Most target tracking is based on a lot of samples training to build the model of the target, which is then carried on the tracking processing. This will need to choose a lot of tracked target samples for learning and ...
详细信息
Learning-based multi-task models have been widely used in various scene understanding tasks,and complement each other,i.e.,they allow us to consider prior semantic information to better infer *** boost the unsupervise...
详细信息
Learning-based multi-task models have been widely used in various scene understanding tasks,and complement each other,i.e.,they allow us to consider prior semantic information to better infer *** boost the unsupervised monocular depth estimation using semantic segmentation as an auxiliary *** address the lack of cross-domain datasets and catastrophic forgetting problems encountered in multi-task training,we utilize existing methodology to obtain redundant segmentation maps to build our cross-domain dataset,which not only provides a new way to conduct multi-task training,but also helps us to evaluate results compared with those of other *** addition,in order to comprehensively use the extracted features of the two tasks in the early perception stage,we use a strategy of sharing weights in the network to fuse cross-domain features,and introduce a novel multi-task loss function to further smooth the depth *** experiments on KITTI and Cityscapes datasets show that our method has achieved state-of-the-art performance in the depth estimation task,as well improved semantic segmentation.
To solve the problem of power system reactive optimization efficiently, a new method, which combines particle swarm optimization algorithm with dynamic weight, was presented. The improved PSO algorithm, which is based...
详细信息
ISBN:
(纸本)9781846260681
To solve the problem of power system reactive optimization efficiently, a new method, which combines particle swarm optimization algorithm with dynamic weight, was presented. The improved PSO algorithm, which is based on dynamic weight, is simple in application and quick in convergence. The weight is changed in every loop according to the swarm evolution degree and aggregation degree factor, which ensures population diversity and avoids premature convergence effectively. Then the algorithm is applied in reactive optimization of power system, compared with PSO, the optimization result s of IEEE-30-bus show that the algorithm has a good stable searching capacity and good parallel efficiency.
The realization of the parameter estimation of chirp signals using the fractional Fourier transform (FRFT) is based on the assumption that the sampling duration of practical observed signals would be equal to the ti...
详细信息
The realization of the parameter estimation of chirp signals using the fractional Fourier transform (FRFT) is based on the assumption that the sampling duration of practical observed signals would be equal to the time duration of chirp signals contained in the former. However, in many actual circumstances, this assumption seems unreasonable. On the basis of analyzing the practical signal form, this paper derives the estimation error of the existing parameter estimation method and then proposes a novel and universal parameter estimation algorithm. Furthermore, the proposed algorithm is developed which allows the estimation of the practical observed Gaussian windowed chirp signal. Simulation results show that the new algorithm works well.
Aiming at the defects of routine settlement measurement methods, such as complicated procedures, time-consuming and labor-intensive, high cost and low measurement accuracy, based on the analysis of existing engineerin...
详细信息
Aiming at the defects of routine settlement measurement methods, such as complicated procedures, time-consuming and labor-intensive, high cost and low measurement accuracy, based on the analysis of existing engineeringmeasurement technical requirements and specifications, a multi-point high precision and high efficiency based on laser reference is proposed. The automatic building settlement real-time monitoring system program gives the principle and system model of single-point settlement observation. The model of multi-point scanning settlement monitoring system and the model of multi-point network settlement monitoring system are designed, and their advantages and disadvantages are analyzed. We focus on the networked multi-point settlement monitoring system for network cumulative error analysis, and propose related evaluation and correction methods. The hardware schematic and software block diagram of the laser reference measurement and measurement system of the single point settlement acquisition system are given. Finally, the risk of subsidence state is quantitatively evaluated based on multi-point settlement monitoring data. The measurement error of this method is less than 300 μm, which can realize the monitoring and evaluation of the overall settlement.
To reduce the impacts of harmonics and noises on frequency measurements, a novel method of frequency tracking is presented based on improved EKF algorithm and wavelet transform. Firstly, the power signal was preproces...
详细信息
Density-based algorithm for discovering clusters in large spatial databases with noise(DBSCAN) is a classic kind of density-based spatial clustering algorithm and is widely applied in several aspects due to good perfo...
详细信息
Density-based algorithm for discovering clusters in large spatial databases with noise(DBSCAN) is a classic kind of density-based spatial clustering algorithm and is widely applied in several aspects due to good performance in capturing arbitrary shapes and detecting outliers. However, in practice, datasets are always too massive to fit the serial DBSCAN. And a new parallel algorithm-Parallel DBSCAN(PDBSCAN) was proposed to solve the problem which DBSCAN faced. The proposed parallel algorithm bases on MapReduce mechanism. The usage of parallel mechanism in the algorithm focuses on region query and candidate queue processing which needed substantive computation resources. As a result, PDBSCAN is scalable for large-scale dataset clustering and is extremely suitable for applications in E-Commence, especially for recommendation.
In order to realize high-precision and real-time flow measurement, in this paper according to the measurement principle of ultrasonic transit time method, a low-power microprocessor MSP430 is used as the core of this ...
详细信息
In this paper, we propose two algorithms based on iterative optimization for UAV path planning. In the first one, a nonlinear programming (NLP) problem of unmanned aerial vehicle (UAV) path planning is transformed int...
详细信息
In order to monitor the crystallization process of L-glutamic acid online, a real-time detection method based on non-invasive image analysis has been proposed to obtain in-situ images, and a deep-learning based networ...
详细信息
暂无评论