The need for efficient content-based image retrieval has increased tremendously in many application areas such as biomedicine, military, commerce, education, and Web image classification and searching. We present here SIMPLIcity (Semanticssensitive Integrated Matching for Picture LIbraries), an image retrieval system, which uses semantics classification methods, a wavelet-based approach for feature extraction, and integrated region matching based upon image segmentation. As in other regionbased retrieval systems, an image is represented by a set of regions, roughly corresponding to objects, which are characterized by color, texture, shape, and location. The system classifies images into semantic categories, such as textured-nontextured, graphphotograph. Potentially, the categorization enhances retrieval by permitting semantically-adaptive searching methods and narrowing down the searching range in a database. A measure for the overall similarity between images is developed using a region-matching scheme that integrates properties of all the regions in the images. Compared with retrieval based on individual regions, the overall similarity approach 1) reduces the adverse effect of inaccurate segmentation, 2) helps to clarify the semantics of a particular region, and 3) enables a simple querying interface for region-based image retrieval systems. The application of SIMPLIcity to several databases, including a database of about 200,000 general-purpose images, has demonstrated that our system performs significantly better and faster than existing ones. The system is fairly robust to image alterations.

This paper treats a multiresolution hidden Markov model for classifying images. Each image is represented by feature vectors at several resolutions, which are statistically dependent as modeled by the underlying state process, a multiscale Markov mesh. Unknowns in the model are estimated by maximum likelihood, in particular by employing the expectation-maximization algorithm. An image is classified by finding the optimal set of states with maximum a posteriori probability. States are then mapped into classes. The multiresolution model enables multiscale information about context to be incorporated into classification. Suboptimal algorithms based on the model provide progressive classification that is much faster than the algorithm based on single-resolution hidden Markov models.

In the last decade, multiscale techniques for gray-level texture analysis have been intensively used. In this paper, we aim on extending these techniques to color images. Weintroduce wavelet energy-correlation signatures and we derive the transformation of these signatures upon linear color space transformations. Experiments are conducted on a set of 30 natural colored texture images in which color and graylevel texture classi#cation performances are compared. It is demonstrated that the wavelet correlation features contain more information than the intensity or the energy features of each color plane separately. The in#uence of image representation in color space is evaluated. Key words: texture analysis, classi#cation, color spaces, feature extraction, wavelet signatures 1 Introduction For image analysis, color and texture are two of the most important properties, especially when one is dealing with real world images. Classical image analysis schemes only takeinto account the pix...

In this paper, an algorithm is developed for segmenting document images into four classes: background, photograph, text, and graph. Features used for classi#cation are based on the distribution patterns of wavelet coe#cients in high frequency bands. Two important attributes of the algorithm are its multiscale nature---it classi#es an image at di#erent resolutions adaptively, enabling accurate classi#cation at class boundaries as well as fast classi#cation overall--- and its use of accumulated context information for improving classi#cation accuracy. Keywords document image segmentation, text and photograph segmentation, multiscale classi#cation, context-dependent classi#cation, wavelet transform, goodness of match. I. Introduction The image classi#cation problem considered in this paper is the segmentation of an image into four classes: background, photograph, text, and graph. Photograph refers to continuous-tone images, such as scanned pictures. Text is interpreted in the na...

This paper presents an algorithm to segment text and picture in an image using two features based on the statistical distribution of the wavelet coefficients in high frequency bands. The algorithm breaks the image into blocks and classifies every block as background, text or picture according to the two features. The block size is variable so that the segmentation can be accurate at the boundary of two types and avoids misclassifying due to over-localized region analysis. 1 Introduction Statistical classification is an important topic in image processing. Classification, which helps to interpret an image, can also be incorporated with other image processing to improve performance. One well-known example is image compression. For training-based image compression algorithms, such as vector quantization [1], a codebook is optimally designed under the assumption that the data to be quantized is statistically consistent with the training data. Hence, different quantizers are required for ...

This paper presents a new general framework for image segmentation. A level set formulation is used to model the boundaries of the image regions and a new Multilabel Fast Marching is introduced for the evolution of the region contours toward the segmentation result. Statistical tests are performed to yield an initial estimate of high-confidence subsets of the image regions. Furthermore, the velocities for the propagation of the region contours are defined in accordance with the a posteriori probability of the respective regions, leading to the Bayesian Level Set methodology described in this paper. Typical segmentation problems are considered and experimental results are given to illustrate the robustness of the method against noise and its performance in precise region boundary localization. C ○ 2002 Elsevier Science (USA) 1.

presented to the University ofWaterloo

Abstract not found

Luminance, colour, and/or texture features may be used, either alone or in combination, for segmentation. In this paper luminance and colour classes are described using the corresponding empirical probability distributions. For texture analysis and characterisation a multichannel scale/orientation decomposition is performed using wavelet frame analysis. Knowing only the number of the di#erent classes of the image, regions of homogeneous patterns are identified. On these regions the features characterising and describing the di#erent classes are estimated. Two labelling algorithms are proposed. The first, a deterministic relaxation algorithm using a quadratic distance measure, yields the labelling of pixels to the di#erent colour --texture classes. The second is a new Multi-label Fast Marching algorithm utilising a level set boundary determination.

This report gives an introduction to the application of wavelet based multiscale image analysis methods to texture analysis. It outlines the basic methods and comments on design issues. It also points to the literature and connections with related methods. The unfinishedness of the research in this area becomes clear from the discussion of some extra issues and from a short list of practical applications. (latest revision 30/06/97) I. Introduction Texture is an important cue for the analysis of many types of images. The term is used to point to intrinsic properties of surfaces, especially those that don't have a smoothly varying intensity. It includes intuitive properties like roughness, granulation and regularity. Some example textures from the Brodatz album [3] are shown in Fig. 1. Texture can be defined as the set of local neighbourhood properties of the gray levels of an image region. Texture analysis is considered a challenging task. The ability to effectively classify and segme...