Content Based Image Retrieval Using Color Layout Descriptor and Generic Fourier Descriptor

摘要

Image databases have been a significant part of systems or applications in many fields including education, forensics and medical sciences. However, due to the increase in the number of digital resources especially images, the demand to manage these databases such as storing and retrieving the images effectively and efficiently from these databases is crucial. Content Based Image Retrieval (CBIR) is one of the best techniques used to retrieve similar images from the image database. Even though it is a good and well-researched technique, still there are some issues to be improved. One of the challenges in CBIR is to represent the image as a feature vector. In this paper, we proposed a new feature vector using Generic Fourier Descriptor (GFD) and Color Layout Descriptor (CLD) to increase the accuracy of the retrieval process. We tested our technique with the Coral Database and compared the results with other CBIR techniques. Proposed method achieved better results than previous techniques. The proposed technique can be used by the forensic department for identification of criminal suspect using forensic database.