Catálogo de publicaciones - libros

In this paper, we implement hand gesture recognition system using union of fuzzy algorithm and Relational Database Management System (hereafter, RDBMS) module for Post PC (the embedded-ubiquitous environment using blue-tooth module, embedded i.MX21 board and note-book computer for smart gate). The learning and recognition model due to the RDBMS is used with input variable of fuzzy algorithm (fuzzy max-min composition), and recognize user’s dynamic gesture through efficient and rational fuzzy reasoning process. The proposed gesture recognition interface consists of three modules: 1) gesture input module that processes motion of dynamic hand to input data, 2) RDBMS module to segment significant gestures from inputted data, and 3) fuzzy max-min recognition module to recognize significant gesture of continuous, dynamic gestures and extensity of recognition. Experimental result shows the average recognition rate of 98.2% for significant dynamic gestures.

Behavioral sequences of the medaka () were continuously investigated through an automatic image recognition system in increasing temperature from 25°C to 35°C. The observation of behavior through the movement tracking program showed many patterns of the medaka. After much observation, behavioral patterns could be divided into basically 4 patterns: active- smooth, active-shaking, inactive-smooth, and inactive-shaking. The “smooth” and “shaking” patterns were shown as normal movement behavior, while the “smooth” pattern was more frequently observed in increasing temperature (35° C) than the “shaking” pattern. Each pattern was classified using a devised decision tree after the feature choice. It provides a natural way to incorporate prior knowledge from human experts in fish behavior and contains the information in a logical expression tree. The main focus of this study was to determine whether the decision tree could be useful in interpreting and classifying behavior patterns of the medaka.

The problem of computing the minimal enclosing sphere (MES) of a set of points in the high dimensional kernel-induced feature space is considered. In this paper we develop an entropy-based algorithm that is suitable for any Mercer kernel mapping. The proposed algorithm is based on maximum entropy principle and it is very simple to implement. The convergence of the novel algorithm is analyzed and the validity of this algorithm is confirmed by preliminary numerical results.

Handwritten character recognition has been an intensive research for last decade. A handwritten character recognition fuzzy system with an automatically generated rule base possesses the features of flexibility, efficiency and online adaptability. A major requirement of such a fuzzy system for either online or offline handwritten character recognition is, the segmentation of individual characters into meaningful segments. Then these segments can be used for the calculation of fuzzy features and the recognition process. This paper describes a new segmentation algorithm for offline handwritten character segmentation, which segments the individual handwritten character skeletons into meaningful segments. Therefore, this algorithm is a good candidate for an offline handwritten character recognition fuzzy system.

As one key step of the automatic fingerprint identification system (AFIS), fingerprint image segmentation can decrease the affection of the noises in the background region and handing time of the subsequence algorithms and improve the performance of the AFIS. Markov Chain Monte Carlo (MCMC) method has been applied to medicine image segmentation for decade years. This paper introduces the MCMC method into fingerprint image segmentation and brings forward the fingerprint image segmentation algorithm based on MCMC. Firstly, it generates a random sequence of closed curves as Markov Chain, which is regarded as the boundary between the fingerprint image region and the background image region and uses the boundary curve probability density function (BCPDF) as the index of convergence. Then, it is simulated by Monte Carlo method with BCPDF as parameter, which is converged to the maximum. Lastly, the closed curve whose BCPDF value is maximal is regarded as the ideal boundary curve. The experimental results indicate that the method is robust to the low-quality finger images.

Phase field model has been well acknowledged as an important method for image segmentation. This paper discussed the problem of jacquard image segmentation by approaching the phase field paradigm from a numerical approximation perspective. For fuzzy theory provides flexible and efficient techniques for dealing with conflicting optimization probelms, a novel fuzzy optimization algorithm for numerical solving of the model was proposed. To achieve global minimum of the model, a multi-start fuzzy strategy which combined a local minimization procedure with genetic algorithm was enforced. As the local minimization procedure does not guarantee optimality of search process, several random starting points need to be generated and used as input into global search process. In order to construct powerful search procedure by guidance of global exploration, genetic algorithm was applied to scatter the set of quasi-local mimizers into global positions. Experimental results show that the proposed algorithm is feasible, and reaches obvious effects in terms of jacquard image segmentation.

The adjunction in lattice theory is an important technique in lattice-based mathematical morphology and fuzzy logical operators are indispensable implements in fuzzy morphology. This paper introduces a set-valued mapping that is compatible with the infimum in a complete lattice and with a conjunction in fuzzy logic. According to the generalized operator, a concept of a fuzzy adjunction is developed to generate fuzzy morphological dilation and erosion. Fundamental properties of the generalized fuzzy morphological operators have been investigated.

The matching similarity measures that can be used in image target recognition are surveyed and a novel similarity measure is proposed in this paper. Two basic factors that affect the image matching performance and the merits and faults of two common types of image matching algorithm are firstly analyzed. Then, based on the systematic study of similarity measures, image matching projection similarity measure is defined by simplify the classical normalized correlation measure. An example of the application of the proposed matching similarity measure in image target recognition and position is given at the end of this paper; the experimental results show its feasibility and effectivity.

The paper uses Bark wavelet filter instead of the FIR filter as front-end processor of speech recognition system. Bark wavelet divides frequency band based on critical band and its bandwidths are equal in Bark domain. By selecting suitable parameters, Bark wavelet can overcome the disadvantage of dyadic wavelet and M-band wavelet dividing frequency band based on octave. The paper gave the concept and parameter setting method of Bark wavelet. For signals that are filtered by Bark wavelet, ZCPA features with noise-robust are extracted and used in speech recognition. And recognition network uses HMM. The results show the recognition rates of the system in noise environments are improved.

In this paper, we proposed an accurate and fast iris location method based on the features of human eyes. Firstly, according to the gray features of pupil, find a point inside the pupil using a gray value summing operator. Next, starting from this point, find three points on the iris inner boundary using a boundary detection template designed by ourselves, and then calculate the circle parameters of iris inner boundary according to the principle that three points which are not on the same line can define a circle. Finally, find other three points on the iris outer boundary utilizing the similar method and obtain the circle parameters. A large number of experiments on the CASIA iris image database demonstrated that the location results of proposed method are more accurate than any other classical methods, such as Daugman’s algorithm and Hough transforms, and the location speed is very fast.