International Journal of Computer Applications (0975 – 8887) Volume 56– No.11, October 2012 28 CBIR using Textural Feature Nilam N Ghuge Research Scholar (JJT UNIV.) Assistant Professor, Electrical B.S.I.O.T.R.(W), Pune (India) Parul S Arora Bhalotra Research Scholar (JJT Univ.) Assistant Professor, ETC G.H.R.C.E.M. Pune (India) B. D. Shinde, PhD. Professor, ETC G.N.V.O.I.T., Tala Raigarh (India) ABSTRACT CBIR system focuses on retrieving images from the database; the system depends on the way the indexing is being implemented. The way or method in which an image is stored will affect how it will be retrieved later and which can save more storage space and improve the retrieval process. Building effective content-based image retrieval (CBIR) systems involves the combination of image creation, storage, security, transmission, analysis, evaluation feature extraction, and feature combination in order to store and retrieve images effectively. The goal of CBIR systems is to support image retrieval based on content i.e. shape, color, texture. In this paper we have implemented CBIR techniques using conventional Histogram and Gabor filter. We have shown results of query image and retrieved image also 2D frequency response of Gabor filter with various angles as it is direction dependent filter. We have used Euclidean distance as a measure to calculate distance between two images. General Terms Hybrid Intelligent Systems for Computer Vision and Pattern Recognition, Image Processing, Texture features, CBIR, Filter. Keywords Content based Image retrieval, Histogram, Gabor function, Euclidean distance, Precision, Recall. 1. INTRODUCTION With the rapid proliferation of the internet and the worldwide- web, the amount of digital image data accessible to users has grown enormously. Image databases are becoming larger and more widespread, and there is a growing need for effective and efficient image retrieval (IR) systems. All the information available is only useful if one can access it efficiently. This does not only mean fast access from a storage management point of view but also means that one should be able to find the desired information without scanning all information manually. An important part of digital media is image data. In contrast to text, images just consist of pure pixel data with no inherent meaning. Commercial image catalogues therefore use manual annotation and rely on text retrieval techniques for searching particular images. However, such an annotation has two main drawbacks: First, the annotation depends on the person who adds it. Naturally the result may vary from person to person and furthermore may depend on the context. Within a general image database it may be sufficient to just add an annotation like “butterfly” whereas this obviously is not sufficient for a biologist’s database consisting of different types of butterflies only. The second problem with manual annotation is that it is very time consuming. While it may be worthwhile for commercial image collections, it is prohibitive for indexing of images within the World Wide Web. One could not even keep up with the growth of available image data. So we need a system which can effectively retrieve the desired image even if the database is not annotated. Imaging is a major factor in areas such as art galleries, interior design and weather forecasting. It is important for those areas to be able to retrieve the stored image quickly and accurately. The more effective the images are being stored, the more efficient the images can be retrieved later; this is where Content-Based Image Retrieval (CBIR) indexing comes in. Several existing applications, such as Query By Image (QBIC) which handles image databases and allows user to insert queries or interact with provided interfaces have focus on CBIR. Some of the applications have even used new algorithms or methods that help bring better result in retrieval process. However, there is potential to improve the existing algorithms, which increases the effectiveness of the retrieval process. Those existing algorithms have their own advantages and disadvantages, but we can use them by trying to combine and come up with a new algorithm that reduces the limitation of existing algorithms [1]. CBIR system focuses on retrieving images from the database; the system depends on the way the indexing is being implemented. The way or method in which an image is stored will affect how it will be retrieved later. This work aim is to develop an indexing algorithm based on existing CBIR studies, which can save more storage space and improve the retrieval process. Image retrieval has become more significant when people start to have large collection of images which they need to use at some point. The idea of searching those collections one by one to match manually with what the user wants (user’s query) is a nightmare. This is where Content-Based Image Retrieval (CBIR) comes in to solve the problem. CBIR has come long way before 1990 and very little papers has been published at that time, however the number of papers published since 1997 is increasing. This indicated that more people have become interested with this area of research. There are many CBIR algorithms as the result of those researches and most of those algorithms process image into several layers of tasks. Those layers of tasks consist of extracting the multidimensional features of an image query and compare it with images in the database are perform after the system populate database with images [2]. Populating database with extracted information from the images and indexed appropriately will affect the performance of retrieval