This work presents a novel approach for detecting and classifying melanocytic skin lesions on macroscopic images. We oversegment the skin lesions using superpixels, and classify independently each superpixel as a benign or malignant using local and contextual information. The overall superpixel classification results allow to calculate an index of malignancy or benignity for the skin lesion. Using the proposed approach it is possible to discriminate a malignant from a benign skin lesion by recognizing early signs of malignancy in parts of the segmented skin lesion. The experimental results are promising, and show a potential accuracy of 99.34% on a popular data set, outperforming the current state-of-art methods.

This paper presents an approach for segmenting natural scenes based on the underlying texture characteristics using a stochastic region merging strategy. Texture region models are constructed from patch-based stochastic texture features using a texton dictionary learning approach. Finally, a stochastic region merging strategy performs the image segmentation based on texture region likelihood. Compared with other state-of-the-art texture segmentation methods, our experimental results suggest that our approach potentially can handle better highly textured regions commonly found in natural scenes, and also can be more robust to color and illumination variations.

Active shape models is an adaptive shape-matching technique that has been used for locating facial features in images. However, when a number of features is extracted for each landmark point, distortions caused by noise or illumination, and the dimensionality of the final representation, have a negative impact in the performance of a classifier. In this paper, an evolutionary wrapper for selection of the most relevant set of features for face recognition is presented. The proposed strategy explores the space of multiple feasible selections using genetic algorithms. Experimental results show that the proposed approach allows to improve the classification performance in comparison with another enhanced method and a state of the art face recognition approach.

The lip motion detection stands out as relevant visual feature for detecting the active speaker and speech recognition. In this paper, a new approach for lips and visual voice activity detection is proposed. First, the algorithm performs skin segmentation to reduce the search area for lip extraction, and the most likely lip and non-lip regions are detected using a Bayesian approach within the delimited area. Then, the final lip segmentation is obtained by thresholding the calculated probability regions and applying simple morphological operators. Finally, the temporal motion of the lips is explored using Hidden Markov Models (HMMs) to detect the likely occurrence of active speech within a temporal window.

In this paper, a computerized approach to segmenting prostate lesions in transrectal ultrasound (TRUS) images is presented. The segmentation of prostate lesions from TRUS images is very challenging due to issues, such as poor contrast, low SNRs, and irregular shape variations. To address these issues, a novel approach is employed to segment the lesions from the surrounding prostate, where region merging is performed via a region-merging likelihood function based on regional statistics, as well as Fisher-Tippett statistics. Experimental results using TRUS prostate images demonstrate that the proposed Fisher-Tippett region-merging approach achieves more accurate segmentation of prostate lesions when compared to other segmentation methods.

An automatic method for segmenting skin lesions in conventional macroscopic images is presented. The images are acquired with conventional cameras, without the use of a dermoscope. Automatic segmentation of skin lesions from macroscopic images is a very challenging problem due to factors such as illumination variations, irregular structural and color variations, the presence of hair, as well as the occurrence of multiple unhealthy skin regions. To address these factors, a novel iterative stochastic region-merging approach is employed to segment the regions corresponding to skin lesions from the macroscopic images, where stochastic region merging is initialized first on a pixel level, and subsequently on a region level until convergence. A region merging likelihood function based on the regional statistics is introduced to determine the merger of regions in a stochastic manner. Experimental results show that the proposed system achieves overall segmentation error of under 10% for skin lesions in macroscopic images, which is lower than that achieved by existing methods.

This paper presents a new method for high efficiency video coding using an adaptive GOP structure based on video content for the H.264/AVC standard. The available H.264/AVC encoders typically use static GOP sizes that define how the frames I (Intra), P (Predictive) and B (Bi-predictive) are positioned during de coding process. However, by analyzing the video content it is possible to identify the optimum position for each type of frame inside the GOP. The proposed method analyses the video content and finds the best position for inserting I frames in the video sequence. Thus the GOP structure can assume different sizes, depending on the video content. The results for test sequences and real videos show that the proposed method can significantly reduce the required bit rate, comparing to the static GOP sizes, with reduced PSNR losses. The proposed adaptive GOP presents a gain, in terms of bit rate reduction for real movies, of 8.6%, 15%, 24.7% and 40.8% in comparison with static GOP sizes 32, 16, 8 and 4, respectively.

This work describes an approach for object-oriented video segmentation based on motion coherence. Using a tracking process based on adaptively sampled points (namely, particles), 2-D motion patterns are identified with an ensemble clustering approach. Particles are clustered to obtain a pixel-wise segmentation in space and time domains. The segmentation result is mapped to an image spatio-temporal feature space. Thus, the different constituent parts of the scene that move coherently along the video sequence are mapped to volumes in this spatio-temporal space. These volumes make the redundancy in the temporal sense more explicit, leading to potential gains in video coding applications. The proposed solution is robust and more generic than similar approaches for 2-D video segmentation found in the literature. In order to illustrate the potential advantages of using the proposed motion segmentation approach in video coding applications, the PSNR of the temporal predictions and the entropies of prediction errors obtained in our experiments are presented, and compared with other methods. Our experiments with real and synthetic sequences suggest that our method also could be used in other image processing and computer vision tasks, besides video coding, such as video information retrieval and video understanding.

This correspondence describes a new superresolution approach for enhancing the resolution of mammographic images using complex wavelet frequency information. This method allows regions of interest of a mammographic image to be viewed in enhanced resolution while reducing the patient exposure to radiation. The proposed method exploits the structural characteristics of breast tissues being imaged and produces higher resolution mammographic images with sufficient visual fidelity that fine image details can be discriminated more easily. In our approach, the superresolution problem is formulated as a constrained optimization problem using a third-order Markov prior model and adapts the priors based on the phase variations of the low-resolution mammographic images. Experimental results indicate the proposed method is more effective at preserving the visual information when compared with existing resolution enhancement methods.

Several continuous manufacturing processes use stochastic texture images for quality control and monitoring. Large amounts of pictorial data are acquired, providing important information about both the materials produced and the manufacturing processes involved. However, it is often difficult to measure objectively the similarity among industrial stochastic images or to discriminate between texture images of stochastic materials with distinct properties. Nowadays, the degree of discrimination required by industrial processes often goes beyond the limits of human visual perception. This paper proposes to model this specific class of textures as colored noise and presents a new approach for multiresolution stochastic texture representation and discrimination in industry (e.g., nonwoven textiles and paper). The wavelet transform is used to represent stochastic texture images in multiple resolutions and to describe them using local orientation and density variability as features. Based on this representation, a multiresolution distance measure for stochastic textures is proposed, and industrial applications of the method and experimental results are reported. The conclusions include ideas for future work

In this paper, a method for lossless and near lossless compression of large digital mammograms is proposed. This method is based on a predictive coder that uses integer-to-integer operations, and generates a two-layer embedded bit stream (i.e. near-lossless and refinement layer). The maximum tolerated pixel distortion is guaranteed in the near lossless component. Comparisons with other proposed approaches are based on a database of high-resolution 12 bits/pixel digital mammograms. The simulation results indicate that our lossless compression method can offer average bit rates 39, 5%, 24%, 7% and 2.5% better than PNG, JPEG 2000, JPEG-lossless, and LOCO, respectively. Besides, for the same images, the proposed method offers near lossless compression at relatively low computational cost, providing an average bit rate of 2.57 bits/pixel and PSNR of 47.77 dB.

Usually, diagnostic video hysteroscopies are recorded without interruptions (i.e. continuous). However, only a few video segments are relevant from the diagnosis/prognosis point of view, and need to be evaluated and referenced later. This paper proposes a hierarchical technique to identify clinically relevant segments in diagnostic hysteroscopy videos, and their associated key-frames, creating a rich video summary. The proposed summarization approach is adaptive to video contents, and represents the clinically relevant video segments hierarchically to facilitate fast video browsing. The preliminary experimental results indicate that our method produces video summaries that contain no false negatives (i.e. all video segments known to be clinically relevant appear in the summaries). Also, compared to other methods published in the literature, our method produces less false positives.

In this paper, a method for lossless compression of large digital mammograms is proposed. In order to efficiently remove the spatial redundancy from the mammographic images, we use a geometric predictor based on irregular sampling and the Delaunay triangulation. The difference between the predicted and the original (i.e. the error image) is calculated and encoded using JPEG-LS approach. For faster convergence, we build the triangulation using the pixels most significant bits, and code separately the least significant bits using the PNG approach. Comparisons with other proposed approaches are based on a database of high-resolution digital mammograms. The preliminary results indicate that our method can offer average compression ratios 43%, 40%, 45%, and 25% higher than JPEG 2000, JPEG-ES, JPEG-lossless, and PNG, respectively.

We outline the information-theoretic differential geometry of gamma distributions, which contain exponential distributions as a special case, and log-gamma distributions. Our arguments support the opinion that these distributions have a natural role in representing departures from randomness, uniformity, and Gaussian behavior in stochastic processes. We show also how the information geometry provides a surprisingly tractable Riemannian manifold and product spaces thereof, on which may be represented the evolution of a stochastic process, or the comparison of different processes, by means of well-founded maximum likelihood parameter estimation. Our model incorporates possible correlations among parameters. We discuss applications and provide some illustrations from a recent study of amino acid self-clustering in protein sequences; we provide also some results from simulations for multisymbol sequences.

This paper proposes a new method for image denoising with edge preservation, based on image multiresolution decomposition by a redundant wavelet transform. In our approach, edges are implicitly located and preserved in the wavelet domain, whilst image noise is filtered out. At each resolution level, the image edges are estimated by gradient magnitudes (obtained from the wavelet coefficients), which are modeled probabilistically, and a shrinkage function is assembled based on the model obtained. Joint use of space and scale consistency is applied for better preservation of edges. The shrinkage functions are combined to preserve edges that appear simultaneously at several resolutions, and geometric constraints are applied to preserve edges that are not isolated. The proposed technique produces a filtered version of the original image, where homogeneous regions appear separated by well-defined edges. Possible applications include image presegmentation, and image denoising.

This paper discusses an approach for detecting edges in color images. A color image is represented by a vector field, and the color image edges are detected as differences in the local vector statistics. These statistical differences can include local variations in color or spatial image properties. The proposed approach can easily accommodate concepts, such as multiscale edge detection, as well as the latest developments in vector order statistics for color image processing. A distinction between the proposed approach and previous approaches for color edge detection using vector order statistics is that, besides the edge magnitude, the local edge direction is also provided. Note that edge direction information is a relevant feature to a variety of image analysis tasks (e.g., texture analysis).