Texture image retrieval using contourlet transform (original) (raw)
Related papers
2010 2nd IEEE International Conference on Information Management and Engineering, 2010
Over the past few years, Content-based image retrieval (CBIR) has been an active research area. A rapid proliferation has been witnessed in the fields of both theoretical research and development of the CBIR system. The most commonly used transformation techniques in CBIR include wavelet and Fourier transformations; in spite of their widespread utilization, they have not been very effective in representing the image regions that are separated by smooth contours. An effective alternative, Contourlet Transformation performs well in representing the time-frequency localization of the images. In this paper, we propose a CBIR system for effective retrieval of images from a database for a given query image. The proposed CBIR system utilizes CT to extract the content of the image in terms of directional contours, horizontal and vertical edges of the image. In addition, the system extracts texture, color and spatial features from the images. In image retrieval, the system measures the similarity between the features of the query image and the images in the database using Squared Euclidean distance. Eventually, the images similar to the query image are effectively retrieved, chiefly based on the contourlet features.
A Novel Content Image Retrieval Method Based on Contourlet
2008
Content-based image retrieval is an active and fast advancing research area since the 1990s as a result of advances in the Internet and new digital image sensor technologies. However, many challenging research problems continue to attract researchers from multiple disciplines. Content-based image retrieval uses the visual contents of an image as features to represent and index the image to be searched from large scale image databases. The quality of the selected features relies mainly on the degree of the invariance property that is ensured under acceptable manipulations. This paper proposes an efficient method for compactly representing color and texture features and combining them for image retrieval. The performance of retrieval based on these compact descriptors obtained by the proposed techniques is analyzed and tested on wang database images yielding satisfactory accuracy rates. A comparative study demonstrated that the developed feature extraction scheme outperformed the other schemes being compared with.
Image Retrieval using Contourlet Transform
International Journal of Computer Applications, 2011
The image retrieval problem has recently become more important and necessary because of the rapid growth of multimedia databases and digital libraries. Different search engines use different features to retrieve images from the database. In this paper, the Contourlet Transform is developed to retrieve similar images from the image database. By combining the Laplacian pyramid and the Directional Filter Bank (DFB), a new image representation is obtained. The direction subbands coefficients are used to form a feature vector for classification. The performance of the Contourlet Transform is evaluated using standard bench marks such as Precision and Recall. An experiment shows that the Contourlet Transform (CT) features provide the best results in Image Retrieval.
Conventional content-based image retrieval schemes may suffer from practical applications. Image databases are often composed of several groups of images and span very different scales in the space of low-level visual descriptors the interactive retrieval of such image classes is then very difficult. To address this challenge, we propose the support vector machine (SVM) algorithm with contourlet transform coefficients distribution. SVM is used to find out the optimal result and to evaluate the generalization ability under the limited training samples. It gives faster result as compared to other. An SVM classifier can be learned from training data of relevance images and irrelevance images marked by users. Features of the face image are extracted in the spectrum domain using contourlet Transform and this transform addresses the problem of representing the images with smooth contours in different directions by providing two additional properties which are directionality and anisotropy. This method will overcome the introduction of the noisy examples by the users. In the proposed technique, multiple feature distances are combined to obtain image similarity. The extensive experiments are performed on two different image data bases to validate the superiority of the proposed method.
Rotated complex wavelet based texture features for content based image retrieval
Proceedings of the 17th International Conference on Pattern Recognition, 2004. ICPR 2004., 2004
In this paper we have proposed a novel approach of extracting texture features for content-based image retrieval. A new set of two-dimensional (2-D) rotated complex wavelet filters (RCWF) is designed with complex wavelet filter coefficients. 2-D RCWF are nonseparable and oriented, which improves characterization of oriented textures. Dual-tree rotated complex wavelet filter (DT-RCWF) and dual-tree complex wavelet transform (DT-CWT) are used jointly for texture analysis in twelve different directions. Texture features are obtained by computing the energy and standard deviation of each subband. Retrieval results obtained using each individual method and in combination are presented. Retrieval performance obtained with the combined filterbank is superior relative to the performance obtained using the other existing methods. New method also retains comparable levels of computational complexity.
Texture feature extraction in the spatial-frequency domain for content-based image retrieval
2010
The advent of large scale multimedia databases has led to great challenges in content-based image retrieval (CBIR). Even though CBIR is considered an emerging field of research, however it constitutes a strong background for new methodologies and systems implementations. Therefore, many research contributions are focusing on techniques enabling higher image retrieval accuracy while preserving low level of computational complexity. Image retrieval based on texture features is receiving special attention because of the omnipresence of this visual feature in most real-world images. This paper highlights the state-of-the-art and current progress relevant to texture-based image retrieval and spatial-frequency image representations. In particular, it gives an overview of statistical methodologies and techniques employed for texture feature extraction using most popular spatial-frequency image transforms, namely discrete wavelets, Gabor wavelets, dual-tree complex wavelet and contourlets. Indications are also given about used similarity measurement functions and most important achieved results.
—An image searching is one of the hottest researchesin the present era because of WWW. The Content Based ImageRetrieval (CBIR) is one of the methods for searching the imageson the bases of visual contents like color, texture and shapeetc. The CBIR uses these contents to search images from therepository instated of metadata like keywords. We have designedsuch a CBIR system that uses the image texture features extractedby applying wavelet transform on the wave properties of theimage. The wavelet transform can provide us with the frequencyof the signals and the time associated with those frequenciessimultaneously making it very convenient for feature extraction ofan image. These extracted features coefficients are given as inputto Artificial Neural Network (ANN) to get normalized valuesby assigning higher weights to coefficients. Also, Color featuresHue, Saturation and Intensity in HSI color space are extractedfrom an image. The weighted sum of these two image featuresis considered to make the retrieval more effective. The proposedsystem also uses a new image as a query which is not there indatabase by accepting input from the user and creates a newcategory by dynamically updating the database. The system istested for different query inputs and calculated the standardmeasures like precision and recall required for evaluating theproposed method. The results have been tabulated and comparedwith existing methods and shows the best performance.
AMALGAMATION OF CONTOUR, TEXTURE, COLOR, EDGE, AND SPATIAL FEATURES FOR EFFICACIOUS IMAGE RETRIEVAL
From the past few years, Content based image retrieval (CBIR) has been a progressive and curious research area. Image retrieval is a process of extraction of the set of images from the available image database resembling the query image. Many CBIR techniques have been proposed for relevant image recoveries. However most of them are based on a particular feature extraction like texture based recovery, color based retrieval system etc. Here in this paper we put forward a novel technique for image recovery based on the integration of contour, texture, color, edge, and spatial features. Contourlet decomposition is employed for the extraction of contour features such as energy and standard deviation. Directionality and anisotropy are the properties of contourlet transformation that makes it an efficient technique. After feature extraction of query and database images, similarity measurement techniques such as Squared Euclidian and Manhattan distance were used to obtain the top N image matches. The simulation results in Matlab show that the proposed technique offers a better image retrieval. Satisfactory precision-recall rate is also maintained in this method.
M-band wavelet based texture features for content based image retrieval
2002
Biorthonormal M-band wavelet transform is used to decompose the image into sub-bands for constructing the feature database in content-based image retrieval of 1856 Brodatz texture images. Texture features are obtained by computing the measure of energy, standard deviation and its combination on each band. Results are far superior and impressive than conventional two-band wavelet decomposition.