Time filter

Source Type

Zhuo Y.,ChangZhou Textile Garment Institute
Journal of Computational and Theoretical Nanoscience | Year: 2016

The conventional Bag of Visual Words model (BoVW) only uses the image feature domain and neglects the contextual semantic information of the image. This paper proposes an image scene classification algorithm based on the contextual semantic information. Firstly, based on the conventional BoVw, we introduce the Markov Random Field model to modeling the image contextual semantic information. Secondly, we use a density-based adaptive selection method to choose the number of the of the optimal probabilistic latent semantic analysis model. At last, we use the Support Vector Machine (SVM) to perform scene classification. The experimental results show that based on the conventional Bag of Visual words model, our algorithm can effectively utilize the contextual semantic information of images and improve image scene classification performance and enhance the classification accuracy. Copyright © 2016 American Scientific Publishers All rights reserved.


Lu Y.,Changzhou Textile Garment Institute
Lecture Notes in Electrical Engineering | Year: 2013

In art and visual perception, grace, ham said, "design form will never be mechanical copy perception of the material, but a creative explain reality, and beautiful pictures have already explained is rich imagination, create and accuracy". In the conversion of the art of design into visual music, it is necessary to focus on the new dynamic audio technology, expression and the way of forming the element, and new esthetic idea for the audience in the actual application. © 2013 Springer-Verlag.


Xing W.,Changzhou Textile Garment Institute
Lecture Notes in Electrical Engineering | Year: 2013

The students who graduate from an artistic design program often receive cold eyes from people at the employment market full of fierce competitions, because they are in shortage of the practical experience. Through an investigation, it is found that the reason for this phenomenon is that the model of colleges and universities to train the art design personnel has been unable to the market demands. The update of the training model of a school is a process for a long time actually, and simultaneously the market demands on personnel changes more and more quickly. © 2013 Springer-Verlag.


Zhuo Y.,Changzhou Textile Garment Institute
Open Automation and Control Systems Journal | Year: 2015

With the development of multimedia technology and computer network, the number of available images increases with an explosive speed. But the technology also brings some trouble to its users, sometimes it’s very difficult for us to find some details that very important from a huge amount of available data. At this time, image scene and emotion categorization technologies are required urgently. The purpose of emotional image classification is that we hope the computer can express the emotion reaction when observing the image, and classify the images into the different emotional categories automatically. The process of scene image classification is that how to make computer systems to classify the image sets automatically which contain semantic information, according to the visual perception mechanism of human. Here, for the scene categorization problem based on the visual words, the dissertation presents a novel learning framework to design discriminating semantic visual words. At last, for the emotion categorization of natural scene images, the dissertation presents an emotion categorization using Affective-probabilistic Latent Semantic Analysis model based on the visual cognitive theory. © Yang Zhuo.


Zhang R.,Changzhou Textile Garment Institute
Lecture Notes in Electrical Engineering | Year: 2015

Prediction of drug–target interactions (DTIs) plays an important role in the drug discovery process. Unfortunately, experimental determination of interactions between drug compounds and target proteins remains a challenging task. Current available heterogeneous data motivate us to develop an effective data integration approach to exploit intrinsic correlations of known interactions between drugs and targets and to predict new interactions. We propose an ensemble learning approach to integrate previously developed methods and to improve the prediction performance. In particular, our algorithm employs a stacking framework, which uses a support vector machine (SVM) classifier as the meta learner to achieve better prediction results. © Springer International Publishing Switzerland 2015.


Wang Z.-M.,Changzhou Textile Garment Institute
Communications in Computer and Information Science | Year: 2012

Network and multimedia technologies combine the graph, writing, sound, animation, phantom and so on with the course content organically, enabling the learners to use multiple sense organs. Learners of visual, auditor and kinesthetic style who prefer different sensation channels can all obtain effective information stimulation, so teaching may be adapted to learners of different cognitive styles. As our country has just started research and utilization of personalized teaching, this paper, through the research of an important individual difference- cognitive style in foreign language study, attempts to study how to aim at cognitive style for implementation of personalized foreign language teaching strategy with the aid of network and multimedia technology. © 2012 Springer-Verlag Berlin Heidelberg.


Yuan X.,Changzhou Textile Garment Institute
Advanced Materials Research | Year: 2013

The effects of poly(vinyl pyrrolidone)(PVP) on dispersing carbon black(CB) particles were investigated by measuring Z-average particle size, Zeta potential and centrifugal stability of CB dispersions. Addition of PVP in a dispersing medium significantly reduces the size of suspended CB particles, especially in water. The dispersing efficiency of PVP was found to have been enhanced by adding an anionic surfactant, sodium methylenedinaphthalene disulphonate(NNO) in the aqueous media. The particle size of CB dispersed with PVP and NNO was 175. 7 nm, significantly smaller than that dispersed with PVP and sodium dodecyl sulfate (SDS). The performance of PVP-encapsulated CB particles and that of PVP-adsorbed CB particles were also compared. The particle size and Zeta potential of PVP-encapsulated CB particles were similar to those of PVP-adsorbed CB particles using ultrasonic method, but the centrifugal stability of PVP-encapsulated CB dispersions was significantly improved. The relative absorbency of PVP/CB dispersions was increased from 42. 0% to 63. 7%. However, the PVP-encapsulated layer can be destroyed by prolonged ultrasonic treatment because the ultrasound over a long period of time can not only break up the flocculation bridge in the crosslinking matrix of PVP, but also flake off the PVP shells on the surface of CB particles. © (2013) Trans Tech Publications, Switzerland.


Hu Y.,Changzhou Textile Garment Institute
Proceedings - 14th International Symposium on Distributed Computing and Applications for Business, Engineering and Science, DCABES 2015 | Year: 2015

In order to solve the current smart phone location-based services LBS (Location Based Service) class of mobile applications in a small area of the environment can not be accurately provide location services issues, Presented to the Campus of LBS system design based on indoor and outdoor location. First introduced system architecture design, and detailed analysis of the Mobile GIS, GPS, system design based on the location of the fingerprint localization algorithm WiFi signal strength values of key technologies, Finally, test each module authentication system. Test results show that the system can achieve terminal location, indoor location services, campus navigation, map services and other functions, and convenient operation, with the feasibility and practicality. © 2015 IEEE.


Zang J.,Changzhou Textile Garment Institute
Advanced Materials Research | Year: 2014

Based on 1-amino-4-hydroxy-2-phenoxy anthraquinone (Disperse Red FB) and ethylene glycol phenyl ether as raw materials, polyethylene glycol (PEG) as catalyst, condensation in alkaline conditions. The optimum reaction temperature is about 140°C, 4 h reaction times, 9g catalyst, the ratio of recovery and fresh consumptions of ethylene glycol phenyl ether is controlled in 1:1 (ml). The reaction end point is easy to arrive in these conditions, and the yield of disperse peach SE-RL (Disperse Red 191) is the best, about 79.2%. © (2014) Trans Tech Publications, Switzerland.


Zang J.,Changzhou Textile Garment Institute
Advanced Materials Research | Year: 2014

A preparation of a kind of green water-based metal cleaning agent has been described in this paper. The reaction is using oleic acid and triethanolamine, by add the corresponding surface active agent. After a series of performance test, all kinds of metal has good corrosion resistance and cleaning effect, and no pollution to the environment. © (2014) Trans Tech Publications, Switzerland.

Loading Changzhou Textile Garment Institute collaborators
Loading Changzhou Textile Garment Institute collaborators