Search

Sign language identification and recognition technique is composed by the gesture images detection and the hand gestures recognition. Gesture images detection is to locate the image part of palm and fingers from the captured pictures, and rotating them to the appropriate gesture posture. Both of them are the important pre-processing for sign language identification and recognition. Lose them, the correctness rate of the sign language recognition algorithms will be dropped down to an unacceptable level. This paper presents novel processing algorithms for the gesture images detection and recognition. In the detection process, it rotates an askew gesture to right position, and to delete the elbow and forearm parts from the captured pictures. In the recognition process, it includes two phases with the model construction and the sign language identification. In the model construction phase, the static hand gesture of sign language is constructed by the Gaussian mixture model, and the unknown gesture image is identified by Gaussian model match. Based on this presented static sign language detection and recognition algorithms, the correct recognition rate is about 94% in average.

In this paper, a pipelining empirical mode decomposition is presented to reduce the computing time of the emotionalized spontaneous speaker or speech recognition processing. This is a novel approach for integrating the pipelining technique into the standard empirical mode decomposition of the Hilbert-Huang transform. In addition, there is reduced about 45% of the computing time when the emotionalized spoken signal through our segmentation and pipelining processes. Based on the designed processing of emotionalized spontaneous speaker or speech recognition, the segmented and processed voice signals are recomposed back for constructing the speech and speaker models, or to identify which existed model is the most similar one. In the final part of this paper, a comparison of the speech recognized rate between standard and pipelining empirical mode decompositions are presented, and an equivalent effect in the recognition will be found. In practice, speaker or speech recognitions in an emotionalized spontaneous speech are very difficult. The existing speech recognition methods often fail to capture inherent voiceprint features from an emotionalized speech, such as the voice with a passionate intonation. And some of the existed methods to extract the pure voiceprint from an emotionalized spoken signal are very expensive in computation and time, so that technique is impossible to use in a real-time environment like smart houses. But, this paper presents a solution to improve the emotionalized spontaneous speaker or speech recognition processing to fit the real-time request.

In this paper, a behavior model of home cared aging family is presented and used by our designed intelligent pervasive-care system. This system uses multiple interlaced and non-visual sensors to recognize the behavior of concerned family in independent and concurrent. It also provides a competition mechanism between those heterogeneous sensing agents for verifying the recognized behavior. Those independent agents process and classify sensed single to some known symptoms. Based on those known symptoms, the reasoning mechanism in our behavior model can infer the occurring events, actions, and purpose of the cared aging family. So the behavior model empowers the intelligent pervasive-care system to monitor the home activities of cared aging family, and assist them to inform suitable peoples when some accidents occurrence.

This paper proposes a qualitative fuzzy modeling technique. The technique takes existing mathematical models of a nuclear power plant to develop causal fuzzy networks. It partitions the value spaces of the observable variables into proper fuzzy spaces according to their physical meaning, and propagates them through the causal fuzzy networks to automatically partition the value spaces of the unobservable variables. This technique allows us to use causal fuzzy networks to model the dynamic behavior of a nuclear power plant. Since the networks are constructed from existing mathematical equations, they can keep detailed causal relationships between variables. The fuzzy space propagation method features the automatic partitioning of the fuzzy spaces of all unobservable variables in the causal fuzzy networks. It also transforms the meaning of the physical partitions of relevant observable variables to dependent unobservable variables. The causal fuzzy networks are used as a qualitative fuzzy model for simulating the dynamic behavior of a nuclear power plant by an on-line simulator.

HASLEARN is a highly intelligent autonomous system that contains multistrategy learning capabilities which are integrated by a four-stages incremental learning processes. So HASLEARN can work like a human operator to learn operationalized reactive action rules, to recover and eliminate any accidents or anomalisms, and to learn prediction rules to warn the development of potential, abnormal states based on the current states when it is applied in a nuclear power plant. This paper describes the embedded learning capabilities in the planner component of HASLEARN. >

Euclid's Greatest Common Divisor (GCD) algorithm is an efficient approach for calculating multiplicative inversions. It relies mainly on a fast modular arithmetic algorithm to run quickly. A traditional modular arithmetic algorithm based on nonrestoring division needs a magnitude comparison for each iteration of shift-and-subtract operation. This process is time consuming, since it requires magnitude comparisons for every computation iteration step. To eradicate this problem, this study develops a new fast Euclidean GCD algorithm without magnitude comparisons. The proposed modular algorithm has an execution time that is about 33% shorter than the conventional modular algorithm.

A low-complexity array multiplier for GF(2m) fields with an irreducible trinomial Xm+Xn+1 is presented. The space complexity of the proposed multiplier is reduced from order O(m2) to O(m) compared with the Lee's array multiplier. The time complexity of the proposed multiplier is about half that of Lee's array multiplier.