this article presents the evaluation analysis of the radial function neural network embedded on an FPGA chip by experiments. the back-propagation algorithm has been embedded and tested for the feasibility for on-line ...
this article presents the evaluation analysis of the radial function neural network embedded on an FPGA chip by experiments. the back-propagation algorithm has been embedded and tested for the feasibility for on-line learning tasks. the nonlinear pattern classification task of the XOR logic has been conducted by the designed hardware. Performances are evaluated extensively by different orders of the Taylor-Maclaurin series expansion for approximating nonlinear functions and compared with results by the MATLAB program. the effects on the performance by the nonlinear function approximation have been analyzed by experimental studies of the XOR classification task.
In this paper, we design linear time algorithms to recognize and determine topological invariants such as genus and homology groups in 3D. these invariants can be used to identify patterns in 3D image recognition and ...
详细信息
ISBN:
(纸本)9781424421749
In this paper, we design linear time algorithms to recognize and determine topological invariants such as genus and homology groups in 3D. these invariants can be used to identify patterns in 3D image recognition and medical image analysis. Our method is based on cubical images with direct adjacency, also called (6,26)-connectivity images in discrete geometry. According to the fact that there are only six types of local surface points in 3D and a discrete version of the well-known Gauss-Bonnett theorem in differential geometry, we first determine the genus of a closed 2D-connected component (a closed digital surface). then, we use the Alexander duality to obtain the homology groups of a 3D object in 3D space. this idea can be extended to general simplicial decomposed manifolds or cell complexes in 3D.
this paper describes a rule-based approach for equivalence checking of reactive systems. the approach is based on new types of dependence and flow graphs that are more appropriate for reactive languages than tradition...
详细信息
this paper describes a rule-based approach for equivalence checking of reactive systems. the approach is based on new types of dependence and flow graphs that are more appropriate for reactive languages than traditional notions intended for transformational languages. Equivalence rules utilizing this static dependence and flow information are derived from language semantics. the rules are then applied in a bottom-up fashion, corresponding to the structures of the programs being checked, until equivalence is shown. A prototype toolset has been implemented, and results indicate speedups of several orders of magnitude over more traditional equivalence checkers. the paper describes our approach and tools, and also outlines how our methods can be used in a general equivalence checking system.
the proceedings contain 37 papers. the topics discussed include: bipartite graph matching for computing the edit distance of graphs;matching of tree structures for registration of medical images;graph-based methods fo...
详细信息
ISBN:
(纸本)9783540729020
the proceedings contain 37 papers. the topics discussed include: bipartite graph matching for computing the edit distance of graphs;matching of tree structures for registration of medical images;graph-based methods for retinal mosaicing and vascular characterization;graph based shapes representation and recognition;a continuous-based approach for partial clique enumeration;a bound for non-subgraph isomorphism;a correspondence measure for graph matching using the discrete quantum walk;a quadratic programming approach to the graph edit distance problem;image classification using marginalized kernels for graphs;comparing sets of 3D digital shapes through topological structures;hierarchy construction schemes within the scale set framework;local reasoning in fuzzy attribute graphs for optimizing sequential segmentation;and graph-based perceptual segmentation of stereo vision 3D images at multiple abstraction levels.
the proceedings contain 37 papers. the topics discussed include: an effective multi-level algorithm based on simulated annealing for bisecting graph;exact solution of permuted submodular MinSum problems;efficient shap...
详细信息
ISBN:
(纸本)9783540741954
the proceedings contain 37 papers. the topics discussed include: an effective multi-level algorithm based on simulated annealing for bisecting graph;exact solution of permuted submodular MinSum problems;efficient shape matching via graph cuts;simulating classics mosaics with graph cuts;an energy minimisation approach to attributed graph regularisation;a pupil localization algorithm based on adaptive gabor filtering and navigating radial symmetry;decomposing document images by heuristic search;skew detection based on elongate feature;active appearance models fitting with occlusion;combining left and right irises for personal authentication;bottom-up recognition and parsing of the human body;an automatic portrait system based on and-or graph representation;and object category recognition using generative template boosting.
A new approach to time frequency transform and patternrecognition of non-Stationary power signals is presented in this paper. In the proposed work Visual localization, detection and classification of non-stationary p...
详细信息
ISBN:
(纸本)9781424409822
A new approach to time frequency transform and patternrecognition of non-Stationary power signals is presented in this paper. In the proposed work Visual localization, detection and classification of non-stationary power signals are achieved using HS-Transform and automatic patternrecognition is carried out using fuzzy C-means based Genetic algorithm. Time frequency analysis and Feature extraction from the non-stationary power signals is done by HS-Transform. Once the feature vectors are extracted is used for patternrecognition of various non-Stationary signals. Various non-stationary power signals are processed through HS-transform with hyperbolic window to generate time-frequency contours for extracting relevant features for pattern classification. the extracted features are clustered using fuzzy C-means algorithm and finally the algorithm is extended using Genetic algorithm to refine the cluster centers. the average classification accuracy of the disturbances is 93.25% and 95% using fuzzy C-means and Genetic based fuzzy C-means algorithm, respectively.
Moving cast shadow causes serious problem while segmenting and extracting foreground from image sequences, due to the misclassification of moving shadow as foreground. this paper proposes a boosting discriminative mod...
详细信息
the existing significant point generation technique such as convex hull is unable to find out the proper significant points for irregular shaped objects. To address this issue, an algorithm namely efficient significan...
详细信息
ISBN:
(纸本)9781424409822
the existing significant point generation technique such as convex hull is unable to find out the proper significant points for irregular shaped objects. To address this issue, an algorithm namely efficient significant point generation (ESG) has been proposed in this paper that is able to find out proper significant points which will further be used to produce required shape descriptor using parametric curve generation techniques. Experimental results confirm the superior performance of ESG algorithm over convex hull in approximating significant points for all types of irregular shaped objects.
Triangle algorithm is used widely in the field of star patternrecognition, but it also has disadvantage that recognition reliability decreases seriously in areas where there are many stars existing small angular sepa...
详细信息
ISBN:
(纸本)9781424409723
Triangle algorithm is used widely in the field of star patternrecognition, but it also has disadvantage that recognition reliability decreases seriously in areas where there are many stars existing small angular separation. the character match algorithm can solve this problem and it has small size of guide star database, but it can not recognize a star image with displacement caused by camera movement and rotation. To overcome this disadvantage, an effective star patternrecognition algorithm is proposed in this paper. this algorithm divided the entire celestial sphere into a lot of square areas based on some bright stars during constructing guide star database, and then it used sub-areas selected from these square areas for star patternrecognition according to characteristic of star image sensed by star sensor. the simulation results show that the algorithm in this paper not only inherits advantages of character match algorithm, but also has strong robustness against star image displacement.
暂无评论