A Comparative Study of the Classification Algorithms Based on Feature Selection
Image classification is a large and growing research field with its applications in the areas of CBIR (Content Based Image Retrieval), image mining and automatic image annotation. In this digital era there is a huge voluminous multimedia data available and the challenge lies in retrieving and classifying the most similar images based upon an input query. Images can be classified according to their nature, content or domain and Feature extraction is the key process to classify the images accordingly. In this paper, an attempt is made to calculate all the possible features of an image based on color, texture, shape, and statistical. Based up on the features the images are further classified, studied and compared with four Classification algorithms namely Naïve Bayes, Instance Based Learning, J48 and Random forest Classification. Further the classification is applied on a prescribed set of features, so as to test the best feature set for the query image to be classified. An image database of 1150 images divided into 17 categories are considered for Classification and a brief comparative study is done.
KeywordsClassification Image Classification Feature Extraction Classification Algorithms Naïve Bayes Random Forest Classification J48 Instance based learning
Unable to display preview. Download preview PDF.
- 1.Le Saux, B., Amato, G.: Image Classifiers For Scene Analysis (2003)Google Scholar
- 2.Harini, D.N.D., Lalitha Bhaskari, D.: Image Mining Issues and Methods Related to Image Retrieval System. International Journal of Advanced Research in Computer Science 2(4) (July-August 2011) ISSN No. 0976-5697Google Scholar
- 3.Gholap, J.: Performance Tuning Of J48 Algorithm for Prediction of Soil FertilityGoogle Scholar
- 4.Kouzani, A.Z., Nahavandi, S., Khoshmanesh: Face classification by a random forest. In: K. TENCON 2007 - 2007 IEEE Region 10 Conference. Deakin Univ., Geelong (October 30-November 2, 2007)Google Scholar
- 5.Harini, D.N.D., Lalitha Bhaskari, D.: Image Retrieval System Based on Feature Extraction and Relevance Feedback. ACM (2012)Google Scholar
- 6.Harini, D.N.D., Lalitha Bhaskari, D.: Identification of Leaf Diseases in TomatoPlant Based on Wavelets and PCA. IEEE (2011), doi:978-1-4673-0125-1_cGoogle Scholar
- 7.Hendrickx, I., van den Bosch, A.: Hybrid Algorithms with Instance-Based Classification. In: Gama, J., Camacho, R., Brazdil, P.B., Jorge, A.M., Torgo, L. (eds.) ECML 2005. LNCS (LNAI), vol. 3720, pp. 158–169. Springer, Heidelberg (2005)Google Scholar
- 8.Kim, S., Jin, X., Han, J.: DislClass: discriminative frequent pattern based image classification. In: MDMKDD 2010, July 25. ACM (2010), doi:978-1-4503-0220-3Google Scholar
- 9.Philippe, M.: A comparison of active classification methods for content- Based Image Retrieval. In: CVDB 2004. ACM, Paris (2004), doi:1-58113-917-9/04/06Google Scholar
- 10.Aha, D.W., Kibler, D., Albert, M.K.: Instance based learning algorithms. In: Machine Learning, vol. 6, pp. 37–66. Kluwer Academic Publishers, Boston (1991), Manufactured in The NetherlandsGoogle Scholar
- 11.Pitchumani Angayarkanni, A.S., Kamal, B.N.B.: Automatic Classification Of Mammogram MRI using Dendograms. AJCSIT 2(4), 78–81 (2012) ISSN 2249-5126Google Scholar
- 12.Bosch, Univ. of Girona, Zisserman, M.: Image classification using Random Forests and Ferns. In: IEEE 11th International Conference on Computer Vision, ICCV 2007, pp. 1–8 (2007) ISSN:1550-5499Google Scholar