This paper proposes a concept and design strategy for the humanoid intelligent management system (HIMS) based on artificial life. Various topics are discussed including the design method and implementation techniques ...
详细信息
This paper proposes a concept and design strategy for the humanoid intelligent management system (HIMS) based on artificial life. Various topics are discussed including the design method and implementation techniques for the dual management scheme (DMS), humanoid intelligent management model (HIMM), central-decentralized management pattern, and multi-grade coordination function.
This paper defines and proves a new closeness degree firstly, and then presents an improved-CFART neural network model, after introducing the closeness degree into the standard fuzzy ART model. This paper also develop...
详细信息
This paper defines and proves a new closeness degree firstly, and then presents an improved-CFART neural network model, after introducing the closeness degree into the standard fuzzy ART model. This paper also develops an information analysis and simulation system based on the recognition of industrial real-time data from field. The results of simulation experiments demonstrate that the CFART model has excellent capability of pattern recognition for dynamic waveform.
In this paper, digital image processing and computer vision techniques are applied to study tongue images for feature extraction with VC++ and Matlab. Extraction and analysis of the tongue surface features are based o...
详细信息
In this paper, digital image processing and computer vision techniques are applied to study tongue images for feature extraction with VC++ and Matlab. Extraction and analysis of the tongue surface features are based on shape, color, edge, and texture. The developed software has various functions and good user interface and is easy to use. Feature data for tongue image pattern recognition is provided, which form a sound basis for the future tongue image recognition.
This paper describes an application of a reusable mobile agent system in network management. A mobile agent reusable system is constructed to realize a new method in forming mobile agent systems. By using this method,...
详细信息
This paper describes an application of a reusable mobile agent system in network management. A mobile agent reusable system is constructed to realize a new method in forming mobile agent systems. By using this method, an agent can change its route dynamically without making any change to its specific behavior. By classifying mobile agents into two categories, the task agent can be reusable in different networks. In this way, a mobile agent system can easily carry out network management tasks.
A color image enhancement approach based on maximum fuzzy entropy and genetic algorithm is proposed in this paper. It enhances color images by stretching the contrast of S and / components respectively in the HSI colo...
详细信息
A color image enhancement approach based on maximum fuzzy entropy and genetic algorithm is proposed in this paper. It enhances color images by stretching the contrast of S and / components respectively in the HSI color representation. The image is transformed from the property domain to the fuzzy domain with S-function. To preserve as much information as possible in the fuzzy the domain, the fuzzy entropy function is used as objective function in a genetic algorithm to optimize three parameters of the S-function. The Sigmoid function is applied to intensify the membership values and the results are transformed back to the property domain to produce the enhanced image. Experiments show the effectiveness of the approach.
The origin and characteristics of computational intelligence, and several typical computational intelligence algorithms such as genetic algorithm and DNA computing are described, and the influence of evolution strateg...
详细信息
The origin and characteristics of computational intelligence, and several typical computational intelligence algorithms such as genetic algorithm and DNA computing are described, and the influence of evolution strategies and convergence properties on the encoding mechanism is discussed. A novel genetic algorithm based on degressive carry number encoding is then proposed. This algorithm uses degressive carry number encoding in the evolutionary process instead of commonly used fixed carry number. Finally a novel encoding mechanism and a new algorithm are proposed, which combine modern computational intelligence with the traditional Chinese methodology.
Implantable rotary blood pump (IRBP) has been promoted to the stage of clinical trial. This paper introduces a unique IRBP without a-shaft. Instead of using thrombogenic pivots or power-drawing magnetic suspension, im...
详细信息
Implantable rotary blood pump (IRBP) has been promoted to the stage of clinical trial. This paper introduces a unique IRBP without a-shaft. Instead of using thrombogenic pivots or power-drawing magnetic suspension, impeller is supported hydrodynamically when rotating, by lubrication flows in the thin spaces between itself and the pump body. To this end, the flow is very difficult to be measured using usual laboratory equipments. Therefore, computational fluid dynamics (CFD) has been applied as an important tool in the IRBP design and its validation procedure. Several CFD results such as pump performance improvement, unsteady hydraulic dynamic analysis, biocapability prediction, validation and verification (V&V), and flow visualization have been performed.
Structures of natural plants are complex and difficult to model. This paper proposes a fast visual modeling for plants based on a small set of images, and establishes a reasonable plant model. Based on knowledge about...
详细信息
Structures of natural plants are complex and difficult to model. This paper proposes a fast visual modeling for plants based on a small set of images, and establishes a reasonable plant model. Based on knowledge about growth patterns of the plant, image segmentation and 3D reconstruction are first performed to construct the plant skeleton (trunk and major branches), from which the remainder of the plant grows. Then the system produces the realistic plant model images based on image synthesis and validation. It is unnecessary to acquire the complex structure (such as the complex production rules of L-systems). The method provides a high degree of control over the final shape by image validation, resulting in realistic reconstruction.
A method for simulation of cutting virtual soft tissue objects made of tetrahedron elements is developed. A linear isotropic elastic model is used for the soft tissue material properties and a tensor-mass model chosen...
详细信息
A method for simulation of cutting virtual soft tissue objects made of tetrahedron elements is developed. A linear isotropic elastic model is used for the soft tissue material properties and a tensor-mass model chosen for the physical deformation. The Verlet leapfrog method is used to perform time integration in solving the dynamic equations. Cutting is simulated by simply removing the tetrahedron elements that are intersected with the virtual scalpel. By making use of the spatial coherence, collision detection between soft tissue objects and the virtual scalpel is sped up. To facilitate the simulation, the soft tissue object is represented by linked lists of vertices, edges and tetra elements with pointers to the related neighboring features. The established software framework can serve as a base for the future development. Results of virtual experiments are shown and discussed. Possible future directions are also given.
Automatic recognition of skin micro-image symptom is important in skin diagnosis and treatment. Feature selection is to improve the classification performance of skin micro-image symptom. This paper proposes a hybrid ...
详细信息
Automatic recognition of skin micro-image symptom is important in skin diagnosis and treatment. Feature selection is to improve the classification performance of skin micro-image symptom. This paper proposes a hybrid approach based on the support vector machine (SVM) technique and genetic algorithm (GA) to select an optimum feature subset from the feature group extracted from the skin micro-images. An adaptive GA is introduced for maintaining the convergence rate. With the proposed method, the average cross validation accuracy is increased from 88.25% using all features to 96.92 % using only selected features provided by a classifier for classification of 5 classes of skin symptoms. The experimental results are satisfactory.
暂无评论