Table of Content

25 January 2014, Volume 42 Issue 1

Previous Issue    Next Issue

Mechanical Engineering

2D Ultrasonic Track Model of Transit- Time Ultrasonic Flowmeter Considering Track Coverage

Liu Gui- xiong Chen Guo- yu Zhu Bin- geng Tan Wen- sheng

2014, 42(1):  1-5.  doi:10.3969/j.issn.1000-565X.2014.01.001

Asbtract ( 381 )   PDF (1526KB) ( 519 )  

Related Articles | Metrics

In order to reveal the effects of coverage and arrangement of ultrasonic track on the measurement accuracyof transit- time ultrasonic flowmeter,the characteristics of the 2D ultrasonic track model of 3D ultrasonic track wereanalyzed,and some formulas were deduced to calculate the coverage area in different covering modes,the numberof track segments as well as the distance between the centerlines of track segments and the center of pipe crosssec-tion,respectively.Moreover,a parameter design of the 2D ultrasonic track model was conducted.The results ofcase study indicate that the calculation method of the 2D ultrasonic track model of 3D ultrasonic track is effective.

Development and Swing Range Analysis of Base Fabric- Traverse Z- Shaped Weaving Device of Artificial Turf

You Dong- dong Liu Peng

2014, 42(1):  6-10.  doi:10.3969/j.issn.1000-565X.2014.01.002

Asbtract ( 304 )   PDF (1703KB) ( 654 )  

Related Articles | Metrics

In order to improve the bearing capacity of base fabric and the distribution uniformity of artificial turf,abase fabric- traverse device with eccentric wheel,which realizes the Z- shaped weaving of artificial turf,was deve-loped. Then,the influence of Z- shaped swing range on the bearing capacity of base fabric of artificial turf was ana-lyzed. Moreover,a virtual prototype model of the device was established,and the corresponding motion simulationand unloaded operation were carried out. Finally,tensile tests of the base fabric with a row spacing of 19 mm(3/4 inches) and with four different swing ranges were conducted. The results indicate that,as compared withline- shaped artificial turf,Z- shaped artificial one is of better distribution uniformity and stronger bearing capacity ofbase fabric,and that the bearing capacity of base fabric becomes superior when the swing range is about half of theline spacing of artificial turf.

In- Situ Synthesis of TiN Coating on Titanium Alloy Surface by EDM with Functional Electrode

Xu An- yang Liu Zhi- dong Li Wen- pei Qiu Ming- bo Tian Zhong- jun

2014, 42(1):  11-16.  doi:10.3969/j.issn.1000-565X.2014.01.003

Asbtract ( 332 )   PDF (3272KB) ( 551 )  

Related Articles | Metrics

Proposed in this paper is a new method to avoid the defects on titanium alloy surface formed during theEDM- induced ablation machining with functional electrode.In this method,nitrogen gas is injected into the pro-cessing area through the original processing system,and the electric spark discharge is adopted to trim the ma-chined surface. With the help of the heat due to the discharge,a reaction of nitrogen gas and titanium occurs,andthe TiN coating is thus synthesized in situ. It is found from the SEM,XRD and microhardness test results of TiNcoating that (1) the electric spark discharge process is stable and continuous in the process of coating formation andis of efficient energy utilization; (2) the coating is a dense,uniform and continuous layer dominated by TiNstrengthening phase,with a thickness of more than 400μm; (3) the microhardness of the coating reaches 1761.8,which is 4.7 times more than that of the matrix; (4) there exists a metallurgical combination with high bondingstrength and even TiN phase distribution between the coating and the matrix; and (5) the synthesis reaction has astrong reverse plating effect on the electrode,which causes a TiN strengthening phase- dominated electrode endface,effectively reduces the electrode spoilage,avoids the coating pollution caused by electrode elements,and effe-ctively maintains the purity of TiN phase in the coating.

Calculation and Analysis of Deflection on Piston Rod of Slim Hydraulic Hoist Cylinder

Ding Wen- si Zhang Xu Fan Ya- jun

2014, 42(1):  17-21.  doi:10.3969/j.issn.1000-565X.2014.01.004

Asbtract ( 365 )   PDF (1302KB) ( 817 )  

Related Articles | Metrics

In this paper,first,the flexural deflections on the cylinder and the piston rod of slim hydraulic hoist arecalculated and analyzed based on an improved ladder pole model,and the effects of some parameters,such as theinstallation angle of the cylinder, the fit clearance between the cylinder and the piston rod, the dead load of the pis-ton rod and the oil mass,on the transverse deformation of the shaft,are taken into consideration. Then,accordingto the basic theory of solid mechanics,the load and deformation of the cylinder and the piston rod are calculated,and an analytical expression describing the deflection on the cylinder and the piston rod is deduced by establishingand solving deflection equations. Finally,an engineering instance is given,and the results show that the proposedmethod is superior to the finite element method because it is of a small calculation error of less than 5%.

Dynamic Stress Test and Transient Analysis of Hydraulic Excavator

Ren Zhi- gui Chen Jin Wang Shu- chun Pang Xiao- ping Huang Ding- hong Ma Jin- cheng

2014, 42(1):  22-28.  doi:10.3969/j.issn.1000-565X.2014.01.005

Asbtract ( 389 )   PDF (2155KB) ( 706 )  

Related Articles | Metrics

In this paper,the dynamic stress characteristics of a working device of hydraulic excavator in a normaldigging process are investigated via dynamic test and transient analysis. Firstly,for a normal digging process ofbackhoe hydraulic excavator,synchronous test platforms were constructed respectively for the tests of dynamicstrain,oil pressure and angular displacement.Then,based on the d'Alembert principle,dynamic balance equa-tions of the boom,the arm and the bucket were respectively established and were used to calculate the dynamic loadspectrum of each hinge point with the test data.Moreover,the calculated dynamic load was transformed into the lo-cal coordinate system from the base one,which helps to perform a transient analysis later.Finally,the test resultsof dynamic stress were compared with those obtained by transient analysis.It is found that the stress obtained by thetwo methods are of the same variation law,and that the transient analysis helps to reveal the overall dynamic stresscharacteristics of the working device and provides a basis for the determination of dynamic stress distribution ruleand dangerous section of the working device.

Design of the Fourth- Order Transmission Error for Spiral Bevel Gears

2014, 42(1):  29-34.  doi:10.3969/j.issn.1000-565X.2014.01.006

Asbtract ( 327 )   PDF (2222KB) ( 451 )  

Related Articles | Metrics

In order to obtain the amplitudes of different meshing transfer points and the minimum fluctuation of thetransmission error under a given predesigned contact ratio and applied load,a design method of the fourth- ordertransmission error is proposed for the spiral bevel gears being cut via the modified roll method on the pinion. In theinvestigation,a symmetrical design of meshing position with predesigned contact ratio of spiral bevel gears isachieved by adjusting the direction of the contact path and the location of the reference point. Then,a high- orderdegeneration coefficient is deduced by means of the anti- TCA (Tooth Contact Analysis) method with a pre- designedexpression of the fourth- order transmission error.Numerical results show that the fourth- order transmission errorswith different amplitudes of meshing transfer points can be obtained by adjusting the former four coefficients of thepolynomial,which affects the fluctuation of the loaded transmission error and the corresponding applied load; andthat the fourth- order transmission error is controlled by the ease- off of the pinion surface.

Transmission Precision Analysis of Planetary Mechanism with Multi- Tooth Meshing

Lin Chao Cai Zhi- qin

2014, 42(1):  35-40,46.  doi:10.3969/j.issn.1000-565X.2014.01.007

Asbtract ( 720 )   PDF (2619KB) ( 417 )  

Related Articles | Metrics

In this paper,first,the transmission principle of a new- type transmission mechanism,namely a planetarymechanism with multi- tooth meshing,was investigated. Next,based on the transmission principle, the influences ofeccentric error,gear phase angle,elastic deformation and load on the transmission accuracy of the mechanism wereanalyzed,and a model was established to describe the transmission accuracy varying with the angle,with the corre-sponding transmission accuracy equations being also deduced. Then,the virtual prototype technology was adoptedto perform a simulation and verification in ADAMS,and a simplified experiment was finally conducted to obtain theoutput angle error and the rotation error of the mechanism. It is found that the simulated results accord well with themathematical ones,and that the transmission accuracy of multi- tooth meshing mechanism is much higher than thatof the traditional planetary mechanism.

Optimization of Repairable Spare Parts for K/N Cold- Standby Redundant System Considering Scraps

Xue Tao Feng Yun- wen Qin Qiang

2014, 42(1):  41-46.  doi:10.3969/j.issn.1000-565X.2014.01.008

Asbtract ( 374 )   PDF (841KB) ( 679 )  

Related Articles | Metrics

K/N cold- standby redundant system is an important way of improving the system reliability.However,inthe existing analytical models of this system,the scrap of repairable spare parts are not taken into consideration.Tosolve this problem,in this paper,first,the characteristics and working procedure of K/N cold- standby redundantsystem are analyzed.Next,by taking the minimum maintenance cost as the optimization target,a single- indenturetwo- echelon optimization model of repairable spare parts is proposed.Then,based on the allocation and modelingtheory of repairable spare parts,the analysis process and method considering the spare parts- supporting efficiencywith scraps are investigated. Moreover, the optimization model is solved via the marginal analysis,and the demandsfor spare parts are subsequently forecasted. Finally,some examples are presented to verify the feasibility and appli-cability of the proposed method.

Maximum 3- Shift Correlated Kurtosis Deconvolution Algorithm for Blind Extraction of Signals

Yang Guang- yong Hu Guo- qing Chen Le Wu Hai- feng Song Jia- sheng

2014, 42(1):  47-51,58.  doi:10.3969/j.issn.1000-565X.2014.01.009

Asbtract ( 343 )   PDF (1580KB) ( 882 )  

Related Articles | Metrics

The traditional low- order adaptive filter is inefficient in suppressing or eliminating the higher- order statis-tical noises of laser displacement signals,and the convergence speed as well as the SNR (Signal- to- Noise Ratio) ofblind source separation based on the maximum entropy or the natural gradient may remarkably decrease under mi-cro- displacement measurement conditions.In order to solve these problems,a maximum 3- shifted correlated kurto-sis deconvolution (M3CKD) algorithm is presented on the basis of the kurtosis variation of mixed signals,and thecorresponding inverse filter is designed.Moreover,the convergence and stability conditions of the algorithm areanalyzed,and a blind extraction is carried out on a constructed experiment platform.The results indicate that theproposed algorithm is efficient in the blind extraction of laser displacement signals and multiple reflection signals,and that it outperforms the FastICA algorithm due to its high convergence speed and large signal- to- noise ratio.

Electronics, Communication & Automation Technology

Secondary User and Power Allocation in Spectrum Sensing of Multi- Channel Cognitive Ｒadio

Liu Xin Zhong Wei- zhi Kong Fan- qiang Jing Qing- feng

2014, 42(1):  52-58.  doi:10.3969/j.issn.1000-565X.2014.01.010

Asbtract ( 314 )   PDF (1740KB) ( 962 )  

Related Articles | Metrics

In order to improve the throughput of multi- channel cognitive radio (CＲ),a joint allocation algorithm ofsecondary user (SU) and transmission power for each sub- channel in the cooperative spectrum sensing is proposed.Based on the present state of the primary user,the proposed algorithm aims to maximize SU's total throughput overall the sub- channels subjected to the constraints such as SU's cooperative detection probability and total power,andachieves the joint allocation of SU and transmission power for each sub- channel through alternative direction optimi-zation.Simulated results show that the proposed algorithm helps to obtain the optimal value with less iteration andimproves SU's throughput by dynamically allocating the SU and the transmission power of each sub- channel accor-ding to the sub- channel state.

Denoising of Coal Combustion Flame Images Based on HMT Model in Dual- Tree Complex Wavelet Domain

Wu Yi- quan Song Yu

2014, 42(1):  59-65.  doi:10.3969/j.issn.1000-565X.2014.01.011

Asbtract ( 309 )   PDF (2054KB) ( 479 )  

Related Articles | Metrics

In order to effectively eliminate the noises existing in boiler coal combustion flame images that are unfa-vorable for the subsequent image feature extraction and temperature reconstruction,an image denoising methodbased on the HMT (Hidden Markov Tree) model in dual- tree complex wavelet domain is proposed.In this method,first,a dual- tree complex wavelet transform is performed for noisy flame image.Next,the real part and the imagi-nary part of the dual- tree complex wavelet coefficients are respectively modeled according to the HMT model.Then,the model parameters are estimated by using the expectation maximization algorithm,and the noiseless dual- treecomplex wavelet coefficients are estimated according to the Bayes minimum mean square error (MMSE) criterion.Finally,an inverse dual- tree complex wavelet transform is conducted to obtain denoised flame images.Experimentalresults show that the proposed method is superior to the wavelet VisuShrink threshold method and the method basedon HMT model in wavelet or Contourlet domain because it helps to reduce the noise more effectively and achievehigher peak signal- to- noise ratio.

Concept and Solving Method of Practicable Optimal Point

Gao Li- qun Kong Xiang- yong Ouyang Hai- bin Zou De- xuan

2014, 42(1):  66-71.  doi:10.3969/j.issn.1000-565X.2014.01.012

Asbtract ( 359 )   PDF (670KB) ( 793 )  

Related Articles | Metrics

From the viewpoint of engineering practice,a concept of practicable point is proposed to solve optimiza-tion problems.However,in a certain situations,not only the global extreme points of the objective function but alsothe characteristics of objective function in the extreme point neighborhood should be considered.In view of thisproblem,a concept of practicable optimal point,namely the global optimal point with a neighborhood constraint,isfurther proposed.Moreover,as neighborhood constraint is difficult to obtain with traditional methods,an approxi-mate method based on neighborhood sampling is put forward,and a rapid search algorithm based on the particleswarm optimization is presented to find the practicable optimal points meeting different requirements.Simulated re-sults indicate that both the proposed practicable optimal point and the corresponding search algorithm are of feasibi-lity,and that the search algorithm is effective in searching the practicable optimal point.

Recognition of Similar Handwritten Chinese Characters Based on CNN and Ｒandom Elastic Deformation

Gao Xue Wang You- wang

2014, 42(1):  72-76,83.  doi:10.3969/j.issn.1000-565X.2014.01.013

Asbtract ( 616 )   PDF (1191KB) ( 1953 )  

Related Articles | Metrics

In order to recognize similar handwritten Chinese characters effectively,a convolutional neural network(CNN) model is proposed,and the topology of the network model is presented.Then,the sample set is extendedby introducing a stochastic elastic deformation to enhance the generalization performance of the model.Experimen-tal results indicate that the recognition accuracy of the proposed CNN model is 1.66% higher than that of the tradi-tional CNN model,especially,for distorted handwritten Chinese characters,the recognition accuracy increases by12.85%; moreover,as compared with the traditional recognition methods,the proposed CNN model reduces therecognition error rate by 36.47%.It is thus concluded that the proposed method is effective.

Localization Algorithm Based on Classification of Distance Estimation Methods for Wireless Sensor Networks

Huang Zhi- yong Zhang Xin

2014, 42(1):  77-83.  doi:10.3969/j.issn.1000-565X.2014.01.014

Asbtract ( 334 )   PDF (1552KB) ( 467 )  

Related Articles | Metrics

Since the traditional localization algorithm of wireless sensor networks,namely the DV- Hop algorithmwithout distance measurement,is of low accuracy due to the imprecise estimated distance,a new one based on theclassification of distance estimation methods is proposed.In this algorithm,the concept of optimal auxiliary anchoris introduced to improve the accuracy of distance estimation.During the estimation,unknown nodes find the opti-mal auxiliary anchor in light of the requirements of distance estimation at the beginning，and then estimate the dis-tance according to the types of the estimation methods.Moreover,in order to further improve the localization accu-racy,a“nonlinearity- hop number”weighting factor is introduced in the trilateral localization and the expectation ofestimated coordinate of unknown nodes is finally obtained.Simulated results demonstrate that the proposed localiza-tion algorithm greatly improves the accuracy of distance estimation only with an appropriate increment of communi-cation cost.

FPGA Implementation for Computing Pixel Purity Index of Hyperspectral Images

Guo Jie Li Yun- song Yi Fang Wu Xian- yun

2014, 42(1):  84-89.  doi:10.3969/j.issn.1000-565X.2014.01.015

Asbtract ( 340 )   PDF (1590KB) ( 817 )  

Related Articles | Metrics

In hyperspectral images,there exists a huge amount of data with hundreds of spectral bands,which mayresult in heavy computational burden of the pixel purity index (PPI).In order to solve this problem,an architec-ture for FPGA (Field Programmable Gate Array) implementation is proposed.In this architecture,the matrix com-putation based on the parallel processing of projection vectors is adopted to reduce the interface data accesses,aninner product circuit with parallel bands is proposed to improve the projection computation parallelism,and simpli-fied projection vectors are employed to avoid the floating- point multiplication in hardware implementation.The re-sults of the experiments on the AVIＲIS Cuprite hyperspectral images with 350 ×350 resolution and 224 bands showthat the proposed architecture outperforms the ENVI software and some other existing architectures in terms of end-member extraction accuracy,and that the PPI computation can be completed within 10 seconds in the correspondingFPGA implementation,which meets the requirement of real- time processing in spaceborne/airborne fields.

Intra- Frame Coding of Depth Maps to Ｒestrain Coding Error Pervasion

Yang Xu Li Bing- bing Ma Xiang Chang Yi- lin

2014, 42(1):  90-96.  doi:10.3969/j.issn.1000-565X.2014.01.016

Asbtract ( 339 )   PDF (1233KB) ( 514 )  

Related Articles | Metrics

In order to improve the coding quality of depth maps and the virtual synthesis quality of 3D videos,anintra- frame coding method to restrain the coding error pervasion is proposed.In this method,the depth map is firstlysegmented into a flat region and an edge region according to the features of depth maps.Then,the distortion perva-sion properties in the two regions are analyzed,based on which the blocks or sub- blocks that cause the distortionpervasion are corrected by using a correction value during the rate- distortion optimization of macro blocks.Thus,alarge number of following blocks or sub- blocks influenced by the distortion pervasion may possess a smaller value ofrate- distortion cost function.Experimental results indicate that,as compared with the intra- coding method presentedby JMVC 8.5 and the intra- frame predication method based on segmentation,the proposed coding method improvesthe average peak signal- to- noise ratio of the virtual synthesis view by 0.30 and 0.25 dB,respectively.

Computer Science & Technology

Algorithms to Generate Location Privacy Area Based on Location Privacy Protection with Spatial Cloaking

Xu Hong- yun Xu Jun Gong Yu- jing Xu Meng- zhen

2014, 42(1):  93-103.  doi:10.3969/j.issn.1000-565X.2014.01.017

Asbtract ( 390 )   PDF (820KB) ( 611 )  

Related Articles | Metrics

In order to improve the switching success rate of location privacy area and thus realize personalized loca-tion privacy protection,two new algorithms to generate location privacy area,namely the primary centroid- offset al-gorithm and the advanced centroid- offset algorithm,are proposed based on the location privacy protection with spa-tial cloaking.The primary centroid- offset algorithm generates the switched location privacy area through offsettingthe centroid of location privacy area before switching, with an offset amplitude being determined by the distributionof neighbor users.The advanced centroid- offset algorithm moves the centroid of switched location privacy area out ofthe one before switching,and,according to the distribution of neighbor users,it generates the switched locationprivacy area without overlapping the one before switching.Experimental results show that,under center attacks,the switching success rate of the primary centroid- offset algorithm is more than 90%,while that of the advancedcentroid- offset algorithm is close to 100%; and that,under indiscriminate attacks，the switching success rate of theadvanced centroid- offset algorithm is also close to 100%.

Management Method of Energy Consumption Optimization Nodes in Cloud Storage System

Lin Wei- wei He Pin- jia Liu Bo

2014, 42(1):  104-110.  doi:10.3969/j.issn.100-565X.2014.01.018

Asbtract ( 324 )   PDF (493KB) ( 603 )  

Related Articles | Metrics

In order to reduce the energy consumption of cloud storage systems,a method to save energy by turningoff some storage nodes in inactive periods is proposed.Then,for the purpose of guaranteeing the availability of datawhen part of storage nodes are turned off,an auxiliary node- based greedy algorithm is designed to properly selectstorage nodes to form a turned- off node set.Moreover,an improved auxiliary node- based greedy algorithm for hete-rogeneous storage systems is proposed for the optimization of energy consumption.Simulated results show that theimproved auxiliary nod- based greedy algorithm is effective because it greatly reduces the energy consumption of he-terogeneous cloud storage systems and outperforms the conventional greedy algorithm.

SoS Filter and Its Application to Industrial Ultrasound Imaging

Dai Guang- zhi Han Guo- qiang Lin Wei- yi OuYang Xian- yue

2014, 42(1):  111-116.  doi:10.3969/j.issn.1000-565X.2014.01.019

Asbtract ( 393 )   PDF (1413KB) ( 484 )  

Related Articles | Metrics

Although the FRI (Finite Rate of Innovation) model may be feasible in reducing the sampling data andsampling frequency of industrial ultrasound imaging systems,the existing FRI model cannot be well applied due tothe high frequency of reflected signals.In order to solve this problem,a new FRI acquisition method named Sum of Sinc (SoS) filter,which is very suitable for the processing of industrial ultrasound signals,is proposed on the basisof the FRI filter group.Afterwards，the necessary conditions for this method are deduced,and the structure of theSoS filter is presented.From the results of raw one- dimension ultrasound imaging data，the proposed method isfound to be effective and feasible.

Extraction of Pavement Cracks Based on Valley Edge Detection of Fractional Integral

Wang Wei- xing Wu Lin- chun

2014, 42(1):  117-122.  doi:10.3969/j.issn.1000-565X.2014.01.020

Asbtract ( 334 )   PDF (2885KB) ( 472 )  

Related Articles | Metrics

As pavement crack images are difficult to segment due to the existence of high noise,weak boundary andsmall cracks,an extraction method of pavement cracks based on the valley edge detection of fractional integral isproposed.In this method,first,neighboring smoothing of the original image is performed to eliminate the noise andexpand the relative width of the cracks.Then,the main cracks are extracted via the valley edge detection of frac-tional integral,and the resulting image is further processed via the morphological approach with short- line noiseelimination. Afterwards,final cracks are extracted by using the gap linking method on maximum entropy thresholdto cause cracks to merge automatically. Experimental results show that the proposed method instantly helps to detectsmall pavement cracks with high accuracy and strong noise robustness.

Parallel Algorithm to Construct CSB⁺- Tree Indexing on Graphic Processing Unit

Liu Yong Xi Jian- qing Huang Dong- ping Jia Lian- yin Miao De- cheng

2014, 42(1):  123-127,134.  doi:10.3969/j.issn.1000-565X.2014.01.021

Asbtract ( 620 )   PDF (913KB) ( 754 )  

Related Articles | Metrics

In order to improve the operation efficiency of cache sensitive B⁺- tree (CSB⁺- tree) indexing,this pa-per deals with the parallel construction and query performance of CSB⁺- tree on graphic processing unit (GPU).Inthe investigation,first,the mapping relationship between each key in internal nodes and the corresponding leafnode of the index tree is analyzed, a lock- free parallel algorithm that once for all builds the CSB⁺- tree internal nodekeys is proposed,and the index tree is constructed at the maximum parallel speed.Moreover, dynamic arrays su-pporting the arbitrary expansion of CSB⁺- tree index data on GPU are designed to implement the dynamic allocationof memory space on GPU,and padding bits are added to the boundary of the internal nodes to reduce the number ofbranches,thus improving the query efficiency of CSB⁺- tree.Experimental results indicate that the proposed algo-rithm is 31. 0 and 1.4 times faster respectively than the parallel algorithms based on single node and tree layer.

Comonadic Corecursions on Strong Coinductive Data Types

Su Jin- dian Yu Shan- shan

2014, 42(1):  128-134.  doi:10.3969/j.issn.1000-565X.2014.01.022

Asbtract ( 267 )   PDF (949KB) ( 651 )  

Related Articles | Metrics

As the unfold on coinductive data types can not describe the corecursive functions with parameters,thefinal coalgebras on Cartesian closed category are proved to be strongly final,and a category- theoretical definition ofstrong coinductive data types as well as the corresponding corecursion with fixed parameters,which is called pun-fold,is proposed.As a result,the corecursion defined on coinductive data types may include extra parameters asthe input of calculation.Moreover,the comonadic corecursions based on comonads is used to give a unified de-scription for unfold and punfold,and various calculation laws for punfold are further analyzed.Thus,the researchesof Pardo on the recursions with parameters via Comonads are successfully extended to coinductive data types.

FP- Growth Algorithm Based on Boolean Matrix and MapReduce

Chen Xing- shu Zhang Shuai Tong Hao Cui Xiao- jing

2014, 42(1):  135-141.  doi:10.3969/j.issn.1000-565X.2014.01.023

Asbtract ( 463 )   PDF (1960KB) ( 979 )  

Related Articles | Metrics

Association rules mining is an important part of data mining.In order to improve the efficiency of associ-ation rules mining,an FP- Growth algorithm based on Boolean matrix and MapReduce,which is marked as BPFP,is proposed,with its time and space complexity being also analyzed.In BPFP algorithm,Hadoop framework andBoolean matrix are used to reduce the number of scans of transaction data,and twice Map- Reduce is adopted tomine frequent item sets.Experimental results on multiple data sets show that the improved FP- Growth algorithm issuperior to the original one due to its high execution efficiency and speedup.

Dictionary Learning via Locality Preserving for Sparse Representation

Chen Si- bao Zhao Ling Luo Bin

2014, 42(1):  142-146.  doi:10.3969/j.issn.1000-565X.2014.01.024

Asbtract ( 380 )   PDF (699KB) ( 554 )  

Related Articles | Metrics

The selection of dictionary is crucial to sparse representation classification.In order to preserve the localinformation of original training samples with less dictionary atoms and include more discriminant information in thelearned dictionary,a new dictionary learning method based on the locality preserving criterion is proposed for sparserepresentation.In this method,the locality preserving criterion is imposed on coding coefficients,which makes thecoding coefficients of neighboring data points in the dictionary close to each other and preserves the local informa-tion of original training samples.Experimental results on extended YaleB,AR and COIL20 databases show that theproposed method is effective because it is of higher classification performance than other methods.