International Journal of Pattern Recognition and Artificial Intelligence

In this paper the problem of nonlinear feature extraction based on the optimization of the Fisher criterion is analyzed. A new nonlinear feature extraction method is proposed. The method does not make use of numerical algorithms and it has an analytical (closed-form) solution. Moreover, no assumptions on the class probability distribution functions are imposed. The proposed method is applied to some standard pattern recognition problems and compared with other classical methodologies already proposed in the literature. The performance of the proposed method turned out to be superior when compared with the other methods studied.

Improving naive Bayes (simply NB)^15,28 for classification has received significant attention. Related work can be broadly divided into two approaches: eager learning and lazy learning.¹ Different from eager learning, the key idea for extending naive Bayes using lazy learning is to learn an improved naive Bayes for each test instance. In recent years, several lazy extensions of naive Bayes have been proposed. For example, LBR,³⁰ SNNB,²⁷ and LWNB.⁸ All these algorithms aim to improve naive Bayes' classification performance. Indeed, they achieve significant improvement in terms of classification, measured by accuracy. In many real-world data mining applications, however, an accurate ranking is more desirable than an accurate classification. Thus a natural question is whether they also achieve significant improvement in terms of ranking, measured by AUC (the area under the ROC curve).^2,11,17 Responding to this question, we conduct experiments on the 36 UCI data sets¹⁸ selected by Weka¹² to investigate their ranking performance and find that they do not significantly improve the ranking performance of naive Bayes. Aiming at scaling up naive Bayes' ranking performance, we present a novel lazy method ICNB (instance cloned naive Bayes) and develop three ICNB algorithms using different instance cloning strategies. We empirically compare them with naive Bayes. The experimental results show that our algorithms achieve significant improvement in terms of AUC. Our research provides a simple but effective method for the applications where an accurate ranking is desirable.

This work presents a flexible framework for recognizing 3D objects from 2D views. Similarity-based aspect-graph, which contains a set of aspects and prototypes for these aspects, is employed to represent the database of 3D objects. An incremental database construction method that maximizes the similarity of views in the same aspect and minimizes the similarity of prototypes is proposed as the core of the framework to build and update the aspect-graph using 2D views randomly sampled from a viewing sphere. The proposed framework is evaluated on various object recognition problems, including 3D object recognition, human posture recognition and scene recognition. Shape and color features are employed in different applications with the proposed framework and the top three matching rates show the efficiency of the proposed method.

In this paper, we present a walking guidance system for the visually impaired pedestrians. The system has been designed to help the visually impaired by responding intelligently to various situations that can occur in unrestricted natural outdoor environments when walking and finding the destinations. It involves the main functions of people detection, text recognition, face recognition. In addition, added sophisticated functions of walking path guidance using Differential Global Positioning System, obstacle detection using a stereo camera and voice user, interface are included. In order to operate all functions concurrently, we develop approaches in real situations and integrate them. Finally, we experiment on a prototype system under natural environments in order to verify our approaches. The results show that our approaches are applicable to real situations.

We present a shape-based method for automatic people detection and counting without any assumption concerning camera motion. In order to evaluate the robustness of the proposed method, we apply it for classifying athletics videos into two classes: videos of individual and videos of team sports. The videos used are real and characterized by dynamic and unconstrained environment. Moreover, in the case of team sport, we propose a shape deformations based method for running/hurdling discrimination (activity recognition). Robust, adaptive and independent from color, illumination changes and the camera motion, the proposed features are combined in the Transferable Belief Model (TBM) framework providing a two-level (frames and shot) video categorization. Experimental results of 97% of accuracy for individual/team sport categorization using a dataset of 252 real videos of athletic meetings, acquired by moving cameras under varying view angles, indicate the stability and the good performance of the proposed scheme.

This paper proposes a method for hiding an important image in a cover image whose size is limited. In this method, in order to save space, a modified search-order coding (MSOC) technique first transforms the important image, then, a randomization procedure permutes the transformed image to further increase the security. Finally, a modulus function embeds the permuted code in a cover image; notably, in the modules function, the modulus base used for a pixel is determined according to the variance of its neighboring pixels. Experimental results show that the images are of high quality. Comparisons with reported methods are provided.

In this paper, a framework for detection of human skin in digital images is proposed. This framework is composed of a training phase and a detection phase. A skin class model is learned during the training phase by processing several training images in a hybrid and incremental fuzzy learning scheme. This scheme combines unsupervised- and supervised-learning: unsupervised, by fuzzy clustering, to obtain clusters of color groups from training images; and supervised to select groups that represent skin color. At the end of the training phase, aggregation operators are used to provide combinations of selected groups into a skin model. In the detection phase, the learned skin model is used to detect human skin in an efficient way. Experimental results show robust and accurate human skin detection performed by the proposed framework.

Polygonal approximation of digitized curves is an important step in shape representation and analysis. One of the approaches to obtain a good polygonal approximation is that of formulating the approximation task as an optimization problem and then solving it using the available tools. Adopting this strategy, in this article, we present a differential evolution-based method for polygonal approximation. We show that our method not only yields near-optimal results, but also does so very efficiently. The proposed method is also a demonstration of the successful application of differential evolution for solving a combinatorial optimization problem.