Journal of Shandong University(Engineering Science)

Adaptive dual learning algorithm for principal (minor) component analysis along geodesic on Stiefel manifold

LIU Lijun1, MA Yumei1, MENG Jiana2

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2014, 44(2): 1-5. doi:10.6040/j.issn.1672-3961.1.2013.026

Abstract ( 1253 )

PDF (1562KB) ( 689 )

Save

Related Articles | Metrics

Using the same topology as that of Oja-Brockett-Xu parallel neural network, a novel dual purpose adaptive algorithm for principal and minor component extraction was proposed by the optimization framework of a weighted Rayleigh quotient on the compact Stiefel manifold. By taking the right translation invariant Killing metric on orthogonal matrix group and search along the geodesic emanating from identity by means of exponential map, a novel dual learning algorithm for principal and minor component analysis was proposed. The proposed algorithm could switch from PCA (Principal Component Analysis) to MCA (Minor Component Analysis) with a simple sign change of its stepsize parameter. Moreover, orthonormality of the weight matrix was guaranteed at any iteration step. The effectiveness of the proposed algorithm was further verified in the section of numerical simulation.

The efficiency of power analysis attack based on S-boxes of block ciphers

WANG Danhui1, WANG An2*

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2014, 44(2): 6-11. doi:10.6040/j.issn.1672-3961.0.2014.041

Abstract ( 1453 )

PDF (1115KB) ( 1101 )

Save

Related Articles | Metrics

In order to make up the safety assessment of S-boxes, which was traditionally tested by mathematical methods, correlation power analysis and template attack were used on different S-boxes of DES, AES and Serpent. The simulations were put forth by MATLAB to obtain the success rates. The S-box of Serpent was the strongest against both the correlation power analysis and template attack. In correlation power analysis, Serpent needed 206 power traces, while AES and DES needed only 112 and 160 respectively. In addition, AES was stronger than DES against template attack. The results showed that the low-wide S-box was stronger than the high-wide S-box with the same total number of bits in one round.

Prediction of protein energy hot spots based on recursion feature elimination

WEI Xiaomin, XU Bin, GUAN Jihong

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2014, 44(2): 12-20. doi:10.6040/j.issn.1672-3961.1.2013.378

Abstract ( 1346 )

PDF (2923KB) ( 713 )

Save

Related Articles | Metrics

18 new features such as residue contact number and the proportion of relative change of accessible surface area et al. were derived based on the analysis of protein-protein interaction energy hot spots. Two recursion feature elimination methods were used to select discriminative feature subsets and two corresponding prediction models were proposed, noted as SVM- RFE and F-Score-RFE. The experimental results showed that the prediction model F-Score-RFE could improve 6.25％ in the value of F1 compared with the best existing method on the same independent test dataset, which indicated that new features defined were significant to improve the performance of prediction.

Multi-target real-time tracking method under harsh environment

QIU Xiaoxin1,2, ZHANG Wenqiang1,2*, QIN Jinxian1,2, DU Zhengyang1,2, ZHANG Defeng1,2

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2014, 44(2): 21-27. doi:10.6040/j.issn.1672-3961.2.2013.322

Abstract ( 1181 )

PDF (2473KB) ( 748 )

Save

Related Articles | Metrics

A new method for multi-target real-time tracking system working under harsh environment was presented. Compared to the current researches in this field, this method could track the invasion target more accurately in the video surveillance scene, and the efficiency of the algorithm was improved greatly. First, based on the current mainstream background modeling method codebook algorithm, the background updating method was improved, which made the computation efficiency and the foreground of the codebook detection accuracy improved greatly compared to the original algorithm. The main performance was better than other mainstream background modeling algorithm. Then, a particle filter algorithm as the multi-target tracking method made large improvement for resampling method that could maintain the effectiveness and diversity of the particles in real time environment. Experiments proved this system could be used in multi-target real-time tracking under harsh environment and had effective performances.

Algorithm of scale change objects tracking with adaptive bandwidth

LI Wu, HOU Zhiqiang*, WEI Guojian, YU Wangsheng

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2014, 44(2): 28-34. doi:10.6040/j.issn.1672-3961.0.2013.214

Abstract ( 1470 )

PDF (3121KB) ( 860 )

Save

Related Articles | Metrics

Taking advantage of the invariant characters for translation and rotation of the Log-polar transformation, a scale tracking method based on the ellipse log-polar transform was investigated to resolve the problem of scale change object tracking. Firstly, a saliency weighted mean shift was presented to locate the objects spatial orientation, then the candidate object region was translated to the ellipse log-polar transformation and integrated along the scale axis. The new proposed method estimated the targets scale parameters according to the maximum correlation coefficient in the transform domain. Experimental results demonstrated that the algorithm could adapt to the objects scale changes and the tracking error was lower. Compared with the traditional, it had a better robustness.

Image fusion method based on saliency computation and adaptive PCNN

YANG Xiulin1, HUANG Shuo2*, DENG Miao1, ZHANG Jihong1,3

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2014, 44(2): 35-42. doi:10.6040/j.issn.1672-3961.2.2013.255

Abstract ( 1201 )

PDF (2315KB) ( 472 )

Save

Related Articles | Metrics

Simple average processing of low-pass subbands was usually adopted in multi-scale transform based image fusion, which could not protect the saliency information in source images very well. To solve this problem, an image fusion method using the saliency computation to drive adaptive pulse-coupled neural network (PCNN) was proposed to optimize the low-pass subbands fusion. First, source images were decomposed by morphological un-decimated wavelet, and low-frequency coefficients and high-frequency coefficients were obtained. Second, low-frequency coefficients were fused by rule based on saliency computation and PCNN, and high-frequency coefficients were selected by strong absolute value. Finally, the fusion image was got by inverse transform. Experimental results indicated that saliency information of source images was obtained to some extent and the fusion indicators, such as mutual information, entropy, average gradient and edge preservation degree were improved.

An improved adaboost algorithm based on new Haar-like feature for face detection

JIANG Weijian1,2, GUO Gongde1,2*, LAI Zhiming1,2

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2014, 44(2): 43-48. doi:10.6040/j.issn.1672-3961.1.2013.003

Abstract ( 1763 )

PDF (2463KB) ( 2756 )

Save

Related Articles | Metrics

To solve problem of highly complexity of multi-angle face detection by the Adaboost algorithm based on the Haar-like, two new groups of extended Haar-like feature were proposed and the calculation method were exploited by the integral image. Then, the frontal faces cascaded classifier and the profile faces cascaded classifier was trained on the face database by the Adaboost algorithm respectively. Finally, the two-channel cascaded classifier was built. On OpenCV which is an open source vision database, the experimental results showed that the proposed method had better performance both in accuracy and computing speed, and could detect face with less weak classifiers. Meanwhile, the cascaded classifier had a good ability of robustness on detecting multi-angle face.

Research of reactive power optimization based on subspace bacterial colony chemotaxis algorithm to in distribution networks with distributed generation#br#

CHEN Jiming, SUN Mingyu, YOU Jujuan, KANG Zhongjian

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2014, 44(2): 49-54. doi:10.6040/j.issn.1672-3961.0.2013.125

Abstract ( 1514 )

PDF (1292KB) ( 967 )

Save

Related Articles | Metrics

The reactive power optimization with distributed generation was discussed. By integrating traditional capacitor reactive power compensation, the distribution generator was used in the reactive power optimization as a continuous adjustable reactive source. Combining subspace conception, the subspace bacterial colony chemotaxis algorithm was proposed, and the global optimization capability of algorithm was enhanced. The application of algorithm in IEEE33 system showed that reactive power optimization with distributed generation was better to reduce the system loss and improve the voltage level. The validity and possibility of the algorithm was also verified.

Modelling and optimization of a production service system with impatient customers

WANG Xiaoyan1, WANG Kangzhou2, QI Zhongbin1, JIANG Zhibin3, LI Na3

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2014, 44(2): 55-63. doi:10.6040/j.issn.1672-3961.0.2013.334

Abstract ( 1278 )

PDF (1465KB) ( 527 )

Save

Related Articles | Metrics

Product and service mix was provied with customers by a service-oriented manufacturing firm. The modelling and optimization of a production service system with two types of items and customers was considered in this paper. Two types of items were produced by the manufacturing facility in the first stage, then two types of orders were fulfilled at the service centre by the dedicated items in the second stage, one unit of item was depleted by each order. The problem is formulated as a Markov chain, this formulation was used to characterize the structure of the optimal scheduling policies. They show that the optimal service policy is a generalization of the well-known optimal production scheduling policy in classical production inventory system and service scheduling policy in classical service system. The benefits of the optimal service policy were showed by numerical experiments against several other policies.

A comprehensive method for traffic lights detection in complex background

SI Junshu, ZHU Wenxing*, SHA Yonghe

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2014, 44(2): 64-68. doi:10.6040/j.issn.1672-3961.0.2013.319

Abstract ( 1628 )

PDF (1956KB) ( 1641 )

Save

Related Articles | Metrics

To solve the auto-recognition problem of multi-type traffic lights in complex background, a novel method based on brightness value division,K-means clustering and histogram analysis of foreground objects was proposed. Firstly, self-localization was realized by brightness value division, geometric features analysis and classification statistics. Along with signal color judgment by K-means clustering algorithm, the type and direction information of traffic lights would be acquired by analyzing the histograms of foreground objects; accordingly auto-detection was realized. Experimental results showed that the proposed method had high detection accuracy for multi-type traffic lights in different scenarios. It was proved that this method was highly reliable and widely applicable.

Preparation of activated carbon with high specific surface area using oily sludge

DENG Hao, WANG Rongsha*, ZHANG Mingdong, REN Wen, XIE Shuixiang, YUE Yong, LIU Peng, LIU Xiaohui

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2014, 44(2): 69-75. doi:10.6040/j.issn.1672-3961.0.2013.264

Abstract ( 1352 )

PDF (1478KB) ( 1012 )

Save

Related Articles | Metrics

An activated carbon with high specific surface area was prepared from oily sludge. The whole preparation was put forth in a static pyrolyzing furnace with nitrogen and sodium hydroxide as the activation agent. The effects of conditions on the activated carbon were studied, such as carbonization temperature, heating way for activation, activation temperature, activation time and the mass ratio (m(NaOH)／m(C)). The specific surface area and pore size distribution were tested by the automated surface area and pore size analyzer. The element and microstructure were characterized by ESEM. The results showed that the best experimental conditions were as the followings, the carbonization temperature was 500℃, the heating way for activation was method c, the activation temperature was 800℃, the activation time was 1h, and m(NaOH)／m(C) was 2∶1. The surface area of the activated carbon was more than 2000m2／g. The activated carbon had uniform pore size and its average pore size was less than 2nm. Its total pore volume was greater than 2cm3／g. Based on those features, activated carbon could be used as storage media for energy, electrode materials, highly effective adsorbents and others basic materials. Therefore, this research could give a new way for resource utilization of oily sludge.

Durability study of PET-brick powder mortar

GE Zhi1, MENG Fanbo2, YUE Hongya1, HU Yizhang1

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2014, 44(2): 76-79. doi:10.6040/j.issn.1672-3961.0.2014.048

Abstract ( 1146 )

PDF (1131KB) ( 543 )

Save

Related Articles | Metrics

PET was used as binder to produce PET-brick powder mortar. The water absorption, sulfate resistance, and chloride penetration resistance of the mortar were investigated. The SEM test was performed to exam the microstructure. The test results showed that the absorption was 0.9％ after 7 days. After 20 cycles, the difference in compressive strength was only 1.83％ for samples in the sulfate solution and water. The sample had low chloride permeability with a test result of 48 C. The test results showed that the PET-brick powder mortar had high durability. The SEM results indicated that the mortar had uniform structure with very small amount of pores inside.

Time-frequency analysis of cabin noise based on ensemble empirical mode decomposition and independent component analysis#br#

WANG Li, ZHOU Yiqi, YU Gang, MI Yongzhen

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2014, 44(2): 80-88. doi:10.6040/j.issn.1672-3961.0.2013.321

Abstract ( 1389 )

PDF (3110KB) ( 935 )

Save

Related Articles | Metrics

In order to effectively control cabin noise and improve noise quality, combined techniques of ensemble empirical mode decomposition (EEMD) and an improvement of independent component analysis (ICA) were applied to analysze cabin noise. Intrinsic mode functions resulting from EEMD analysis were used as the virtual channels of ICA, which solved an undetermined problem of blind source identification from single sampling signal. The feasibility of EEMD-ICA method for complex non-stationary signals was verified by analyzing the simulated signals. With the experiment and coherence analysis for cabin interior and outside noise and vibration signals of a bulldozer, the technique of EEMD-ICA was effective on the noise source separation and identification. Meanwhile, techniques of coherence analysis and time-frequency analysis were used for accurate noise sources identification. The results showed that the main cabin noise source was combustion noise of diesel engine and mechanical radiation noise of cabin parts caused by 1／2 order and 1 order rotation frequency of the diesel engine. And generation details of noise source, transmission paths of air-borne or structure-borne noise were well analyzed in order to control the interior noise and identify faults by the techniques of coherence analysis and time-frequency analysis.

Experiment on mercury speciation transformation during coal pyrolysis with calcium chloride addition

SUI Hui, DONG Yong*, ZHANG Mengze, YU Min, WANG Peng

JOURNAL OF SHANDONG UNIVERSITY (ENGINEERING SCIENCE). 2014, 44(2): 89-94. doi:10.6040/j.issn.1672-3961.0.2013.308

Abstract ( 1234 )

PDF (1605KB) ( 465 )

Save

Related Articles | Metrics

Adding CaCl2 cooperating with SCR catalyst to accelerate the transformation of Hg0 during coal pyrolysis was proposed based on the fact that elemental mercury was the dominant species in the coal pyrolysis gas. A temperature-programmed tube furnace was set up to investigate the mercury release rate and cumulative release rate during coal pyrolysis. The mercury speciation transformation was determined by integration of the total released amount of various forms of mercury. The effect of different chlorine content and synergy of chlorine and catalyst on mercury transformation was investigated. The results showed that mercury emission from coal pyrolysis significantly depended on temperature. The release rate of Hg in coal increased with increasing temperature before 430℃, and it decreased subsequently with increasing temperature; Cumulative release rate of mercury showed significant positive linear correlation between 400~500℃, and about 90％ mercury was released at 600℃. With the amount of chloride was increased, percentage of Hg2+in gas increased dramatically, most of element mercury was oxidized forming Hg2+ under the synergy of catalyst.

Table of Content