Huibin Wang

Combination of Interacting Multiple Models with the Particle Filter for Three-Dimensional Target Tracking in Underwater Wireless Sensor Networks

Tracking underwater targets is a focused application area in modern underwater defence systems. Using traditional techniques based on imaging or sensor arrays may be difficult and impractical in some mission-critical systems. Alternatively, the underwater wireless sensor networkUWSN� is able to offer a promising solution. This paper tackles the problem of accurately tracking underwater targets moving through the UWSN environment. This problem is considered nonlinear and non-Gaussian where the present solution methods based on the particle filter technique are powerful and simple to implement. For three-dimensional underwater maneuvering target tracking,the traditional particlefilter tracking algorithmmay fail to track the targets robustly and accurately. Thus, the interacting multiple model method is combined with the particle filter to cope with uncertainties in target maneuvers. Simulation results show that the proposed method is a promising substitute for the traditional imaging-based or sensor-based approaches.

Multiple Feature Fusion for Tracking of Moving Objects in Video Surveillance

Recently video surveillance techniques have been widely applied to intelligent transportation systems. Tracking of moving objects such as vehicles has become a major topic in video surveillance applications. This paper presents a multi-feature fusion model based on a particle filter for moving object tracking. The particle filter combines color and edge orientation information by a stochastic fusion scheme. The scheme randomly selects single observation model to evaluate the likelihood of some particles. The stochastic selection probability is adjusted adaptively by the uncertainty associated with a feature model. The experiment shows that the proposed method has strong tracking robustness and can effectively solve the occlusion problem.

Experimental studies on hydro-mechanical properties of metamorphic rock under hydraulic pressures

The research on the hydro-mechanical coupling of metamorphic rock is of great importance in rock engineering. The main purpose of this study was to investigate the hydro-mechanical properties of metamorphic rock based on the experimental data from triaxial compression tests. Metamorphic rock samples taken from an underground oil storage facility were tested under different hydraulic pressures using a rock servo-controlled triaxial rheology test system. The results indicated that hydraulic pressure had impact on the strength, deformation and permeability of the metamorphic rock. Based on the test results, the stress–strain behaviour and permeability evolution of the rock samples show three stages. The mechanical properties of the metamorphic rock are presented in detail for the three stages in the whole stress–strain process. The derived mechanical parameters are correlated to the division of the phases. The relationship between hydraulic pressure and axial strain, lateral strain, volumetric strain and permeability are analysed and discussed, especially the influence of hydraulic pressure on various parameters in the hydro-mechanical process. The influence of the hydraulic pressure on the failure mode of the rock under coupled conditions is described.

A research on stereo matching algorithm for underwater image

To realize the accuracy and stability of the underwater binocular image matching, this paper design a stereo matching algorithm based on Harris corner detection. Firstly, apply median and homomorphic filtering for underwater stereo image preprocessing to minimize noise and enhance image contrast. Secondly, apply the corner feature extraction and matching method based on Harris for obtaining the characteristic information of underwater target. And then apply the constraint condition to eliminate the error matched points, so as to improve the underwater image matching stability. Finally, the experiments verify the algorithm proposed in this paper, and the experimental results have shown the effectiveness of this algorithm.

Monocular Vision-Based Underwater Object Detection

In this paper, we propose an underwater object detection method using monocular vision sensors. In addition to commonly used visual features such as color and intensity, we investigate the potential of underwater object detection using light transmission information. The global contrast of various features is used to initially identify the region of interest (ROI), which is then filtered by the image segmentation method, producing the final underwater object detection results. We test the performance of our method with diverse underwater datasets. Samples of the datasets are acquired by a monocular camera with different qualities (such as resolution and focal length) and setups (viewing distance, viewing angle, and optical environment). It is demonstrated that our ROI detection method is necessary and can largely remove the background noise and significantly increase the accuracy of our underwater object detection method.

Fuzzy Integral Based Information Fusion for Water Quality Monitoring Using Remote Sensing Data

To improve the monitoring precision of lake chlorophyll a (Chl-a), this paper presents a fusion method based on Choquet Fuzzy Integral (CFI) to estimate the Chl-a concentration. A group of BPNN models are designed. The output of multiple BPNN model is fused by the CFI. Meanwhile, to resolve the over-fitting problem caused by a small number of training sets, we design an algorithm that fully considers neighbor sampling information. A classification experiment of the Chl-a concentration of the Taihu Lake is conducted. The result shows that, the proposed approach is superior to the classification using a single neural network classifier, and the CFI fusion method has higher identification accuracy.

Multi-Camera Tracking Based on Information Fusion in Video Surveillance

Recently, intelligent video surveillance system with multiple cameras is a research hotspot due to its wide application. This system has many advantages on improving the surveillance performance and extending surveillance functions. However, taking multi-camera resources and video data into account, many problems also need to explore and research. The paper designs a system that fuses tracking information from multiple cameras. The system use an approach called map-view mapping for camera calibration and present a fusion strategy to fuse multi- camera tracking data. The experiment verified the performance of the fusion method, and showed this method can effectively eliminate occlusion in overlapping region and integrate all the events from multiple cameras in the ground plane.

A Knowledge-Based Target Recognition Method For Remote Sensing Image

This paper proposed a knowledge-based target bridge detection method from remote sensing images. Firstly, the prior knowledge about bridge can supervise the low and middle image processing and analysis to detect the edges of bridges and riversides. Secondly, knowledge about the different characters of the bridge edges and riverside edges was used to wipe off the riverside edges and remain the lines of bridges, so that the purpose of recognition on bridge can be achieved. The method was tested on spot remote sensing images and the result shows it can detect the bridges exactly from images.

Image Denoising Based on Wavelet Edge Detection by Scale Multiplication

A novel image denoising scheme based on edge detection by scale multiplication in wavelet domain is presented. The dyadic wavelet transforms of B-spline at two adjacent scales are multiplied as a function to magnify the edge structures and suppress the noise. In view of this correlation, the edges of an image are determined by the dual-threshold judgment to the scale product. Apply hard thresholding to the non-edge coefficients in the high frequency subbands at each scale and retain the edge coefficients. Then reconstruct the modified coefficients and obtain the denoised image. The experiment shows that image denoising using this new method performs better than that using the traditional wavelet denoising.

Experimental Investigation of the Anisotropic Mechanical Properties of a Columnar Jointed Rock Mass: Observations from Laboratory-Based Physical Modelling

Because of the complex geological structure, determination of the field mechanical parameters of the columnar jointed rock mass (CJRM) was a challenging task in the design and construction of the Baihetan hydropower plant. To model the mechanical behaviour of the CJRM, uniaxial compression tests were conducted on artificial CJRM specimens with geological structure similar to that found in the actual CJRM. Based on the test results, the anisotropic deformation and strength were mainly analysed. The empirical correlations of evaluating the field mechanical parameters were derived based on the joint factor approach and the modulus reduction factor method. The findings of the physical model tests were then used to estimate the field moduli and unconfined compressive strengths of the Baihetan CJRM. The results predicted by physical model tests were compared with those obtained from the field tests and the RMR classification system. It is concluded that physical model tests were capable of providing valuable estimations on the field mechanical parameters of the CJRM.