Medical Image Retrieval using Dual Tree Complex Wavelet Transform and Principal Component Analysis with Haralick Texture Features
Keerthika C.1, Rajakumar K.2
1Keerthika C., School of Computer Science and Engineering, Vellore Institute of Technology, Vellore, India.
2Dr. Rajakumar K.*, School of Computer Science and Engineering, Vellore Institute of Technology, Vellore, India.

Manuscript received on January 01, 2020. | Revised Manuscript received on January 20, 2020. | Manuscript published on January 30, 2020. | PP: 3425-3435 | Volume-8 Issue-5, January 2020. | Retrieval Number: E6511018520/2020©BEIESP | DOI: 10.35940/ijrte.E6511.018520

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: Noise and distortion occurs in all types of medical images (Computed Tomography (CT), Magnetic Resonance Imaging (MRI.)..) and are unavoidable during the stages of image acquisition. We use medical image retrieval to extract the images from database by texture, shaptrix or color features. We use Dual Tree Complex Wavelet Transform (DTCWT) and Principal Component Analysis (PCA). DTCWT extracts the information of images. PCA compress the images. It also minimizes the feature vectors dimensions of all images. Haralick texture features are extracted from images with the co-occurrence matrix. This matrix describes the relationship of pixels. The similar images are found by calculating the similarity measure of the query image and all images in database by Mahalanobis distance. This method retrieves the similar images from database with respect to the input image provided by the user. The performance of the proposed algorithm can be found by precision and recall measures for evaluation. This system can be used in hospitals, clinics etc., for detecting diseases earlier.
Keywords: Dual Tree Complex Wavelet Transform (DTCWT), Haralick texture features, Medical Image Retrieval, Principal Component Analysis (PCA).
Scope of the Article: Biomedical Computing.