An Efficient Image Retrieval System using GLCM Features and Kullback-Leibler Divergence
Ch. Kodanda Ramu1, T. Sita Mahalakshmi2
1Ch. Kodanda Ramu*, research scholar in CSE Department, GITAM (Deemed to be university), Vizag, Andhra Pradesh, India.
2Dr. T. Sita Mahalakshmi, professor in the CSE Department, GITAM (Deemed to be University), Vizag, Andhra Pradesh, India.
Manuscript received on January 26, 2020. | Revised Manuscript received on February 05, 2020. | Manuscript published on February 30, 2020. | PP: 3180-3184 | Volume-9 Issue-3, February 2020. | Retrieval Number: C6062029320/2020©BEIESP | DOI: 10.35940/ijeat.C6062.029320
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Image processing is a process of extracting features from an image. The features of the image are extracted using the correlation model, based on Gray-Level Co-Occurrence Matrix (GLCM). Each of the images considered for data set are converted into gray level before applying Gaussian Mixture Model (GMM). The features extracted from GLCM are given as an input to the model-based technique so that the relative Probability Density Functions (PDF) are extracted. The comparison is carried out in the same manner by identifying the relative PDF of the training set and test data by using KullbackLeibler divergence method (KL-Divergence). In this paper an attempt is made for developing an effective model to retrieve the images based on features by considering GLCM and GMM. The results derived show that the proposed methodology is able to retrieve images more effectively.
Keywords: GLCM, GMM, PDF, Correlation, KL-Divergence, MSE, RMSE, PSNR, IF.