Airborne vehicle detection and tracking systems equipped on unmanned aerial vehicles (UAVs) are difficult to develop because of factors like UAV motion, scene complexity and so on. In this paper, we propose a new fram...
详细信息
Airborne vehicle detection and tracking systems equipped on unmanned aerial vehicles (UAVs) are difficult to develop because of factors like UAV motion, scene complexity and so on. In this paper, we propose a new framework of multi-motion layer analysis to detect and track moving vehicles in airborne platform. Moving vehicles are firstly detected by registration and temporal differencing to establish motion layers. After motion layers are constructed, they are maintained over time for tracking vehicles. All vehicles are tracked by maintaining their corresponding motion layers. Our experimental results showed that compared with other previous algorithms, our method can achieve better results in terms of detection and tracking performance.
Based on the QSIM reasoning of Kuipers, comparative constraint is presented to reduce the diagnosis space. The transfer regulation of qualitative constrain is used in the simulation and reasoning of system diagnosis. ...
详细信息
Based on the QSIM reasoning of Kuipers, comparative constraint is presented to reduce the diagnosis space. The transfer regulation of qualitative constrain is used in the simulation and reasoning of system diagnosis. This arithmetic takes observed faulty state as the beginning, to diagnosis the discrepancy of variable and location of the causation according to the transfer regulation of qualitative constraint, reasons from faulty diagnosed to examine the result and delete the redundancy of diagnosis result. In the example of Condensation refrigeration system, the constraint relation is built according to qualitative difference equation. Aim at the bad efficiency of refrigeration, the mixture of air or Freon superfluously is diagnosed as the result faulty source, is consistent with the factual system.
According to the miniaturization characteristics of step impedance resonators, a step impedance bandpass filter with a center frequency of 16.55 GHz and a bandwidth of 500MHz is designed. The effects of parameter chan...
详细信息
According to the miniaturization characteristics of step impedance resonators, a step impedance bandpass filter with a center frequency of 16.55 GHz and a bandwidth of 500MHz is designed. The effects of parameter changes such as substrate resistivity and the shape of metal via etching on the performance of the filter are studied. Finally, the filter is fabricated and its performance is tested under MEMS technology.
parallel framework for NURBS-based isogeometric analysis on multi-core cpus is presented in this *** framework is composed of two parts:(1) construction of the stiffness matrix and right hand side vector; (2) a parall...
parallel framework for NURBS-based isogeometric analysis on multi-core cpus is presented in this *** framework is composed of two parts:(1) construction of the stiffness matrix and right hand side vector; (2) a parallel sparse solver.
Noise is an omnipresent phenomenon. It obscures the real behavior of dynamical system. Lots of methods are proposed to remove the noise contaminating time series. However, almost all the methods consider the noise red...
详细信息
Noise is an omnipresent phenomenon. It obscures the real behavior of dynamical system. Lots of methods are proposed to remove the noise contaminating time series. However, almost all the methods consider the noise reduction in the phase space and often sharp points are kept. Different with these methods, this paper proposes a method directly on the time series itself, considering the gauss noise feature and the smoothness of the real data, uses curve-fitting way to eliminate the sharp points. The numeral results verify the effectiveness of our method.
To enhance the reliability and the fault coverage of the test results, a local error diagnostic mechanism was introduced into dynamic protocol conformance test, a new dynamic test method was proposed, whose principle ...
详细信息
To enhance the reliability and the fault coverage of the test results, a local error diagnostic mechanism was introduced into dynamic protocol conformance test, a new dynamic test method was proposed, whose principle and correctness was analyzed on determinate FSM model. The new method took full advantage of the dynamic information generated during test to avoid the incorrect locating of local error. Also a local error diagnostic method was presented on error sub-sequence, which was not restricted to single error assumption and more practical. The experimental results showed that the method can effectively avoid the descending of the test reliability and give better fault coverage.
Artificial intelligence (AI) is transforming scientific research, including proteomics. Advances in mass spectrometry (MS)-based proteomics data quality, diversity, and scale, combined with groundbreaking AI technique...
详细信息
3D object detection is an essential perception task in autonomous driving to understand the environments. The Bird's-Eye-View (BEV) representations have significantly improved the performance of 3D detectors with ...
3D object detection is an essential perception task in autonomous driving to understand the environments. The Bird's-Eye-View (BEV) representations have significantly improved the performance of 3D detectors with camera inputs on popular benchmarks. However, there still lacks a systematic understanding of the robustness of these vision-dependent BEV models, which is closely related to the safety of autonomous driving systems. In this paper, we evaluate the natural and adversarial robustness of various representative models under extensive settings, to fully understand their behaviors influenced by explicit BEV features compared with those without BEV. In addition to the classic settings, we propose a 3D consistent patch attack by applying adversarial patches in the 3D space to guarantee the spatiotemporal consistency, which is more realistic for the scenario of autonomous driving. With substantial experiments, we draw several findings: 1) BEV models tend to be more stable than previous methods under different natural conditions and common corruptions due to the expressive spatial representations; 2) BEV models are more vulnerable to adversarial noises, mainly caused by the redundant BEV features; 3) Camera-LiDARfusion models have superior performance under different settings with multi-modal inputs, but BEV fusion model is still vulnerable to adversarial noises of both point cloud and image. These findings alert the safety issue in the applications of BEV detectors and could facilitate the development of more robust models.
With the popularity of small storage embedded devices, recent researches on deep convolutional neural networks no longer only focus on improving the classification accuracy but start to attach great importance to the ...
With the popularity of small storage embedded devices, recent researches on deep convolutional neural networks no longer only focus on improving the classification accuracy but start to attach great importance to the size of convolutional neural networks. This paper proposes a small deep convolutional neural network for vehicle type classification, based on the Fire module and the residual module. The Fire module is applied to maintain a small number of parameters. The residual module is employed to accelerate the training process and improves the classification accuracy. Experimental results show that our model achieves 95.57% accuracy of vehicle type classification on an open source dataset BIT-Vehicle. This outperforms previous researches and some classical convolutional neural networks such as AlexNet, ResNet-32 and SqueezeNet. What is more, the size of our model is only 5.3 MB, which has 40X fewer parameters than AlexNet and has 15X fewer parameters than ResNet-32.
3D object detection is an essential perception task in autonomous driving to understand the environments. The Bird’s-Eye-View (BEV) representations have significantly improved the performance of 3D detectors with cam...
详细信息
暂无评论