Home > Highlights of Sciencepaper Online > Current Issue

 
 
 
ISSN 1674-2850
CN 11-9150/N5
 
Current Issue
Previous Issue  |  Next Issue
August 30,2016
Volume 9,Issue 16
Pages -
Subject Area:Image Processing,Autocontrol Technology,Communications Technology,Computer Applications,Computer Software,Computer Network,Computer Architecture
 
Title: Development of a 3D virtual city system for waypoint navigation
Authors: JIA Zhen, LI Jianqing, XIAO Yong
PP: 1726-1732
Abstract: As the real urban environment is very complex and uncontrollable, this paper developed a three-dimensional virtual city system based on VC-OpenGL. In order to assist pedestrians to navigate in unknown places, this simulation system would provide them with real-time spatial navigation information. The city model of 3ds format is exported from Google SketchUp, mapped again in 3Ds Max, and displayed in the OpenGL program frame. The user roams in the virtual urban scenario in the first person, and receives the dynamic spatial information from the simulation system, which makes him follow the predefined route from the origin to the destination. This system also provides the user with the fixed-point aerial view and the roaming timer, and can save and display the user’s roaming routes. Experimental results indicate that, with the aid of different types of spatial navigation information, the user can navigate to the destination successfully in a strange urban environment.
Keywords: other subjects of information processing technology; virtual city; navigation information; target orientation; waypoint; route deviation
 
Title: Multi object detection and tracking based on ship monitoring video
Authors: PENG Jingjing, HUANG Chaobing
PP: 1719-1725
Abstract: A multi moving target detection and tracking algorithm is proposed for ship monitoring video. Firstly, the extracted image sequences are filtered to remove the noise. After the target is detected by the frame difference method, the morphological operation and the analysis of the connected domain are carried out in order to overcome the defect of the frame difference method for the detection of long or slow moving targets. Finally, Kalman filter is used to track the feature of the hull. The method can effectively solve the occlusion problem, and has strong robustness and applicability.
Keywords: image processing; computer vision; object detection and tracking; frame difference method; connected domain labeling; Kalman filter
 
Title: A method for painting assurance based on binary features
Authors: ZHOU Haode, LIU Jianguo
PP: 1712-1718
Abstract: The goal of this paper is to study painting assurance by binary-feature-based image matching. With feature-based matching technique, we match the recorded and current images of the artwork, to judge whether these two sets of images come from the same artwork or not. Painting assurance based on handheld and platform imaging device, means to sample the painting with platform when records, and to identify based on the images of poor quality from handheld device. Because the images from handheld device are of low quality, normal methods cannot get a satisfactory matching result. This paper proposes an assurance method with ORB feature and BRISK feature. Experimental results show that the multi-feature fusion method can solve the matching problem of images of different quality, and the painting assurance result is of high reliability.
Keywords: image processing; painting assurance; binary feature; multi-feature fusion
 
Title: Selection of p using ART-TVp on CT image reconstruction
Authors: YU Chunyu, SUN Jingyang, MIAO Yajian
PP: 1705-1711
Abstract: Algebraic reconstruction technique (ART)-TVp based image reconstruction, using either full projections or partly projections, is studied. All projections data for image reconstruction is from the shepp-logan phantom. The p value can affect the image quality of the image reconstruction and p=0.5 is found to be the best value for ART-TVp image reconstruction. Then ART-TVp setting p=0.5 is compared with the other two traditional algorithms, filtered-back projection (FBP) and ART, on a computered tomography (CT) medical image reconstruction. Conclusions are ART-TVp is superior for it can reconstruct CT image from both full projections and partly projections, especially when p=0.5, its reconstructed image is with higher quality than FBP and ART.
Keywords: image processing; image reconstruction; norm; total variance; image quality; image noise
 
Title: A fast measuring method for response time of auto-darkening welding filter
Authors: LIU Shiyi1, LIU Qinggang1, FAN Zhiguo1, GUO Dehua2, GUO Ya2, HU Xiaotang1
PP: 1698-1704
Abstract: In order to improve the situation of lacking effective measuring equipment for the response time of auto-darkening welding filters in China, a fast and automatic measuring system for response time was developed. Microcontroller was used creatively as the core processor of the measuring system, improving the signal acquisition and processing speed, making the system compact and easy to use. A wide measuring range of 0.013 to 30 ms, along with a small repeatability error of 0.008 ms, was achieved, greatly improving the measuring accuracy of response time compared to similar systems. The developing of this measuring system will provide technical support for product quality supervision and testing standardization of auto-darkening welding filters, and will help to reduce the occurring of welder’s occupational disease.
Keywords: mechatronics technology; welding filter; auto-darkening; response time; measuring system; microcontroller
 
Title: Design of networked predictive controllers for permanent magnet synchronous motor
Authors: HU Jibao, ZHANG Yan
PP: 1689-1697
Abstract: The linear model was presented to improve stability of permanent magnet synchronous motor (PMSM). And it was used to design the speed control strategy with generalized predictive control (GPC) algorithm. For PMSM control system with slowly varying parameters, a generalized predictive self-tuning algorithm was proposed. A networked compensation algorithm was presented to compensate data packet loss and time delay based on four time sequence of the actuator. And it could also solve the problem of out of order. An improved compensation algorithm took a weighted average of the predictive control increment vector in the past. And it was presented to compensate the continuous data loss that long delay produced. So, the new predictive compensation algorithm was used to design networked predictive controllers for PMSM. Simulation results show the improved networked predictive controllers can be well adapted to given speed variation and improve performance of the control system.
Keywords: automatic control technology; networked control system; generalized predictive control; permanent magnet synchronous motor; compensation algorithm
 
Title: Low sidelobe beamforming algorithm on nonuniform linear array
Authors: LIAO Yanping, SUN Yushuang
PP: 1681-1688
Abstract: Simulation experiments are carried out on the example of nonuniform linear array. In the first part, the notch noise-field method is used to reduce the sidelobe. Based on the maximum signal to interference and noise ratio (SINR) criterion, a large amount of interference is applied to the field of the sidelobe, and the power of the interference is corrected to obtain the pattern with the uniform low sidelobe. In the second part, the optimization theory is used to reduce the sidelobe by adding some sidelobe constraints. By comparing the notch noise-field method with the convex optimization algorithm, the conclusion is drawn that the second one is better. Also, the robustness is studied. At last, a new hybrid algorithm is proposed, which can further reduce the sidelobe by combining the notch noise-field method with the convex optimization, and its robustness is discussed.
Keywords: communication technology; array signal; beamforming; low sidelobe; notch noise-field; convex optimization
 
Title: A hybrid genetic algorithm about logistics distribution on ordering platform
Authors: LI Huibin1, MA Jianxin1, ZHANG Chengwen2
PP: 1672-1680
Abstract: In order to improve the quality of delivery service on the O2O (online to offline) mode’s ordering platform, the logistics scheduling vehicle problem was studied. First of all, corresponding mathematical model was put forward. Then an insertion heuristic algorithm was used to solve the vehicle allocation problem and hybrid genetic algorithm was used to improve vehicle routing. In the end, the test proved that the hybrid genetic algorithm is better than genetic algorithm.
Keywords: computer application; vehicle routing problem; genetic algorithm; logistics scheduling
 
Title: Bi-directional signal transmission system based on ZigBee technology
Authors: JIN Wanting, LIU Xirui, LI Zhen, GUO Fang, CHANG Wennan, DUAN Feng
PP: 1661-1671
Abstract: Animal-robots have an extremely important practical significance in many fields due to their unique flexibility and adaptability to the environment. This paper introduces a part of the wireless transmission of animal-robots experiment. This experiment uses two pieces of CC2530 development board and one MPU6050 module to realize the bi-directional transmission of the motion data collected by MPU6050 and the control command sent by experimenter and eventually realize periodic switching of the transmit and receive functions in both of PC end and rat end to accomplish the expected target.
Keywords: computer application; ZigBee; Basic RF; bi-directional transmission
 
Title: Image classification algorithm based on saliency detection and visual vocabulary
Authors: WEN Jing, ZHAO Xue
PP: 1654-1660
Abstract: In this paper, an image classification algorithm is proposed to deal with the blind image classification, which means the content of image is unknown, but the classification task could be guided by human visual perception. Firstly, the salient regions are extracted from natural images, then the speed-up robust feature (SURF) of the salient region is computed, after that, the visual vocabulary of the salient region is built up, the classification can be worked out by 1-norm support vector machine’ s sparse property. The experimental results show the efficiency and effectiveness of the presented approach.
Keywords: computer application; image classification; Gist; speed-up robust feature; saliency segmentation; multiple instance learning
 
Title: Research on node localization of dual-microphone array
Authors: YANG Shuailing, JIN Naigao
PP: 1647-1653
Abstract: This paper presents a event-driven localization method for the dual-microphone node localization problem. Firstly, the proposed method generate the driven event by controlling the speaker in the each node. Then the node sequence is achieved by the time of arrival, and binary 0/1 information is obtained on each node by time delay estimated based on sparse Fourier transform (SFT). Finally the node position and orientation are estimated by processing multiple node sequences and binary 0/1 sequences. Simulation and actual experiments are carried out to verify the effectiveness of the proposed method.
Keywords: computer softwove node localization; wireless sensor network; dual-microphone; event driven localization
 
Title: On the delay less than a round trip time at input-queued switches with distributed schedulers
Authors: ZHOU Qian, HU Bing
PP: 1636-1646
Abstract: This paper proposes a novel request prediction (RP) mechanism. By adopting the RP mechanism, many existing monolithic iterative scheduling algorithms can be easily extended to the distributed implementation and cut down the minimal queuing delay to less than an round trip time (RTT). To be more specific, an idle output j can smartly predict the request en route and issue a grant in advance. If this prediction is precise, the delay can be cut down. Otherwise, the performance/bandwidth is not suffered yet as output j is originally idle. RP is very simple to implement with complexity O(1). Simulation results show that RP optimizes the existing algorithms round robin with longest queue first (RR/LQF) and highest rank first with request compression (HRF/RC) to achieve the best performance under light and moderate traffic. Last but not least, the analytical results for the performance of RP are provided.
Keywords: Input-queued switch; Distributed iterative scheduling algorithms; Round trip time
 
Title: Implementation and verification of routing algorithm in network on chip based on Dijkstra algorithm
Authors: MA Xiantong, ZHU Yongxin
PP: 1627-1635
Abstract: By abstracting network on chip (NOC) nodes into points, weights are attached to the network channels based on the degree of congestion. Then the problem of routing selection is converted into a problem of finding the shortest paths between nodes in graph. According to the characteristics of NOC topologies, the calculation process of Dijkstra algorithm is optimized, in which adjacent nodes ID are calculated directly and the minimum congestion’s node ID is got within one clock cycle by parallel comparison in unvisited nodes. It is designed with the verilog hardware description language (HDL). A testbench is created to verify it. The simulation result is compared with the C model. Finally, the algorithm is implemented on a Xilinx Zynq series field programmable gate array (FPGA).
Keywords: computer architecture; network on chip; shortest path algorithm; verilog hardware description language; field programmable gate array
 
Title: Opinion target extraction using stacked denoising auto-encoders
Authors: LI Juan, GAO Zhiqiang
PP: 1617-1626
Abstract: We propose a new model based on stacked denoising auto-encoders (SDAE), which only uses word embeddings as input features. The original SDAE model is a classification model, however, we regard the opinion target extraction task as a sequential labeling task in this paper. Thus, in order to adapt the original classification model to a sequential labeling model, we propose to use a greedy algorithm to extend the SDAE model. Experimental results demonstrate that the proposed model rival the top performing constrained model which only uses the provided training data in SemEval-2014.
Keywords: natural language processing; opinion target extraction; stacked denoising auto-encoders; opinion mining