Catálogo de publicaciones - libros

First, a modified Neighborhood-Based Clustering (MNBC) algorithm using the directed tree for data clustering is presented. It represents a dataset as some directed trees corresponding to meaningful clusters. Governed by Neighborhood-based Density Factor (NDF), it also can discover clusters of arbitrary shape and different densities like NBC. Moreover, it greatly simplify NBC. However, a failure applying in image segmentation is due to an unsuitable use of Euclidean distance between image pixels. Second, Gray NDF (GNDF) is introduced to make MNBC suitable for image segmentation. The dataset to be segmented is all grays and thus MNBC has the constant computational complexity O(256). The experiments on synthetic datasets and real-world images shows that MNBC outperforms some existing graph-theoretical approaches in terms of computation time as well as segmentation effect.

In this paper, we present an efficient unsupervised mixture model image segmentation method. The idea of this method is that individual image region classes are modeled as mixtures of fuzzy subclasses of mixture distributions, and classification is performed based on the Expectation-Maximization algorithm. To overcome the difficulty of classical mixture model method for noisy image segmentation, spatial contextual information should be taken into account. In particular, the proposed approach based on Markov Random Field was shown to provide more accurate classification of images than traditional Expectation-Maximization algorithm and traditional Markov Random Field image segmentation techniques. The effectiveness of the proposed method is illustrated with synthetic and real images data. The experiments results have shown that the proposed method can achieve more robust segmentation for noisy images.

The polarimetric synthetic aperture radar (PSAR) images are modeled by a mixture model that results from the product of two independent models, one characterizes the target response and the other characterizes the speckle phenomenon. For the scene interpretation, it is desirable to separate between the target response and the speckle. For this purpose, we proposed a new speckle reduction approach using independent component analysis (ICA) based on statistical formulation of PSAR image. In addition, we apply four ICA algorithms on real PSAR images and compare their performances. The comparison reveals characteristic differences between the studied neural ICA algorithms, complementing the results obtained earlier.

Independent component analysis (ICA) has shown success in the separation of sources in lots of applications. However, in synthenic aperture radar (SAR) images the noise is multiplicative, so the applicability of ICA is seriously reduced. This paper proposes a new robust independent component analysis neural network (RICANN) that improves the robustness of ICA by adding outlier rejection rule. Its application in synthetic aperture radar (SAR) is discussed. The results show the potential usage in SAR image processing problems.

Kernel fisher discriminant analysis (KFDA) has received extensive study in recent years as a dimensionality reduction technique. KFDA always encounters an intrinsic singularity of scatter matrices in the feature space, namely ‘small sample size’ (SSS) problem. Several novel methods have been proposed to cope with this problem. In this paper, kernel uncorrelated discriminant analysis (KUDA) is proposed, which not only can bear on the SSS problem but also extract uncorrelated features, a desirable property for many applications. And then, we have conducted a comparative study on the application of KUDA and other variants of KFDA in radar target recognition problem. The experimental results indicate the effectiveness of KUDA and illustrate the utility of KFDA on the problem.

There are increasing demands for high-resolution (HR) images in various applications. Image superresolution (SR) reconstruction refers to methods that increase image spatial resolution by fusing information from either a sequence of temporal adjacent images or multi-source images from different sensors. In the paper we propose a hybrid Bayesian method for image reconstruction, which firstly estimates the unknown point spread function(PSF) and an approximation for the original ideal image, and then sets up the HMRF image prior model and assesses its tuning parameter using maximum likelihood estimator, finally computes the regularized solution automatically. Hybrid Bayesian estimates computed on actual satellite images and video sequence show dramatic visual and quantitative improvements in comparison with the bilinear interpolation result, the projection onto convex sets (POCS) estimate and Maximum A Posteriori (MAP) estimate.

A retrieval-aware image format (rim format) is developed for the usage in the similar-image retrieval. The format is based on PCA and ICA which can compress source images with an equivalent or often better rate-distortion than JPEG. Besides the data compression, the learned PCA/ICA bases are utilized in the similar-image retrieval since they reflect each source image’s local patterns. Following the format presentation, an image search viewer for network environments (Wisvi; Waseda image search viewer) is presented. Therein, each query is an image per se. The Wisvi system based on the “rim” method successfully finds similar-images from non-uniform network environments. Experiments support that the PCA/ICA methods are viable to the joint compression and retrieval of digital images. Interested test users can download a -version of the tool for the joint image compression and retrieval from a web site specified in this paper.

25% of the total revenue earning is achieved from Textile exports for some countries like Bangladesh. It is thus important to produce defect free high quality garment products. Inspection processes done on fabric industries are mostly manual hence time consuming. To reduce error on identifying fabric defects requires automotive and accurate inspection process. Considering this lacking, this research implements a Textile Defect detector. A multi-layer neural network is determined that best classifies the specific problems. To feed neural network the digital fabric images taken by a digital camera and converts the RGB images are first converted into binary images by restoration process and local threshold techniques, then three different features are determined for the actual input to the neural network, which are the area of the defects, number of the objects in a image and finally the shape factor. The develop system is able to identify two very commonly defects such as Holes and Scratches and other types of minor defects. The developed system is very suitable for Least Developed Countries, identifies the fabric defects within economical cost and produces less error prone inspection system in real time.

A modeling and classification model for MPEG video traffic data using a Fuzzy C-Means algorithm with a Divergence-based Kernel (FCMDK) for clustering GPDF data is proposed in this paper. The FCMDK is based on the Fuzzy C-Means clustering algorithm and thus exploits advantageous features of fuzzy clustering techniques. To further improve classification accuracies and deal with nonlinear data, the input data is projected into a feature space of a higher dimensionality. Consequently, nonlinear problems existing in the input space can be solved linearly in the feature space. The divergence-based kernel method adopted in the FCMDK employs a divergence measure between two probability distributions for its similarity measure. By adopting the divergence-based kernel method for probability data, the FCMDK can not only utilize advantageous features of the kernel method but also exploit the statistical nature of the input data. Experiments and results on several MPEG video traffic data sets demonstrate that the classification model employing the FCMDK for clustering GPDF data can archive improvements of 28.19% and 34.60% in terms of False Alarm Rate (FAR) over the models using the conventional k-means and SOM algorithms, respectively.

Replays are key cues for events detection in sport videos since they are the immediate consequence of highlights or important events happened in sports. In many sports videos, replays are usually sandwiched with two identical logo transitions, prompt the beginning and end of a replay. A logo transition is a kind of special digital video effects, usually contains 12-35 consecutive frames, describe a flying or variable object. In this paper, a novel automatic logo detection approach is proposed. It contains two main stages: a logo transition template is automatically learned by dynamic programming and unsupervised clustering, a key frame is also extracted; then the extracted key frame and the learned logo template are used jointly to detect logos in sports videos. The optical flow features are used to depict the motion characteristics of the logo transitions. Experiments on different types of sports videos show that the proposed approach can reliably detect logos in sports videos efficiently.