processing. IEEE Trans Acoust Speech Signal Process 29(6):1153–1160MathSciNetCrossRefzbMATHGoogle Scholar
  24. Krasser S, Tang Y, Gould J, Alperovitch D, Judge P (2007) Identifying image spam based on header and file properties using C4. 5 decision trees and support vector machine learning. In: Information assurance and security workshop, IAW’07. IEEE SMC, IEEEGoogle Scholar
  25. Li F, Shen Q, Li Y, Parthaláin NM (2015) Handwritten Chinese character recognition using fuzzy image alignment. Soft Comput 20(8):2939–2949CrossRefGoogle Scholar
  26. Liu T-J, Tsao W-L, Lee C-L (2010) A high performance image-spam filtering system. In: Ninth international symposium on distributed computing and applications to business engineering and science (DCABES). IEEEGoogle Scholar
  27. Lucas SM, Panaretos A, Sosa L, Tang A, Wong S, Young R (2003) ICDAR 2003 robust reading competitions. In: Seventh international conference on document analysis and recognition, proceedings. IEEE Computer SocietyGoogle Scholar
  28. Mehta B, Nangia S, Gupta M, Nejdl W (2008) Detecting image spam using visual features and near duplicate detection. In: Proceedings of the 17th international conference on World Wide Web. ACMGoogle Scholar
  29. Narudin FA, Feizollah A, Anuar NB, Gani A (2014) Evaluation of machine learning classifiers for mobile malware detection. Soft Comput 20(1):343–357CrossRefGoogle Scholar
  30. Saito T, Rehmsmeier M (2015) The precision–recall plot is more informative than the ROC plot when evaluating binary classifiers on imbalanced datasets. PLoS ONE 10(3):e0118432CrossRefGoogle Scholar
  31. Saraubon K, Limthanmaphon B (2009) Fast effective botnet spam detection. In: Fourth international conference on computer sciences and convergence information technology, ICCIT’09. IEEEGoogle Scholar
  32. Sathiya V, Divakar M, Sumi T (2011) Partial image spam e-mail detection using OCR. Int J Eng Trends Technol 1(1):55–59Google Scholar
  33. Sharaff A, Nagwani NK, Dhadse A (2016) Comparative study of classification algorithms for spam email detection. In: Shetty NR, Patnaik LM, Hamsavath PN, Nalini N (eds) Emerging research in computing, information, communication and applications. Springer, New Delhi, pp 237–244CrossRefGoogle Scholar
  34. Smith C (2018) 400 Interesting twitter stats and facts | By the Numbers. https://expandedramblings.com/index.php/twitter-stats-facts/
  35. Steinwart I, Christmann A (2008) Support vector machines. Springer, BerlinzbMATHGoogle Scholar
  36. Wakade SV (2011) Classification of image spam. University of Akron, AkronGoogle Scholar
  37. Wang K, Babenko B, Belongie S (2011) End-to-end scene text recognition. In: IEEE international conference on computer vision (ICCV). IEEEGoogle Scholar
  38. Xu J, Huang Y (2006) Using SVM to extract acronyms from text. Soft Comput 11(4):369–373CrossRefGoogle Scholar
  39. Xu Z, Wang H-G, Shao Z-Z (2009) Evaluation of image spam classification system based on AHP. In: International conference on computational intelligence and software engineering, CiSE 2009. IEEEGoogle Scholar

Copyright information

© Springer-Verlag GmbH Germany, part of Springer Nature 2019

Authors and Affiliations

  1. 1.Department of Electrical and Computer Engineering, Semnan BranchIslamic Azad UniversitySemnanIran
  2. 2.Department of Electrical and Computer Engineering, Semnan BranchIslamic Azad UniversitySemnanIran
  3. 3.Department of Electrical and Computer Engineering, Garmsar BranchIslamic Azad UniversityGarmsarIran

Personalised recommendations