Histogram of Oriented Gradients-Based Digit Classification Using Naive Bayesian Classifier
Classification helps in grouping the objects according to their characteristics or features, which is essential for predicting the behavior of objects, simplifying the process of searching in a large database, detecting specific objects, etc. Advancement in information technology has increased the need for classification of text documents, image, video, audio dataset for easy and accurate retrieval of required information. Selecting features where the most relevant information lies is one of the important steps before classification. In this paper, gradient information is used for feature extraction with the help of histogram of oriented gradients technique. The simplicity of naive Bayesian classifier makes it suitable for large databases. The accuracy and ROC curve prove the effectiveness of the proposed method.
KeywordsHistogram of oriented gradients Bayes’ theorem Naive Bayesian classifier Supervised learning Digit classification
- 1.Kotsiantis, S.: Supervised machine learning: A review of classification techniques. Informatica (2007) 249–268.Google Scholar
- 2.Nilsback, M., Zisserman, A.: Automated flower classification over a large number of classes. Sixth Indian Conference on Computer Vision, Graphics and Image Processing, IEEE (2008).Google Scholar
- 3.Jagannathan, S., Desappan, K., Swami, P., Mathew, M., Nagori, S., Chitnis, K., Marathe, Y., Poddar, D., Narayanan, S., Jain, A.: Efficient object detection and classification on low power embedded systems. International Conference on Consumer Electronics, IEEE (2017).Google Scholar
- 4.Tuermer, S., Kurz, F., Reinartz, P., Stilla, U.: Airborne vehicle detection in dense urban areas using HOG features and disparity maps. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing (2013).Google Scholar
- 5.Geismann, P., Schneider, G.: A two-staged approach to vision-based pedestrian recognition using Haar and HOG features. IEEE Intelligent Vehicles Symposium (2008) 554–559.Google Scholar
- 6.Subramanyam, A., Emmanuel, S.: Video forgery detection using HOG features and compression properties. International Workshop on Multimedia Signal Processing, IEEE (2012).Google Scholar
- 7.Dahmane, M., Meunier, J.: Emotion recognition using dynamic grid-based HOG features. Internatioanl Conference on Automatic Face and Gesture Recognition and Workshops, IEEE (2011) 884–888.Google Scholar
- 8.Zhang, H., Ren, J., Kang, Y., Bo, P., Liang, J., Ding, L, Kong, W., Zhang, J.: Development of novel in Silico model for developmental toxicity assessment by using naïve bayes classifier method. Reproductive Toxicology. https://doi.org/10.1016/j.reprotox. 2017. 04. 005 (2017).
- 9.Zhang, M., Peña, J., Robles, V.: Feature selection for multi-label naive bayes classification. Information Sciences, ELSEVIER (2009) 3218–3229.Google Scholar
- 10.Yamajuchi, T., Nakano, Y., Maruyama, M., Miyao, H., Hananoi, T.: Digit classification on signboards for telephone number recognition. International Conference on Document Analysis and Recognition, IEEE (2003).Google Scholar
- 11.Sermanet, P., Chintala, S., LeCun, Y.: Convolutional neural networks applied to house numbers digit classification. International Conference on Pattern Recognition, IEEE (2012).Google Scholar
- 12.Ebrahimzadeh, R., Jampour, M.: Efficient handwritten digit recognition based on histogram of oriented gradients and SVM. International Journal of Computer Applications, Vol. 104, No. 9 (2014) 10–13.Google Scholar
- 13.Tsai, G.: Histogram of Oriented Gradients. http://web.eecs.umich.edu/~silvio/teaching/EECS598_2010/slides/09_28_Grace.pdf (2010).
- 14.Han, J., Kambar, M.: Data mining concepts and techniques, Elsevier (2006).Google Scholar