Abstract:In this correspondence, we propose a novel, efficient, and effective Refined Histogram (RH) for modeling the wavelet subband detail coefficients and present a new image signature based on the RH model for supervised texture classification. Our RH makes use of a step function with exponentially increasing intervals to model the histogram of detail coefficients, and the concatenation of the RH model parameters for all wavelet subbands forms the so-called RH signature. To justify the usefulness of the RH signatur… Show more
“…It is further worth noting why we omitted comparative tests against some of the approaches mentioned in Section I, such as the Refined Histogram technique [17] or a Hidden Markov Tree based approach. First, the Refined Histogram technique is particularly designed to enhance runtime behavior of the GG based margin-only approach of Do & Vetterli [5].…”
Section: Discussion and Concluding Remarksmentioning
confidence: 99%
“…A considerable amount of research work has been devoted to the statistical modeling of DCT or DWT subband coefficients, primarily focusing on the Generalized Gaussian (GG) distribution [5], [14], [15]. Recently, the Generalized Gamma (GGamma) distribution [16] and the Refined Histogram [17] have been proposed as reasonable alternatives. In particular, the Refined Histogram approach is based on the idea of modeling DWT coefficients by Product Benoulli distributions [18], [19] which can be estimated very efficiently.…”
Abstract-In this article, we investigate a novel joint statistical model for subband coefficient magnitudes of the Dual-Tree Complex Wavelet transform which is then coupled to a Bayesian framework for Content-Based Image Retrieval. The joint model allows to capture the association among transform coefficients of the same decomposition scale and different color channels. It further facilitates to incorporate recent research work on modeling marginal coefficient distributions. We demonstrate the applicability of the novel model in the context of color texture retrieval on four texture image databases and compare retrieval performance to a collection of state-of-the-art approaches in the field. Our experiments further include a thorough computational analysis of the main building blocks, runtime measurements and an analysis of storage requirements. Eventually, we identify a model configuration with low storage requirements, competitive retrieval accuracy and a runtime behavior which enables the deployment even on large image databases.
“…It is further worth noting why we omitted comparative tests against some of the approaches mentioned in Section I, such as the Refined Histogram technique [17] or a Hidden Markov Tree based approach. First, the Refined Histogram technique is particularly designed to enhance runtime behavior of the GG based margin-only approach of Do & Vetterli [5].…”
Section: Discussion and Concluding Remarksmentioning
confidence: 99%
“…A considerable amount of research work has been devoted to the statistical modeling of DCT or DWT subband coefficients, primarily focusing on the Generalized Gaussian (GG) distribution [5], [14], [15]. Recently, the Generalized Gamma (GGamma) distribution [16] and the Refined Histogram [17] have been proposed as reasonable alternatives. In particular, the Refined Histogram approach is based on the idea of modeling DWT coefficients by Product Benoulli distributions [18], [19] which can be estimated very efficiently.…”
Abstract-In this article, we investigate a novel joint statistical model for subband coefficient magnitudes of the Dual-Tree Complex Wavelet transform which is then coupled to a Bayesian framework for Content-Based Image Retrieval. The joint model allows to capture the association among transform coefficients of the same decomposition scale and different color channels. It further facilitates to incorporate recent research work on modeling marginal coefficient distributions. We demonstrate the applicability of the novel model in the context of color texture retrieval on four texture image databases and compare retrieval performance to a collection of state-of-the-art approaches in the field. Our experiments further include a thorough computational analysis of the main building blocks, runtime measurements and an analysis of storage requirements. Eventually, we identify a model configuration with low storage requirements, competitive retrieval accuracy and a runtime behavior which enables the deployment even on large image databases.
“…Nevertheless, the WNISM has some limitations, which are investigated in detail by Ma et al [13] and are summarized as: (a) the coefficient connectivity is not taken into account in the wavelet subbands [10]; (b) it performs well on a single distortion type but is not effective for multiple distortions when analyzed together [14]; (c) in the case of KLD, it is computationally expensive when two different GGDs are computed [15], which indeed is not feasible for practical applications. In addition, since KLD is asymmetric [16], the distance between the distorted and original images is not the same.…”
Abstract:Reduced reference image quality assessment does not require the presence of the original image for assessing the quality of a degraded image. This work proposes an intelligent method for reduced reference image quality assessment based on a reorganized discrete cosine transform (RDCT). A genetic algorithm (GA) is used to compute optimized estimation of the generalized Gaussian distribution (GGD), which then approximates the coefficient distribution in the RDCT domain. Experimental results validate that such an intelligent estimation produces far superior results compared to conventional empirical estimation methods as presented in the literature. We compare the proposed technique with a number of contemporary techniques in the literature and demonstrate the generalization capability and effectiveness of the proposed technique as compared to prior works.
“…Recently, models based on wavelet subband coefficients have also been used on texture classification. The existing models in literatures contain the Characteristic Generalized Gaussian Density (CGGD) model [12], the Bit-plane Probability (BP) model [13,14], the Refined Histogram [15], the Local Energy Histogram [16], and so on. Particularly, the Bit-plane Probability (BP) signature is a very competitive feature by modeling wavelet high-frequency subband coefficients via the Product Bernoulli Distributions (PBD).…”
A new texture classification method based on singular value decomposition(SVD) and wavelet transform is presented. Wavelet transform is employed on texture images having been preprocessed with SVD. The elements of the signature vector of an image are the fractal dimensions and barycentric coordinates of the bit planes of the wavelet coefficients in both the 3-Level high frequency domains and the third low frequency domain. The one-nearest-neighbor classifier with standard L ଵ -norm distance is utilized to perform supervised texture classification. Compared with some other classification methods, the method is experimentally proved more efficient and less time-consuming.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.