Comparison of Feature Selection Method with ReliefF base Multi Label Algorithm to Improve Semantic Image Classification

Main Article Content

Tejtasin Phiasai
Nutchanun Chinpanthana

Abstract

Image classification is developed as part of a framework in digital image processing. The image extraction in relevant features is the most challenging part of classification. The performance of algorithm depends on features considered from the dataset. There are many algorithms attempt to analyze and combine all features but high dimensional dataset degrades the performance of algorithm. To overcome this problem, we propose a ReliefF base on Multi Label Algorithm (ReliefF-ML) to improve semantic image classification. Feature selection technique is used as a first step to analyze in large data set that is to find relevant features and removes redundant features from high dimensional dataset. This paper presents four steps including (1) data preprocessing, (2) feature extraction, (3) ReliefF-ML feature selection model, and (4) efficiency measurement and evaluation of experimental results. The experimental results indicate that our framework offers performance improvements with ReliefF-ML feature selection algorithm. The proposed model can achieve significant improvements for image classification with maximum success rate of 78.87% with 12 features.

Downloads

Download data is not yet available.

Article Details

How to Cite
[1]
T. Phiasai and N. Chinpanthana, “Comparison of Feature Selection Method with ReliefF base Multi Label Algorithm to Improve Semantic Image Classification”, JIST, vol. 11, no. 1, pp. 88-96, Jun. 2021.
Section
Research Article: Information Systems (Detail in Scope of Journal)

References

[1] Galleguillos C. and Belongie S., “Context Based Object Categorization: A Critical Survey”, Computer Vision and Image Understanding, 114. pp. 712-722, 2010.

[2] N. Chinpanthana, “A Study of Feature Extraction Techniques used for Content Based Image Retrieval System”, Christian University Journal, Vol.23, No.1pp. 130-139, 2017.

[3] Russell, B.C., Torralba, A., Murphy, K.P. et al., “LabelMe: A Database and Web-Based Tool for Image Annotation”, Int J Comput Vision, 77. pp. 157–173, 2008.

[4] Junshi Huang, Rogerio S. Feris, Qiang Chen, Shuicheng Yan, “Cross-domain Image Retrieval with a Dual Attribute-aware Ranking Network”, Computer Vision and Pattern Recognition. pp. 1062-1070, 2015.

[5] I. Simon, N. Snavely, and S. Seitz., “Scene summarization for online image collections”, Computer Vision. pp. 1–8, 2007.

[6] Tang, J., S. Alelyani, and H. Liu, “Feature selection for classification: A review”, Data classification: Algorithms and applications. p. 37, 2014.

[7] Janecek, A., Gansterer, W., Demel, M. and Ecker, G., “On the relationship between feature selection and classification accuracy”, In New Challenges for Feature Selection in Data Mining and Knowledge Discovery, pp. 90-105, 2008.

[8] Parikh R. B., Obermeyer Z., and Bates D. W., “Making Predictive Analytics a Routine Part of patient Care.” 2016.

[9] Liu, H. and H. Motoda, “Feature selection for knowledge discovery and data mining,” Springer Science & Business Media, Vol. 454, 2012.

[10] El-Naggar, N., Y. El-Sonbaty, and M.A. El-Nasr, “Sentiment analysis of modern standard Arabic and Egyptian dialectal Arabic tweets,” IEEE Computing Conference, 2017.

[11] Omar, N., et al., “A comparative study of feature selection and machine learning algorithms for arabic sentiment classification,” Asia Information Retrieval Symposium. Springer, 2014.

[12] N. Spolaor, E. A. Cherman, M. C. Monard, “Using ReliefF for multi-label feature selection,” Proceedings of the Conferencia Latinoamericana de Informatica, Brazil, pp. 960–975, 2011.

[13] N. Spolar, E. Cherman, M. Monard, H. Lee, “Filter approach feature selection methods to support multi-label learning based on ReliefF and Information Gain”, Proceedings of the Advances in Artificial Intelligence - SBIA2012, LNCS, Springer Berlin Heidelberg, pp. 72–81, 2012.

[14] D. Kong, C. Ding, H. Huang, and H. Zhao, “Multi-label ReliefF and F-statistic Feature Selections for Image Annotation,” Proceedings of Computer Vision and Pattern Recognition, pp. 2352–2359, 2012.

[15]Liu, H. and H. Motoda, “Feature selection for knowledge discovery and data mining,” Springer Science & Business Media, Vol. 454, 2012.

[16] Tang, J., S. Alelyani, and H. Liu, “Feature selection for classification: A review,” Data classification: Algorithms and applications, p. 37, 2014.

[17] A. Kuznetsova, et al., “The Open Images Dataset V4: Unified image classification, object detection, and visual relationship detection at scale,” IJCV, 2020.

[18] Papadopolous et al., “Extreme clicking for efficient object annotation,” Computer Vision and Pattern Recognition ICCV, 2017.

[19] T. Joachims, “Text categorization with support vector machines learning with many relevant features,” In Proceedings of the 10th European Conference on Machine Learning, Chemnitz, Germany. Berlin: Springer. pp.137–42, 1998.

[20] J. Long, E. Shelhamer, and T. Darrell. “Fully convolutional networks for semantic segmentation,” In CVPR, 2015.

[21] Ren S., He K., Girshick R. and Sun J., “Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks,” IEEE International Conference on Computer Vision, 2015.