The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
An interesting target detection framework with transferred deep convolutional neural network (CNN) is proposed. For CNN, many labeled samples are needed to train the multi-layer network. However, for target detection tasks, only few target spectral signatures are available, or they are unknown in anomaly detection. In this work, we employ a reference data and further generate pixel-pairs to enlarge...
It would be difficult and stressful for a single operator to operate an underwater manipulator using his/her both hands in deep sea environments while the operator has to monitor or manipulate additional equipment. In order to reduce the operating pressure and make full use of the operator potentials, in this paper we propose a control strategy for operating the underwater manipulator via P300 brainwaves,...
This paper concerns on inefficiency or even failure in secret key generation caused by the imperfect channel state information. We propose a secret key generation scheme based on wavelet analysis. Firstly, the channel estimates are pre-processed by wavelet analysis to improve the correlation. Secondly, to ensure the randomness of the secret keys, an adaptive equal probability quantization approach...
In this paper, convolutional neural networks (CNNs) is employed for remote-sensing scene classification, which fully utilizes the semantic features extracted from the images while ignoring some traditional features. Consider the limited labeled samples, CaffeNet model as the pre-trained architecture is adopted. By fine-tuning the pre-trained models, the proposed method is expected to be robust and...
Collaborative representation-based classification with distance-weighted Tikhonov regularization (CRT) has offered high accuracy and efficiency. Due to its per-pixel classification nature without a training step, this paper develops a parallel implementation by using compute unified device architecture (CUDA) on graphics processing units (GPUs). To further improve classification accuracy, local binary...
With the rapid development of technology, Internet has changed people's lives and online community has become a comprehensive platform for exchanging ideas and shopping. This study develops and tests an environmental psychology model of online impulse purchase. The model captures the moderating effect of search and experience product type on the relationship between the online impulse buying intention...
In hyerspectral remote sensing community, sparse representation based classification (SRC) is a novel concept — a testing pixel is linearly represented by labeled data, and weight coefficients are often solved by an ℓ1-norm minimization. In this work, an extension of SRC is proposed by imposing an adaptive similarity measurement between the testing pixel and labeled data on the ℓ1-norm penalty, named...
In hyerspectral image analysis, representation-based classification is a novel concept — a testing pixel is linearly represented by using the labeled samples. The weight coefficients can be solved by an ℓ1-norm penalty for sparse representation, or solved by an ℓ2-norm penalty for collaborative representation. In this work, a convex combination of these two representations using the elastic net model...
With the increasing number of services on the Internet, it has become a great challenge to help users find services according to their demands. Personalized recommendation technology is an effective way to solve the problem. Existing service recommendation approaches make recommendations among services with same or similar functionalities to meet the non-functional requirements of users, while the...
Software defined infrastructure greatly reduces the deployment cost of distributed applications. Many distributed applications employ message oriented middleware (MOM) for the integration of heterogeneous components and to achieve scalability and fault tolerance. The structure of a distributed application can be very complex. In addition, the asynchronous message delivery model of MOM further complicates...
We propose a novel collaborative representation based k-nearest neighbors algorithm for hyperspectral image classification. The proposed method is based on a collaborative representation computed by an ℓ2-norm minimization with a Tikhonov regularization matrix. More specific, the testing sample is represented as a linear combination of all the training samples, and the weights for representation are...
This paper presents a P300 model for controlling a humanoid robot with mind, including an off-line phase with a fixed trial number for training the model and an on-line phase with an adaptive strategy for generating commands to the humanoid robot. Our control scheme includes a procedure of acquiring P300 signals, topographical distribution analysis of P300 signals, and a classification approach to...
A method of unsupervised nearest regularized subspace is proposed for anomaly detection in hyperspectral imagery. Based on a dual window, an approximation of each testing pixel is a representation of surrounding data via a linear combination, for which the weight vector is calculated by distance-weighted Tikhonov regularization. Proposed detector returns the similarity measurement between the testing...
Power transformer applied voltage test is one of the most effective and direct way to identify power transformer insulation strength and it is also an important part of preventive test. The methodology of diagnosing the transformer winding insulation defect using the transient voltage and current in voltage regulation process is put forward in this paper. In this methodology, the transient voltage...
This paper describes the function, hardware and software design of structural defect nondestructive testing (NDT) system of a certain unmanned aerial vehicle (UAV) airframe, and analyzes the key technology in the detection system.
An ANN-based inverse scattering algorithm for estimation of the thickness and permittivity of multi-layered medium is proposed in this paper, which demonstrates the simplification and effectiveness of feature extraction of reflected signal of each layer. Neural network is adopted to investigate the nonlinear model between the feature of reflected wave and thickness and permittivity of medium. The...
A concept of local gray entropy is introduced to solve the problem of the human body detection difficulty under the dim contrast environment. This paper analyses the trait of local gray entropy, and then proposes an algorithm of human target detection based on the trait, which is according to the principle that local gray entropy could reflect the discrete level accurately, and it has no relationship...
In this paper, a usage model is proposed to simulate users' behaviors realistically in load testing of web applications, and another relevant workload model is proposed to help generate realistic load for load testing. It also demonstrates an eclipse-based load testing tool “Load Testing Automation Framework (LTAF)” which is based on these two models and can perform load testing of web applications...
The Digitally Controlled Potentiometers (DCP) is a new type of electronic device with great developing foreground, which can replace the traditional mechanical potentiometer in many fields. The programmable gain amplifier, programmable filter and others programmable analogy devices can be built using SCM and DCP through programming. Thereby it is realizable to “Set the analogy device onto the bus”...
A common method for real-time video detection in image sequences involves ??background subtraction??. The numerous approaches to this problem differ in the type of background model. This paper discuss a new background updating method based on classification. A growth template that can detect the target and interference, like noise, has put forward. This template can choose the growth direction automatically...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.