This paper presents an off-line Arabic handwriting recognition system using the Hidden Markov Model Toolkit(HTK).HTK is a portable toolkit for speech recognition *** recognition system extracts a set of features on bi...
详细信息
This paper presents an off-line Arabic handwriting recognition system using the Hidden Markov Model Toolkit(HTK).HTK is a portable toolkit for speech recognition *** recognition system extracts a set of features on binary handwritten images using sliding widow,builds character HMM models and learns word HMM models using embedded training without character *** this paper studies the relationship between frame overlap and number of *** that have been implemented on the benchmark IFN/ENIT database show the average recognition rate of this system is85.43%.
For recognizing the objects in the image, the existing image recognition methods only can identify certain kinds of objects in the image recognition field and the recognition rate is also not high. This paper proposes...
详细信息
For recognizing the objects in the image, the existing image recognition methods only can identify certain kinds of objects in the image recognition field and the recognition rate is also not high. This paper proposes an image recognition method which is based on shape feature and texture feature. This method divides the objects images into two types, foreground images and background images. Then extract shape features to foreground images and extracted texture features to background images. Finally match the extracted features with corresponding established feature library. This method uses Mahalanobis distance to match the objects in the images. Experiment results have proved that this method not only can identify several different kinds of objects but also has a good recognition rate.
In this paper, the two dimension incompressible Navier-Stokes equations with pseudo-spectral method are solved using the related subroutines in FFTW and CUFFT. Compared with the codes on CPU, the performance of the co...
详细信息
In this paper, the two dimension incompressible Navier-Stokes equations with pseudo-spectral method are solved using the related subroutines in FFTW and CUFFT. Compared with the codes on CPU, the performance of the codes on GPU is much better, especially when the resolution increases. For the resolution of 2048 2048, the acceleration reaches 14.45 times. We also try to combine MPI(Message Passing Interface) and CUDA(Compute Unified Device Architecture) in our solver. Due to the inevitable frequent data transfers between Host and Device, the speedup is not so ideal compared with that of the single node, and 1.82 times acceleration is obtained in double precision for the resolution of 4096 4096.
This paper proposes a FCM-based QPSO algorithm for evolutionary fuzzy-system design. The objective of this paper is to learn TSK type fuzzy rules with high accuracy. In the designed fuzzy system,data is firstly cluste...
详细信息
This paper proposes a FCM-based QPSO algorithm for evolutionary fuzzy-system design. The objective of this paper is to learn TSK type fuzzy rules with high accuracy. In the designed fuzzy system,data is firstly clustered into classes by fuzzy c-means algotithm so that each rule defines its own fuzzy sets, the number of fuzzy rules is also determined by the number of clusters..Then Quantum-behaved particle swarm optimization learning algotithm then used for optimising the parameters of the fuzzy system. We illustrates the algorithm in details with computer simulation to solve nonlinear problems and compare the rusults between basic PSO and our algorithm.
there're so many kinds of localization algorithms with a big localization error. And how to reduce this error is really a very important problem. This paper studies four distributed localization algorithms in WSN,...
详细信息
there're so many kinds of localization algorithms with a big localization error. And how to reduce this error is really a very important problem. This paper studies four distributed localization algorithms in WSN, and introduces a pretty popular optimization algorithm named quantum particle swarm optimization algorithm into it. In the end, this paper does some simulations with matlab. And it proves that the QPSO algorithm works well in node localization algorithm.
In order to retrieve the semantic relations among RDF data, a new query pattern is introduced to SPARQL [1]. This paper presents the research work of extending the ARQ engine to realize the new SPARQL query in a more ...
详细信息
In order to retrieve the semantic relations among RDF data, a new query pattern is introduced to SPARQL [1]. This paper presents the research work of extending the ARQ engine to realize the new SPARQL query in a more efficient way by using a digraph retrieval model for searching RDF data instead of the iterator model adopted by the original ARQ engine. The design and implementation of the digraph retrieval model and its application to the semantic relation retrieval are discussed in detail. With the new query pattern, many interesting semantic relations among the RDF data can be explicitly and easily enquired.
A large number of devices are involved in bulk grain silos system and the control flows are particularly complicated. To cope with the complexity, a structured system program design approach is proposed in this study ...
详细信息
A large number of devices are involved in bulk grain silos system and the control flows are particularly complicated. To cope with the complexity, a structured system program design approach is proposed in this study while the ControlLogix system is adopted as the operating platform. Meantime, an intelligent flow path selection method by means of a flow table is used to select the control flow. The readability of the programming codes is greatly improved and the complication of sequence control for large-scale equipment is simplified. A range of functions are realized by help of the configuration software iFIX, including flow select, start, stop and suspend, as well as display of the running status of the devices and real-time alarming, etc.
The mutual exclusion of roles as an effective means of implementing separation of duty, has an important role on access control in RBAC systems. This paper deeply discusses the types that the mutual exclusion of roles...
详细信息
The mutual exclusion of roles as an effective means of implementing separation of duty, has an important role on access control in RBAC systems. This paper deeply discusses the types that the mutual exclusion of roles come from, and puts forward a model of describing mutual exclusion of roles. The model is similar to the structure of adjacency multiple table, and its basic unit is the role structure which can describe the relations with other roles. The paper also explores the situation after the digestion of the mutual exclusion, and finally gives out a simple application.
There are two method frequently used for parallelization: data parallel and task parallel. According to the characteristics of analysis of digital terrain, data parallel is more applied, so Petri nets is introduced in...
详细信息
There are two method frequently used for parallelization: data parallel and task parallel. According to the characteristics of analysis of digital terrain, data parallel is more applied, so Petri nets is introduced in this paper to describe the parallel relationships within data patitions based on granularity model. As for different storage environment, corresponding scheduling algorithms are proposed for load balance.
Building k-nearest neighbor(kNN) graphs is a necessary step in such areas as data mining and machine learning. So in this paper, we attempt to study the kNN furthermore, we first propose a parallel algorithm for appro...
详细信息
Building k-nearest neighbor(kNN) graphs is a necessary step in such areas as data mining and machine learning. So in this paper, we attempt to study the kNN furthermore, we first propose a parallel algorithm for approximate kNN graph construction and then apply the kNN graph to the application of clustering. Experiments show that our MPI/OpenMP mixed mode codes can make the construction of approximate kNN graph faster and make the parallelization and implementation easier. Finally, we compare the results of agglomerative clustering methods by using our parallel algorithm to illustrate the applicability of this method.
暂无评论