We discuss how approximation spaces considered in the context of rough sets and information granule theory have evolved over the last 20 years from simple approximation spaces to more complex spaces. Some research tre...
详细信息
For spruce (Picea abies (L.) Karst.), as with most other species, the value of a log without major defects like decay or compression wood is to great extent determined by its knot structure. that is the reason why saw...
详细信息
In this paper we show how inexact multisubgraph matching can be solved using methods based on the projections of vertices (and their connections) into the eigenspaces of graphs - and associated clustering methods. Our...
详细信息
In this work a new algorithm for texture analysis is presented. Over a region with size NxN in the image, a texture print is found by means of counting the number of changes in the sign of the derivative in the gray l...
详细信息
We compare two diverse classification strategies on real-life biomedical data. One is based on a genetic algorithm-driven feature extraction method, combined with data fusion and the use of a simple, single classifier...
详细信息
One of the important reasons for poor recognition rate in optical character recognition (OCR) system is the error in character segmentation. Existence of touching characters in the scanned documents is a major problem...
详细信息
One of the important reasons for poor recognition rate in optical character recognition (OCR) system is the error in character segmentation. Existence of touching characters in the scanned documents is a major problem to design an effective character segmentation procedure. In this paper, a new technique is presented for identification and segmentation of touching characters. the technique is based on fuzzy multifactorial analysis. A predictive algorithm is developed for effectively selecting possible cut columns for segmenting the touching characters. the proposed method has been applied to printed documents in Devnagari and Bangla: the two most popular scripts of the Indian sub-continent. the results obtained from a test-set of considerable size show that a reasonable improvement in recognition rate can be achieved with a modest increase in computations.
the proceedings contain 90 papers. the special focus in this conference is on Graphs, Languages, Strings and Grammars. the topics include: Spectral methods for view-based 3-D object recognition using silhouettes;machi...
ISBN:
(纸本)3540440119
the proceedings contain 90 papers. the special focus in this conference is on Graphs, Languages, Strings and Grammars. the topics include: Spectral methods for view-based 3-D object recognition using silhouettes;machine learning for sequential data;graph-based methods for vision;reducing the computational cost of computing approximated median strings;tree k-grammar models for natural language modelling and parsing;algorithms for learning function distinguishable regular languages;non-bayesian graph matching without explicit compatibility calculations;spectral feature vectors for graph clustering;identification of diatoms by grid graph matching;string edit distance, random walks and graph matching;learning structural variations in shock trees;a comparison of algorithms for maximum common subgraph on randomly connected graphs;inexact multisub graph matching using graph eigenspace and clustering models;optimal lower bound for generalized median problems in metric space;structural description to recognising arabic characters using decision tree learning techniques;feature approach for printed document image analysis;example-driven graphics recognition;estimation of texels for regular mosaics using model-based interaction maps;using graph search techniques for contextual colour retrieval;comparing shape and temporal PDMs;linear shape recognition with mixtures of point distribution models;curvature weighted evidence combination for shape-from-shading;probabilistic decisions in production nets;an application of machine learning techniques for the classification of glaucomatous progression;estimating the joint probability distribution of random vertices and arcs by means of second-order random graphs.
In this paper, according to the concept of Generalized Fisher Discriminan(GFD) presented by *** and ***, Generalized Kernel Function Fisher Discriminant(GKFD) is investigated and proved based on Linear Fisher Discrimi...
详细信息
ISBN:
(纸本)0780374886
In this paper, according to the concept of Generalized Fisher Discriminan(GFD) presented by *** and ***, Generalized Kernel Function Fisher Discriminant(GKFD) is investigated and proved based on Linear Fisher Discriminant(LFD) and Kernel Function Fisher Discriminant(KFD). It generalizes the solution of two-class patternrecognition nonlinearly, and decision function is obtained In the press of decision, competition principle is used, each test sample is determined as the class withthe largest decision function value, and a valid approach is provided for multi-class patternrecognition. GKFD has the characteristic of solid theory foundation and strong generalization capability, which embraces important meanings and application merits in multi-class patternrecognition.
this study attempts to apply the principle of neural networks and patternrecognition (PR) technologies to real-time recognition by client-Server network structure into a web-based recognizing system. In this paper, w...
详细信息
ISBN:
(纸本)0769516564
this study attempts to apply the principle of neural networks and patternrecognition (PR) technologies to real-time recognition by client-Server network structure into a web-based recognizing system. In this paper, we recommend a Web-Based PR technology, which is unproved recurrent neural network (RNN) from possessing feedback and non-linear activation function with its input, be taken out threshold. the purpose of this article is to construct a Client-Server network structure for PR system with associative memory. the Server-end is built a databases management system for storage sample patterns. In proceed with training, the user can real-time assign any pattern, which is a record in the Server-end databases. In deal with retrieve task, we propose a novel PR method via databases matching, it can efficient solve spurious states problem from RNN in the WBPR system. On the other hand, taking advantage of Database Matching is to overcome the capacity restrictions on RNN. In order to clarify, and corroborate the above Web-Based PR technology, thus a simulation experiment will be presented and their algorithms are also discussed.
:On the basis of DBF nets proposed by Wang Shoujue,. the model and implement of DBF neural network were discussed in this paper. When applied in patternrecognition, the algorithm and implement on hardware were presen...
详细信息
ISBN:
(纸本)0780374886
:On the basis of DBF nets proposed by Wang Shoujue,. the model and implement of DBF neural network were discussed in this paper. When applied in patternrecognition, the algorithm and implement on hardware were presented respectively. Compared with traditional RBF, DBF neural networks have been shown to offer advantages in accuracy, high speed and easy for hardware implement.
暂无评论