In the field of face recognition and analysis, eye state detection is an essential step, which is the prerequisite and breakthrough of drowsiness estimation and auxiliary driving. This paper presents an eye state dete...
详细信息
In the field of face recognition and analysis, eye state detection is an essential step, which is the prerequisite and breakthrough of drowsiness estimation and auxiliary driving. This paper presents an eye state detection method based on Weight Binarization Convolution Neural Network(WBCNN). The weight of the network is constrained by binarization, which can limit the weight to 1 or-1, reducing the power dissipation and internal storage considerably. The human eye state features which can be extracted by convolution neural network effectively, and binary network not only contributes to reducing the storage size of the model, but also accelerates the computation. Experiments on eye state detection were conducted on the Closed Eyes in the wild(CEW) and FER2013 Databases, from which the results show that our method achieved average test accuracy of 97.41%on CEW. We used the FER2013 facial expression database for pre-training, which can make up for the lack of CEW training samples. The computational speed of non-binary is slower than binary network. Moreover, less storage capacity is required by our method.
In this paper, the object detection technology based on deep learning is applied to the assembly process of space power station simulation, which can provide assistance for the attitude adjustment and navigation of th...
详细信息
In this paper, the object detection technology based on deep learning is applied to the assembly process of space power station simulation, which can provide assistance for the attitude adjustment and navigation of the aircraft through the detection of some components. Firstly, the 3 D modeling and rendering of the space power station are carried out, on which the image dataset is collected and established. Then, based on the YOLOv3 network, we improve the structure of feature *** fusing the information of shallow and deep features, we can improve the detection ability of the network for different scale *** and quantitative experimental results show that the improved YOLOv3 network can accurately and effectively detect the key components of the Space solar power station.
With the development of single-cell sequencing technology, it is a hot research topic to identify the cell types using single-cell sequencing data, and many single-cell clustering algorithms have been developed to stu...
详细信息
With the development of single-cell sequencing technology, it is a hot research topic to identify the cell types using single-cell sequencing data, and many single-cell clustering algorithms have been developed to study this issue. These methods capture partial information of single-cell sequencing data, and obtain the different performance on the same data set. Combining these different results into one can improve the accuracy and validity. Here, we proposed ECBN, Ensemble Clustering based on B ayesian Network. ECBN can ensemble several different results of state-of-the-art single cell clustering methods, such as Seurat, CIDR, SC3 and t-SNE+k-means, and generate a more optimal clustering result through Bayesian network. Experiments are carried on the 5 single cell data sets and compared with 4 individual single cell clustering methods and 3 integrative *** size of experiment data sets ranges from 822 to 3605 and the results show that our method can achieve good ***, ECBN can also use the graphical regularization to lighten the limitation which is generated by the different basis results.
The regenerative chatter during milling seriously affects the stability of the *** paper proposes a method based on Lyapunov-Krasovski functional analysis for the stability of the milling ***,the mechanism analysis of...
详细信息
The regenerative chatter during milling seriously affects the stability of the *** paper proposes a method based on Lyapunov-Krasovski functional analysis for the stability of the milling ***,the mechanism analysis of the milling process is performed,then the state-space equation of the time-varying delay system caused by the regeneration effect is ***,based on the model of a time-delay system,a stability criterion is developed by constructing an augmented LyapunovKrasovski functional(LKF) and using auxiliary function inequality with reciprocally convex combination ***,the validity of the method is verified through an example,and the milling stability domain lobe diagram with a parameter combination of spindle speed-cutting depth is obtained which provides operational guidelines to guarantee a stable vibration-free process.
Inconsistent feeder impedances in microgrids easily lead to uneven reactive power output of the inverter. This paper proposes an improved current-based droop control strategy for this problem. The droop coefficient is...
详细信息
Inconsistent feeder impedances in microgrids easily lead to uneven reactive power output of the inverter. This paper proposes an improved current-based droop control strategy for this problem. The droop coefficient is adjusted by the inverter capacity ratio, and the reactive current measured at the point of common coupling(PCC) is used as a reference value for improved differential control to compensate the voltage and control its reactive power output. It is found that the improved droop control has good adaptive ability and stability. The simulation results also prove the correctness and feasibility of the proposed strategy.
In order to solve the problem that ultrasonic ranging is difficult to obtain the first wave, an ultrasonic ranging system based on cross-correlation method is designed. In this system, the signal generating unit is ma...
详细信息
In order to solve the problem that ultrasonic ranging is difficult to obtain the first wave, an ultrasonic ranging system based on cross-correlation method is designed. In this system, the signal generating unit is mainly responsible for transmitting high frequency sound waves, and the data acquisition unit is mainly responsible for collecting and storing the sound waves. Then,according to the time delay between the collected acoustic signals, the time difference between the signals received by the signal acquisition unit is obtained, so as to measure the distance.
This paper presents a sampled-data frequency consensus control of microgrid with additive noise based on multiagent system. The distributed generations in microgrid are regarded as the agents and form the multiagent c...
详细信息
This paper presents a sampled-data frequency consensus control of microgrid with additive noise based on multiagent system. The distributed generations in microgrid are regarded as the agents and form the multiagent communication network. For a continuous-time multiagent system, due to the low efficiency and high cost, we apply the sampled-data approach to analyze the frequency consensus under additive noise. Here, based on primary droop control of microgrid, Alternating Direction Multiplier Method is applied to get the optimal frequency reference set points, and the frequency consensus based on leader-following multiagent system is obtained by the mean square consensus theory. The efficiency of the proposed method for control of the multiagent-based microgrid system under the additive noise is simulated in MATLAB.
This paper studies the dissipative problem of neural networks with time-varying delay and external disturbance. A suitable augmented Lyapunov-Krasovskii functional(LKF) is constructed by taking full advantage of the d...
详细信息
This paper studies the dissipative problem of neural networks with time-varying delay and external disturbance. A suitable augmented Lyapunov-Krasovskii functional(LKF) is constructed by taking full advantage of the delay information of the system and the conditions of the excitation function. Then by employing auxiliary function inequalities, the reciprocally convex combination and a vector zero-value method to deal with the derivative of the LKF, a less conservative delay-dependent dissipative criterion is obtained. Finally, a numerical example is given to show the effectiveness of this criterion.
Modeling dynamic multi-objective optimization problems (DMOPs) has been one of the most challenging tasks in the field of dynamic evolutionary optimization. Based on the analysis of the existing DMOPs, several feature...
详细信息
Modeling dynamic multi-objective optimization problems (DMOPs) has been one of the most challenging tasks in the field of dynamic evolutionary optimization. Based on the analysis of the existing DMOPs, several features widely existed in real-world applications are not taken into account: different objectives may have different function models and variables to be optimized; and the number of conflicting variables should be independent from the number of objectives; the time-linkage property is not considered. In order to overcome the above issues, a novel framework for constructing DMOPs is proposed, where all objectives can be designed independently, and the number of the conflicting variables can be tuned by users. Moreover, it is easy to add new dynamic features to this framework. Several classical dynamic multi-objective optimization algorithms are tested on four scenarios, results show that these characteristics are challenging for the existing algorithms.
In evolutionary algorithms, how to effectively select interactive solutions for generating offspring is a challenging problem. Though many operators are proposed, most of them select interactive solutions (parents) ra...
详细信息
In evolutionary algorithms, how to effectively select interactive solutions for generating offspring is a challenging problem. Though many operators are proposed, most of them select interactive solutions (parents) randomly, having no specificity for the features of landscapes in various problems. To address this issue, this paper proposes a reinforcement-learning-based evolutionary algorithm to select solutions within the approximated basin of attraction. In the algorithm, the solution space is partitioned by the k-dimensional tree, and features of subspaces are approximated with respect to two aspects: objective values and uncertainties. Accordingly, two reinforcement learning (RL) systems are constructed to determine where to search: the objective-based RL exploits basins of attraction (clustered subspaces) and the uncertainty-based RL explores subspaces that have been searched comparatively less. Experiments are conducted on widely used benchmark functions, demonstrating that the algorithm outperforms three other popular multimodal optimization algorithms.
暂无评论