The Content-Based Image Retrieval using the Pulse Coupled Neural Network

Abstract

The image matching is one of the most important techniques in the image processing and the method using Pulse Coupled Neural Network (PCNN) was proposed in conventional study. Authors had been proposed a learning method to define the parameters in the PCNN for image matching. The learning method improves the performance of image matching using the PCNN. On the other hand, recently, a lot of researches on the Content-Based Image Retrieval (CBIR) have been studied. Also, some applications such as "Image search by the image" have been developed recently and these are based on this CBIR technique. In this study, we propose the method of the CBIR based on the image matching using PCNN. In our CBIR system, we assume that the image data is used as a query for image search. In the simulation results, we show the results of the retrieval for learned or unlearned query image. Also, we evaluate the performances of the system using recall and precision characteristics. From these results, we discuss the applicability of the method and we show the validity of the implementation of the image matching technique using PCNN to CBIR system.

Hardware implementation of IC-PCNN for the color image segmentation

“…Recently, a lot of engineering applications of the PCNN have been proposed, especially in the field of image processing [3]∼ [9]. For example, the PCNN achieves the image segmentation, the image matching, and so on.…”