Authors : Jinsa Kuruvilla and K. Gunavathi
Abstract: In this study, researchers present Content Based Image Retrieval (CBIR) System for Computed Tomography (CT) images of lungs. When a query image is given to the system, the system will retrieve the images which are similar to the query image from a database of cancerous and non-cancerous images. In CBIR Systems, the visual contents of the images in the database are extracted and stored as feature vectors to form a feature database. The Gray Level Co-occurrence Matrix (GLCM) parameters and statistical parameters are used to form the feature vectors. The parameters which are most relevant for retrieval process are found by artificial neural network classifier. Similarity measure plays an important role in CBIR Systems. The similarity comparison is done by different distance measures like Euclidean, Cityblock, Chebychev, Tversky, Manhattan, Canberra, Bray-Curtis, Squared Chord and Chi Squared. They calculate the similarities between the query image and images in the database. Different similarity measures have different effects in an Image Retrieval System. It is important to find the best similarity measure for CBIR System. The performance of the system is evaluated by Precision Rate (PR). The maximum retrieval rate obtained for cancerous images is 95% by GLCM parameters contrast and dissimilarity with modified Bray-Curtis distance.
Jinsa Kuruvilla and K. Gunavathi, 2014. Content Based Image Retrieval for CT Images of Lungs. International Journal of Soft Computing, 9: 386-390.