Table of Content

25 March 2017, Volume 45 Issue 3

Previous Issue    Next Issue

Electronics, Communication & Automation Technology

Residual Structure Characteristics- Based Block Classifying Reconstruction Algorithm for CVS

YANG Chun-ling LI Wen-hao

2017, 45(3):  1-10.  doi:10.3969/j.issn.1000-565X.2017.03.001

Asbtract ( 240 )   PDF (2284KB) ( 236 )  

Related Articles | Metrics

Most existing compressed video sensing ( CVS) algorithms with best reconstruction performance adopt a “prediction-residual reconstruction”strategy,which helps obtain high reconstruction quality by taking good advantage of intra-frame and inter-frame correlation.However,all of them ignore the residual structure characteristics and simply use SPL reconstruction algorithm which is only suitable for natural image compressed sensing.In order to solve this problem,a block classifying reconstruction algorithm on the basis of residual structure characteristics is proposed,which firstly classifies residual blocks according to their average energy and then adopts suitable algorithms to reconstruct residual blocks corresponding to their structure characteristics.Simulated results show that the proposed algorithm helps achieve higher reconstruction quality than SPL algorithm for video sequences with fast movements.

Feature Similarity Image Quality Assessment on the Basis of Human Visual System

SUN Yan-jing LIU Dong-lin XIE Xin-xin WANG Yan-fen

2017, 45(3):  11-19.  doi:10.3969/j.issn.1000-565X.2017.03.002

Asbtract ( 287 )   PDF (1905KB) ( 303 )  

Related Articles | Metrics

As the existing image quality evaluation methods of feature similarity ( FSIM) is inefficient in image information uncertainty measurement and edge information detection,a novel algorithm named HFSIM is proposed on the basis of the internal generative mechanism of human visual system ( HVS) .In this algorithm,the auto-regressive ( AR) model is employed to decompose distorted images,and the original image is decomposed into two portions,one is the predicted portion and the other is the disorderly portion.By combining FSIM with edge structural similarity ( ESSIM) algorithm,the predicted portion of image is measured,and,by employing the multi-scale peak signal-to-noise ratio ( PNSR) ,the distortion of the disorderly portion is measured.Finally,the overall image quality score is obtained according to the above-mentioned measured results of the predicted and the disorderly portions.It is found from the experiments on six public benchmark databases that the proposed algorithm is highly consistent with human perception,and that it possesses high performance in the assessment of different types of distorted images.

Fireworks Algorithm-Based Dimensionality Reduction for Hyperspectral Imagery Classification

CUI Ying SONG Guo-jiao CHEN Li-wei LIU Shu-bin WANG Li-guo

2017, 45(3):  20-28.  doi:10.3969/j.issn.1000-565X.2017.03.003

Asbtract ( 260 )   PDF (783KB) ( 381 )  

Related Articles | Metrics

In order to reduce the size and computational complexity of hyperspectral data as well as to avoid Hughes phenomenon in the following classification,a novel hyperspectral image dimensionality reduction method on the basis of fireworks algorithm ( FWA) is proposed.In this method,the weighted sum of compactness coefficient and separation coefficient is used as the criterion for band selection,and a search is constantly performed and updated until the algorithm is convergent.Thus,an optimal band combination is successfully obtained.Then,a comparison among the dimensionality reduction methods respectively based on FWA,genetic algorithm ( GA) and tabu search ( TS) algorithm is made with AVIRIS and ROSIS datasets.The results indicate that FWA-based method is of lower computational complexity and smaller time consumption because it helps select fewer band combinations; and that,when the obtained band combination is used for hyperspectral image classification,the proposed FWA-based method is the best because it improves classification accuracy and Kappa coefficient by 0. 06% ～ 4. 72% and 0. 00 ～ 0. 09,respectively.

LS-ORTSEG Method for Segmentation of Buildings in Digital Surface Model

YAN Yi-ming ZHAO Chun-hui CUI Ying

2017, 45(3):  29-34.  doi:10.3969/j.issn.1000-565X.2017.03.004

Asbtract ( 271 )   PDF (1967KB) ( 327 )  

Related Articles | Metrics

Segmenting buildings in digital surface model ( DSM) is a key technique of 3D city reconstruction based on remote sensing data.In order to overcome the poor performance caused by the negative interference of topographic relief and edge in the segmentation of buildings in DSM,a hierarchical segmentation method named LSORTSEG is proposed in light of both level-set ( LS) method and the model of occlusions of random texture ( ORTSEG) .In this method,firstly,the potential building regions are extracted roughly by means of the LS method.Secondly,the regions are expanded properly.Then,the expanded regions are segmented by using the ORTSEG for an optimized segmentation of buildings.Experimental results show that the proposed method can improve the segmentation accuracy of buildings effectively.

A KMP-RBF Fusion Method to Forecast Duty Vehicle's Travel Time

JIN Shan JIN Zhi-gang LIU Yong-lei

2017, 45(3):  35-41,47.  doi:10.3969/j.issn.1000-565X.2017.03.005

Asbtract ( 260 )   PDF (1950KB) ( 378 )  

Related Articles | Metrics

Proposed in this paper is a KMP-RBF fusion method for forecasting the travel time of duty vehicle.In this method,the signal source consisting of GPS information and SCATS ( Sydney Coordinated Adaptive Traffic System) is utilized to establish a traffic information fusion model that combines fuzzy inference knowledge representation,MAPSO ( Multi-Agent Particle Swarm Optimization) and RBF ( Radial Basis Function) training together,the key parameters are optimized adaptively,and the time and space data are matched and obtained from historical training database.Experimental results show that the travel time after fusion and prediction is identical to the actual data measured by the traffic monitoring system,and that the proposed KMP-RBF fusion method is effective and reliable in the aspects of error rate，iterative degree and accuracy.

Design of a Low-Power Consumption D Flip - Flop on the Basis of a-IGZO TFTs

YAO Ruo-he LIN Shao-long

2017, 45(3):  42-47.  doi:10.3969/j.issn.1000-565X.2017.03.006

Asbtract ( 271 )   PDF (473KB) ( 309 )  

Related Articles | Metrics

Proposed in this paper is a low-power consumption D flip-flop circuit with asynchronous reset on the basis of Pseudo-CMOS logic gates,which consists of n-type a-IGZO TFTs ( Thin Film Transistors) ,replaces the diodeload in Pseudo-CMOS topology with dynamic load,and decreases the static power consumption by reducing the conduction probability of the circuit.The output stage of the circuit is a latch,and the effect of dynamic loadcaused output swing decrement on the delay is reduced through a feedback path.The proposed D flip-flop is then applied to the design of a ring shift register.The results show that the trigger circuit can reduce the static power consumption in NOR gate logic circuit effectively.

Computer Science & Technology

Resource Selection Algorithm on the Basis of Topic Model

DONG Shou-bin XIE Yi-fan YUAN Hua CHEN Jian-hao

2017, 45(3):  48-53.  doi:10.3969/j.issn.1000-565X.2017.03.007

Asbtract ( 194 )   PDF (302KB) ( 292 )  

Related Articles | Metrics

In the federated search environment with multiple real search engines,the small-document approach,which is inefficient in estimating the accurate number of indexed files in the process of resource description,may result in poor performance of resource selection methods.In order to solve this problem,a resource library description method on the basis of topic model is proposed,which adopts LDA topic model to obtain the description word of each resource library.Then,a new resource selection algorithm is proposed,which combines with both vertical weight and word vector to calculate the correlation between resource library and query request,and to obtain the final resource selection results according to the correlation.Experimental results show that the proposed resource selection algorithm on the basis of topic model improves the performance of resource selection and can be effectively applied in the federated search environment of distributed search engines.

LDA Subtopic Detection Algorithm with Background Noise Restraint

LI Jing-yuan QIU Zhi-jie LIU Yue CHENG Xue-qi REN Yan

2017, 45(3):  54-60.  doi:10.3969/j.issn.1000-565X.2017.03.008

Asbtract ( 216 )   PDF (279KB) ( 268 )  

Related Articles | Metrics

Special article set is a collection of articles with common background knowledge.In order to more effectively detect the subtopics form special article set with complex information correlation,an LDA subtopic detection algorithm with background noise restraintnamed BLDA is proposed,which improves the precision of subtopic detection from article set by firstly extracting the common background knowledge and then reproducing the keywords in each iteration step.By a series of experiments on a set of WeChat documents from public accounts,it is proved that the detection results obtained by BLDA are much better than those obtained by LDA,with a topic recall rate increment of about 170%,a Purity index increment of 143% and a NMI index increment of 160%.

Chinese Word Segmentation Method on the Basis of Bidirectional Long-Short Term Memory Model

ZHANG Hong-gang LI Huan

2017, 45(3):  61-67.  doi:10.3969/j.issn.1000-565X.2017.03.009

Asbtract ( 331 )   PDF (364KB) ( 314 )  

Related Articles | Metrics

Chinese word segmentation is one of the fundamental technologies of Chinese natural language processing.At present,most conventional Chinese word segmentation methods rely on feature engineering,which requires intensive labor to verify the effectiveness.With the rapid development of deep learning,it becomes realistic to learn features automatically by using neural network.In this paper,on the basis of bidirectional long short-term memory ( BLSTM) model,a novel Chinese word segmentation method is proposed.In this method,Chinese characters are represented into embedding vectors from a large-scale corpus,and then the vectors are applied to BLSTM model for segmentation.It is found from the experiments without feature engineering that the proposed method is of high performance in Chinese word segmentation on simplified Chinese datasets ( PKU,MSRA and CTB) and traditional Chinese dataset ( HKCityU) .

Convolutional Network-Based Semantic Similarity Model of Sentences

HUANG Jiang-ping JI Dong-hong

2017, 45(3):  68-75.  doi:10.3969/j.issn.1000-565X.2017.03.010

Asbtract ( 202 )   PDF (400KB) ( 295 )  

Related Articles | Metrics

Computing the semantic similarity between two sentences is an important research issue in natural language processing field,and,constructing an effective semantic model of sentences is the core task of natural language processing for paraphrase identification,textual similarity computation,question /answer and textual entailment.In this paper,a parallel convolutional neural network model is proposed to represent sentences with fixedlength vectors,and a similarity layer is used to measure the similarity of sentence pairs.Then,two tasks,namely paraphrase identification and textual similarity test,are used to evaluate the performance of the proposed model.Experimental results show that the proposed model can capture sentence s semantic information effectively; and that,in comparison with the state-of-the-art baseline,the proposed model improves the F1-score in paraphrase identification by 7. 4 percentage points,while in comparison with the logistic regression method,it improves the Pearson correlation coefficient in semantic similarity by 7. 1 percentage points.

Bacteria Biotope Extraction on the Basis of Neural Network

WANG Jian LI Hong-lei LIN Hong-fei YANG Zhi-hao ZHANG Shao-wu

2017, 45(3):  76-81.  doi:10.3969/j.issn.1000-565X.2017.03.011

Asbtract ( 218 )   PDF (268KB) ( 383 )  

Related Articles | Metrics

Proposed in this paper is a neural network-based method for extracting the relationship between bacteria and their habitats.In this method,implicit senior features are learnt automatically to avoid the complexity and redundancy of the traditional artificial design of features,and,distributed vector representation with rich syntactic and semantic knowledge for words and entities,two different neural network models,as well as integrated document- level extraction results,are comprehensively employed to make an evaluation on the BB-event corpus from BioNLP-ST 2016.Experimental results show that the proposed method achieves preferable F1 score,which means that it is effective in bacteria biotope extraction.

“Word Frequency-Filtering”Hybrid Feature Selection Method Applied to Spam Identification

CHEN Jun-ying ZHOU Shun-feng MIN Hua-qing

2017, 45(3):  82-88.  doi:10.3969/j.issn.1000-565X.2017.03.012

Asbtract ( 274 )   PDF (262KB) ( 349 )  

Related Articles | Metrics

In order to solve the increasingly rampant spam problem,naive Bayes and support vector machine classification methods are used to identify spam emails in this paper.In this method,"word frequency-filtering”hybrid feature selection method is applied to classification models to improve the identification performance of classifiers,and the identification performance of naive Bayes classification method is enhanced by considering more comprehensive classification probability cases.Moreover,some experiments are designed to test and verify the identification performance of the spam detection system in terms of accuracy rate,recall rate and F1 score.The results show that the proposed“word frequency-filtering”hybrid feature selection method can improve the identification performance of spam classifiers effectively,and that the classification output adjustment module based on the cost-sensitive method can greatly reduce the probability that the classifier mistakes a non-spam email as a spam email.In conclusion,the spam identification system designed and implemented in this paper possesses strong practicability and applicability in practical work and life.

A Symmetric Locally-Preserving Semi-Supervised Dimensionality Reduction Algorithm

XU Jin-cheng

2017, 45(3):  89-96.  doi:10.3969/j.issn.1000-565X.2017.03.013

Asbtract ( 240 )   PDF (443KB) ( 242 )  

Related Articles | Metrics

As many natural images are symmetrical and most of data distributions exhibit a manifold structure,a symmetric locally-preserving semi-supervised dimensionality reduction ( SLPSDR) algorithm is proposed.In the algorithm,a matrix is used to define the relationship between dimensionality reduction mapping matrix elements,so as to minimize the difference between the matrix elements of symmetric pixel points in an image.In order to keep the manifold structure of data by using the training samples without a label,it is required that the neighborhood relationship of each point in a low-dimension space is similar to that in a high-dimension space.The experimental results on CMU PIE,Extend YaleB,ORL and AR face databases show that the symmetric feature of image data causes the SLPSDRalgorithm to be superior to other contrastive dimensionality reduction algorithms.

A Hybrid Differential Evolution Algorithm with Multiple Search Strategies for Large-Scale Optimization

LUO Jia-xiang NI Xiao-ye HU Yue-ming

2017, 45(3):  97-103,116.  doi:10.3969/j.issn.1000-565X.2017.03.014

Asbtract ( 212 )   PDF (621KB) ( 283 )  

Related Articles | Metrics

Large-scale optimization problem with multiple peaks and high dimension is a hot topic in current optimization research field.By using the co-evolutionary algorithm as the framework,this paper proposes a hybrid differential evolution ( DE) algorithm with multiple search strategies to solve the large-scale optimization problem.In this algorithm,firstly,based on the thought of decomposition,a self-adaptive DE operator is applied to a local optimization of sub-problems.Then,a random search mechanism based on simulated annealing is introduced to improve the algorithm s global search ability,and a local search chain is combined to search the solution space deeply.Finally,a set of benchmark functions is employed to evaluate the proposed algorithm.The results show that the algorithm is prior to the existing ones because it helps obtain better average value and optimized solution.

Food Science & Technology

Effects of Acid Hydrolysis Time on Properties of Small-Particle Starch and Stability of Pickering Emulsion

HUANG Qiang YANG Yin-zhou WANG Chan FENG Kang

2017, 45(3):  104-110.  doi:10.3969/j.issn.1000-565X.2017.03.015

Asbtract ( 241 )   PDF (2012KB) ( 296 )  

Related Articles | Metrics

In this paper,firstly,by using maize starch as a raw material,small-particle starch granules were prepared by means of acid hydrolysis,octenyl-succinic anhydride ( OSA) modification and freeze-grinding.Then,the effects of acid hydrolysis time on the properties of small-particle starch and the stability of Pickering emulsion were investigated.Finally,the morphology and particle size distribution of the prepared small-particle starch were characterized by using a scanning electron microscopy and a particle size analyzer,and the application of the smallparticle starch to Pickering emulsion was investigated.The results show that ( 1) with the prolonging of hydrolysis time,the particle size of starch granules decreases significantly,and the volume average particle size reduces from 15. 6μm to 6. 6μm; ( 2) the prepared starch granules are fragmented and disperse evenly; ( 3) the degree of substitution of OSA modified starch decreases as acid hydrolysis time prolongs; ( 4) the prepared Pickering emulsion,which uses the small-particle starch as the emulsifier,possesses a stability increasing with the decrease of particle size; ( 5) the emulsifying index of Pickering emulsion decreases with the prolonging of storage time and stabilizes in the end; and ( 6) the Pickering emulsion prepared with the starch after an acid hydrolysis for 48 h or 72 h possesses good stability,for instance,the emulsifying index achieves 60% even after a storage for 30 d.

Physicochemical Properties of Highly-Substituted Carboxymethyl Starch

ZHANG Ben-shan GUO Chen-feng LI Wei-min WANG Jian-ping

2017, 45(3):  111-116.  doi:10.3969/j.issn.1000-565X.2017.03.016

Asbtract ( 244 )   PDF (804KB) ( 505 )  

Related Articles | Metrics

Firstly,carboxymethyl starches ( CMS) with high degree of substitution ( DS) were prepared from native corn starch via a reaction in aqueous alcohol.Secondly,the physicochemical properties of the prepared starches were measured and analyzed by means of FT-IR,SEM,XRD,DV-I Prime rotary viscometry and ultraviolet spectrophotometry,which were then compared with those of native corn starch.The results indicate that ( 1) after the reaction,native corn starch is successfully carboxymethylated with a DS of more than 1. 0,and starch granules are rough,swollen and cracked,with huge burst holes in the their center; ( 2) the crystalline sharp peaks of CMS damage and even disappear after the carboxymethylation; and ( 3) in comparison with native corn starch,CMS possesses much better transparency and freeze-thaw stability that are both positively related to DS.

Investigation into Sugar /Water Hot Gelatinization Mechanism of Chinese Yam Starch by Means of Integral Optical Density Method

LI Qian GAO Qun-yu

2017, 45(3):  117-124.  doi:10.3969/j.issn.1000-565X.2017.03.017

Asbtract ( 264 )   PDF (1537KB) ( 266 )  

Related Articles | Metrics

Firstly,an image analysis method,namely integral optical density ( IOD) method,was combined with the model of response difference of crystallite change ( MRDCC) to analyze the gelatinization property of Chinese yam starch.Then,XRD was adopted to discuss the crystalline structure of the starch.Finally,the effects of concentration and sugars variety on the gelatinization process were analyzed.The results demonstrate that ( 1) Chinese yam starch is in a C-type structure; ( 2) MRDCC shows double peaks that correspond to the gelatinization process of types B and A paramorph; ( 3) different sucrose concentrations as well as various monosaccharide,disaccharide,trisaccharide and tetrasccharide have different effects on the gelatinization process; ( 4) the inhibition effect due to sugar shows good correlation with dynamic hydration number,concretely,the more the equatorial hydroxyl number ( e-( OH) ) of sugar molecules,the stronger the inhibition effect; ( 5) sugar-water hydration effect is also related to the size and three-dimension structure of sugar molecules; and ( 6) a helical structure bonded through hydrogen bond is ideal for tetrasccharide because this structure weakens the hydratability,which means that the gelatinization inhibitiion effect of stachyose is smaller than that of raffinose.

Effects of Pulsed Electric Field on Preparation and Structure of Starch-Zn Complex

CHEN Shan XIE Zheng HONG Jing ZENG Xin-an HAN Zhong

2017, 45(3):  125-131.  doi:10.3969/j.issn.1000-565X.2017.03.018

Asbtract ( 252 )   PDF (2381KB) ( 223 )  

Related Articles | Metrics

At various pulse electric field ( PEF) strength ( 0. 0,2. 5,5. 0,7. 5,10. 0 and 12. 5kV/cm) and pulse wave number ( 0,180,360,540,720 and 900) ,rice starch-Zn complex was prepared via the reaction of rice starch with zinc acetate ( 0. 001mol /L) ,and the Zn content in the complex was determined,finding that,under the optimal conditions for preparing rice starch-Zn complex,namely a PEF strength of 7. 5 kV/cm and a pulse wave number of 360,the zinc content of the sample increases from 468. 24 mg /kg to 738. 62 mg /kg,which means that PEF can promote the reaction of Zn with starch.Besides,the optical characteristics and structural properties of rice starch-Zn complex were investigated by means of SEM,IR,X-ray photoelectron spectrometry and XRD.The results show that zinc acetate reacts with the hydroxyl groups of glucose units on rice starch; and that,with the increase of PEF strength and pulse number,the surface of rice starch particle is damaged,the relative degree of crystallinity is decreased,the coordination reaction is promoted and the amount of rice starch-Zn complex is increased.

Effect of Enzymatic Interesterification on Physicochemical Properties of Fast-Frozen Special Oil

ZHU Ting-wei LIU Qin ZONG Min-hua LI Bing WU Hong

2017, 45(3):  132-137,152.  doi:10.3969/j.issn.1000-565X.2017.03.019

Asbtract ( 225 )   PDF (1656KB) ( 288 )  

Related Articles | Metrics

Firstly,an enzymatic interesterified blend was made with palm stearin ( PS) and soybean oil ( SO) ( with a PS-to-SO mass ratio of 7∶ 3) by the catalysis of immobilized lipase.Then,the fast-frozen special oil-based enzymatic interesterified blend ( EIFSO) and the fast-frozen special oil-based physical blend ( PBFSO) were prepared in laboratory scale,respectively.Finally,the molecular structure and physical properties of EIFSO and PBFSO were comparatively investigated.The results show that ( 1) enzymatic interesterification has no significant influence on fatty acid content,but may result in the decrease of high-melting and low-melting triacylglycerol content of fast-frozen special oil; ( 2) the solid fat content ( SFC) of EIFSO is lower than that of PBFSO at 0 ～ 45℃ and the hardness of EIFSO is smaller than that of PBFSO except at 30℃; and ( 3) in comparison with PBFSO,EIFSO is of higher β' crystal content,smaller crystals and denser crystal network structure.

Construction of Casein Nanoparticles for Quercetin Delivery System

YOU Li-jun LIU Min LIN Lian-zhu ZHAO Mou-ming

2017, 45(3):  138-145.  doi:10.3969/j.issn.1000-565X.2017.03.020

Asbtract ( 194 )   PDF (681KB) ( 292 )  

Related Articles | Metrics

Self-assembled casein nanoparticleswere prepares via acid induction and thermal induction,and aquercetin delivery system was constructed.Then,the interaction mechanism and the effect of quercetin on casein conformation were investigated by means of fluorescence spectroscopy and dynamic light scattering.The results indicate that ( 1) quercetin can quench the intrinsic fluorescence of casein; ( 2) casein nanoparticles at pH5. 5 after heat treatment possesses the strongest binding with quercetin; ( 3) the interaction between quercetin and casein is a spontaneous and exothermal process,and the main force in this process is hydrogen bonding; ( 4) the interaction between quercetin and casein nanoparticles has no great effect on the self-assembly characteristics of casein; ( 5) quercetin after a 30-day storage keeps stable; and ( 6) casein nanoparticles after a heat treatment at pH5. 5 is of the highest retention rate up to 75. 38%.

Investigation into Identification，Antioxidant Activity and Anti-Hemolysis Activity of Phenolics Compounds in High Acid Apple

SUN Wei-zheng YANG Kun LIN Lian-zhu ZOU Ying ZHAO Mou-ming

2017, 45(3):  146-152.  doi:10.3969/j.issn.1000-565X.2017.03.021

Asbtract ( 274 )   PDF (267KB) ( 367 )  

Related Articles | Metrics

The composition of a kind of high acid apple from Xinjiang,China,namely Hongxun Ⅰ ( red flesh) ,was determined,finding that the high acid apple is of a titratable acidity of 1. 31%,a total phenolics and anthocyanins content in an order of peel ＞ pomace ＞ whole fruit ( with a total phenolics content of ( 2026. 88 ±28. 30) mg /kg and a total anthocyanins content of ( 244. 92 ± 13. 88) mg /kg) ＞ flesh.Then,ten phenolic compounds including chlorogenic acid,phlorizin,epicatechin,ferulic acid,myricetin,catechins,vanillic acid,p-coumaric acid,rutin and caffeic acid were indentified by means of high-performance liquid chromatography,finding that the high acid apple is rich in chlorogenic acid ( ( 391. 19 ± 18. 37) mg /kg) ,epicatechin ( ( 698. 82 ± 17. 21) mg /kg) and catechins ( ( 570. 30 ± 9. 21) mg /kg) .Moreover,the antioxidant and anti-hemolysis activities of the high acid apple were determined by means of DPPH radical scavenging activity assay,oxygen radical absorption capacity assay and AAPH-induced oxidative damage human erythrocytes model,finding that the peel,flesh,pomace and whole fruit of the high acid apple all exhibit excellent antioxidant activity and that the anti-hemolysis activities are indicative of the following order: peel ＞ pomace ＞ whole fruit ＞ flesh.All these above-mentioned statements indicate that,as a valuable product,high acid apple is worth processing and developing.