Data Mining-Based CBIR System

Introduction

Data mining is the process of extracting useful non-trivial knowledge from huge data repositories. These techniques have gained popularity in various application domains including medicine where a huge amount of data is available in the form of medical images. However, the subjective analysis of these images by radiologists is vulnerable to inter and intra-observer variability. These variations depend upon image quality as well as the expertise of the radiologists. Computer-aided diagnosis (CAD) systems are therefore employed as second readers to improve upon the precision in diagnosis; thereby enhancing the survival rates. Hence, there is a considerable interest among researchers to develop medical image mining techniques to design efficient CAD systems using medical images obtained through various imaging modalities, like X-Ray, Computed Tomography (CT), Magnetic Resonance Imaging (MRI) and Ultrasound (US).

Images are the effective medium for presenting visual data in many applications of industry and academia. With the development of technology, a large amount of images are being generated every day. Therefore, managing and indexing of images become essential in order to retrieve similar images effectively. In conventional systems, images are generally indexed with textual annotation. However, as the database grows larger, the use of keywords based method to retrieve a particular image becomes inefficient. Besides, skilled manual labor is required to annotate every single image with appropriate keywords. As consequences, it becomes a time-consuming and tedious task. Seeing these limitations in the conventional system, content based image retrieval (CBIR) systems came into existence. Content-based image retrieval (CBIR) or Content-base visual information retrieval (CBVIR) is an application of computer vision technique. In CBIR, retrieval of images is based on their visual attributes / features such as color, texture, and shape etc (R.Senthil Kumar et al., 2013). In CBIR, each image that is stored in database has its features extracted and compared to the features of the query image. It involves two steps:

• •

  Feature Extraction: The first step in the process is extracting image features to a distinguishable extent.

• •

  Matching: The second step involves matching these features to yield a result that is visually similar (Shadma Parveen et al., 2013).

Based on the matching criteria, images can be retrieved from a larger database efficiently. This technique can be used successfully in the following areas:

• •

  Security Check: Finger print or retina scanning for access privileges.

• •

  Medical Diagnosis: Using CBIR in a medical database of medical images to aid diagnosis by identifying similar past cases.

• •

  Crime Prevention: Automatic face recognition system, used by police forces.

• •

  Automated Inspections system for the manufacturing industries.