International Journal of Information Acquisition

This paper presents a fast palmprint verification system based on fractal coding. In the stage of registration, a sub-image from user's training palmprints is intentionally extracted and stored as his or her template. In the stage of verification, the step of region of interest extraction is not needed, the sample image is directly matched with the template based on fractal coding, which can reduce the whole response time. Whether the sample image and the template are from the same person or not is decided by their matching scores. Experimental evaluation results on two databases clearly demonstrate the effectiveness of the proposed approach.

A novel variational multiphase level set mathematical model is derived for image segmentation with two contributions. By virtue of eliminating the time-consuming re-initialization procedure and neglecting the property of the level set function during the evolution process, we in this paper present two strategies that may be taken as our contributions to solving these problems. Two scenarios are considered, namely, first, the distance regularization term which is defined by double-well potential function with two minimum points is introduced to our mathematical model for avoiding the re-initialization process. Second, by combining a Tikhonov-like regularization term which can guarantee the smoothness for the evolution curve over the previous method. Numerical simulation studies are presented to verify our new model via evaluating and comparing with existing algorithms.

Breast cancer is the leading cause of death in women. Early detection and early treatment can significantly reduce the breast cancer mortality. Texture features are widely used in classification problems, i.e., mainly for diagnostic purposes where the region of interest is delineated manually. It has not yet been considered for sonoelastographic segmentation. This paper proposes a method of segmenting the sonoelastographic breast images with optimum number of features from 32 features extracted from three different extraction methods: Gray Level Co-occurrence Matrix (GLCM), Local Binary Pattern (LBP), and Edge-Based Features. The image undergoes preprocessing by Sticks filter that improves the contrast and enhances the edges and emphasizes the tumor boundary. The features are extracted and then ranked according to the Sequential Forward Floating Selection (SFFS). The optimum number of ranked features is used for segmentation using k-means clustering. The segmented images are subjected to morphological processing that marks the tumor boundary. The overall accuracy is studied to investigate the effect of automated segmentation where the subset of first 10 ranked features provides an accuracy of 79%. The combined metric of overlap, over- and under-segmentation is 90%. The proposed work can also be considered for diagnostic purposes, along with the sonographic breast images.

In this paper, a multi-sensor condition monitoring scheme is developed to diagnose machine faults in the presence of sensor failure. The signals from the monitored machine are decomposed using the wavelet packet transform (WPT). Two feature reduction schemes, using genetic algorithms are developed for feature selection in condition monitoring. One scheme assumes no prior knowledge about system costs or failure characteristics, and the other scheme aims to minimize the operating costs over a period of time. Two classifiers, radial basis function networks and support vector machines, are developed and compared in their ability to classify machine faults under conditions of sensor failure. The developed methodology is implemented in an experimental system, an industrial fish processing machine. The machine is instrumented with multiple accelerometers and microphones to continuously acquire signals of machine vibration and sound. The performance of the implemented fault diagnosis methodology is evaluated though experimentation.

Recently, applying control theory to regulate the intracellular mRNA level was introduced as a new direction for gene regulation. However, the high nonlinearity in the gene regulatory networks imposes significant challenges in control design. As a well understood benchmark example, the GAL regulatory network in S. cerevisiae was recently proposed as a test-bed system for validating theoretical control algorithms in cellular systems. A simple proportional feedback control approach was previously proposed for regulating the intracellular mRNA level in the GAL network, however, there were still limitations with its use to control the nonlinear GAL network. To improve the performance and effectiveness, this paper proposes an advanced nonlinear control strategy. The reduced mathematical model for the GAL network is reorganized into a nonlinear affine system. Then, a partial feedback linearization control approach was employed to regulate the concentration of a protein at a desired level. For validating the control approach in experimental studies, we choose Gal1p as a measurable output, instead of GAL1 mRNA used in the previous study. Simulation results demonstrate that this control approach can shorten the convergence time between states comparing with the proportional feedback control.

Particle swarm optimization (PSO), a prevalent optimization algorithm, has been successfully applied to various fields of science and engineering. However, PSO still suffers from some problems such as premature convergence. To solve these problems, we propose a mutation PSO (MPSO) in this paper. Compared with the traditional PSO, there are two main improvements of the proposed MPSO. First, a new particle update rule is explored. The new rule updates a particle's position according to not only its best known position and the global best known position of the swarm, but also a number of other particles' best known positions. The second improvement is that a mutation operator is employed. Mutation operator is used to avoid premature convergence. The MPSO is utilized to train a multilayer perceptron (MLP). The MLP trained by MPSO is finally applied to two classification problems: Iris flower classification and scene classification. For comparison purposes, traditional PSO, genetic algorithm (GA), and back-propagation (BP) are also investigated. Experimental results demonstrate the superior performance of the proposed MPSO for MLP training.

Particle swarm optimization (PSO) is one of the important evolutionary algorithms. However, the traditional PSO suffers from the premature convergence problem. In view of this, a new PSO, named mutation PSO (MPSO), is proposed in this paper. The proposed MPSO not only makes use of a mutation operator to update particles/individuals, which was originally designed for genetic algorithm (GA). But also a new weighted update rule is proposed for MPSO to produce the new swarm. Then we use the proposed MPSO to train multilayer perceptron (MLP) with two tasks: curve fitting and classification. In particular, the performance investigation is concentrated on scene classification. For a comparison purpose, MLPs trained using the error backpropagation (BP), traditional PSO and GA are also investigated. The advantages and disadvantages of these algorithms are also analyzed. Experimental results show that the proposed MPSO outperforms than other algorithms for the training of an MLP.

Adaptive robust synchronization method is proposed for chaos synchronization of Lurie systems with time-varying delay. To weaken the restrictions on the change rate of time-varying delay in synchronization problems, adaptive estimations related to the derivative's upper bound of the time-varying delay are designed, so this upper bound can be unknown. The Lipschitz constant of the nonlinear link is also adaptively estimated, rather than being pre-calculated. Whole design is in the framework of robust control, and the synchronization can endure disturbance effectively. In addition, the synchronization of chaotic Lurie system with time-varying delay is realized as an example.

In many fields of science, IT applications and business environments successfully evolved systems to receive vast amount of electronic data and information. Due to increasing electronic data and information, most recent researches have tried to find a solution to resolve the crisis of information overload. These solutions include a combination of techniques of data mining, machine learning, natural language processing and information retrieval, information extraction, and knowledge management. A great challenge is how to exploit those information and knowledge resources and turn them into useful knowledge available to concerned people. The value of knowledge increases when people can share and capitalize on it. Thus, approaches that can help researchers to benefit from existing hidden knowledge are needed. For this, tools that can analyze, extract and explore relevant and useful information with relations are required. So, the main contribution of this paper is to integrate the technology of XML with text analysis for introducing an efficient concept-based structure model, where this model can represent the text in a form that can be easily understood, shared, managed and mined. This paper describes an efficient object oriented text analysis (OOTA) approach by generating an object oriented model that transforms unstructured text to a specific structured form and stored in XML format. The experimental results show that this approach has a good promotion on results.

With the increasing role of computing devices, facilitating natural human computer interaction (HCI) will have a positive impact on their usage and acceptance as a whole. For long time, research on HCI has been restricted to techniques based on the use of keyboard, mouse, etc. Recently, this paradigm has changed. Techniques such as vision, sound, speech recognition allow for much richer form of interaction between the user and machine. The emphasis is to provide a natural form of interface for interaction. Gestures are one of the natural forms of interaction between humans. As gesture commands are found to be natural for humans, the development of gesture control systems for controlling devices have become a popular research topic in recent years. Researchers have proposed different gesture recognition systems which act as an interface for controlling the applications. One of the drawbacks of present gesture recognition systems is application dependence which makes it difficult to transfer one gesture control interface into different applications. This paper focuses on designing a vision-based hand gesture recognition system which is adaptive to different applications thus making the gesture recognition systems to be application adaptive. The designed system comprises different processing steps like detection, segmentation, tracking, recognition, etc. For making the system as application-adaptive, different quantitative and qualitative parameters have been taken into consideration. The quantitative parameters include gesture recognition rate, features extracted and root mean square error of the system while the qualitative parameters include intuitiveness, accuracy, stress/comfort, computational efficiency, user's tolerance, and real-time performance related to the proposed system. These parameters have a vital impact on the performance of the proposed application adaptive hand gesture recognition system.

A new watermarking approach based on affine Legendre moment invariants (ALMIs) and local characteristic regions (LCRs) which allows watermark detection and extraction under affine transformation attacks is presented in this paper. It is a non-blind watermarking scheme. Original image color image is converted into HSV color space and divided into four parts. LCR is constructed and a set of affine invariants are derived on LCRs based on Legendre moments for each part. These invariants can be used for estimating the affine transform coefficients on the LCRs. ALMIs are used for watermark embedding, detection and extraction as they provide synchronization and invariant feature which is necessary for a robust watermarking scheme. The proposed scheme shows resistance to geometric distortion, cropping, filtering, compression, and additive noise than the existing ALMI based scheme [Alghoniemy, M. and Tewfik, A. H. [2004] "Geometric invariance in image watermarking," IEEE Trans. Image Process13(2), 145–153] and affine geometric moment invariant (AGMI) based scheme [Seo, J. S. and Yoo, C. D. [2006] "Image watermarking based on invariant regions of scale-space representation," IEEE Trans. Signal Process. 54(4), 1537–1549].