Feature representations play a crucial role in modern face recognition systems. Most hand-crafted image descriptors usually provide low-level information. In this paper, we propose a novel feature learning method base...
详细信息
ISBN:
(纸本)9781479923427
Feature representations play a crucial role in modern face recognition systems. Most hand-crafted image descriptors usually provide low-level information. In this paper, we propose a novel feature learning method based on deep neural networks to obtain high-level, hierarchical representations for face verification. Learning proceeds in two phases. In the pre-training phase, we train Restricted Boltzmann Ma-chine(RBM) networks for each modular region in the image separately. In the fine-tuning phase, in order to develop good discriminative ability, we stack the RBM networks of each region in deep architecture and combine deep learning with side information constraints in the whole image scale. Finally, we formulate the proposed method as an appropriate optimization problem and adopt gradient descent algorithm to get the optimal solution. We evaluate our method on the LFW dataset. Representations learned from the networks achieve comparable performance (93.11%) to the state-of-art method.
This article describes the control algorithm and software implementation of astronomical tracking based on the mechanical analysis for Five-hundred-meter Aperture Spherical Telescope(FAST) that is being built in Guizh...
详细信息
This article describes the control algorithm and software implementation of astronomical tracking based on the mechanical analysis for Five-hundred-meter Aperture Spherical Telescope(FAST) that is being built in Guizhou province of China, which aims to help astronomers resolve questions in cosmology. On the basis of position and attitude feedback algorithm of cable driven parallel robot, the control of X-Y positioner with feed-forward was introduced and the astronomical tracking experiments were conducted on scale model to verify the validity of control strategy. The future works for focus cabin suspension of FAST are also pointed out.
Chemical emergency response process is a dynamic complex process. To improve the effectiveness of emergency management, Petri nets are applied to model dynamic process of chemical emergency response process in this pa...
详细信息
ISBN:
(纸本)9781479905300;9781479905294
Chemical emergency response process is a dynamic complex process. To improve the effectiveness of emergency management, Petri nets are applied to model dynamic process of chemical emergency response process in this paper. A dynamic process model of emergency management is built based on hybrid Petri nets and hierarchical object-oriented Petri nets. The result reveals that the model can indicated the properties of Continuity, timeliness, interactive and multiple possibilities of emergency response process.
Facility layout problem (FLP) which deals with the placement of facilities in the plant area is one of the most critical problems in manufacturing systems design. This paper presents a novel facility layout approach b...
详细信息
ISBN:
(纸本)9781479927456
Facility layout problem (FLP) which deals with the placement of facilities in the plant area is one of the most critical problems in manufacturing systems design. This paper presents a novel facility layout approach based on differential evolution (DE) algorithm. Firstly, mathematical model with constraints for the FLP is established. Then, a new facility layout algorithm based on DE is proposed, which can improve the convergence rate by using the interference checking method. Finally, a simulation platform with three-dimensional visualization is developed. The comparison experiments of the proposed method and genetic algorithm (GA) based on robot work cell layout problem show that the proposed method converges faster and can obtain better optimal layout.
Due to illumination variation, view changes, and complex road conditions, automatic lane detection is crucial for path finding and planning for autonomous car-like robots. In this paper, a robust lane detection method...
详细信息
ISBN:
(纸本)9781467362498;9781467362481
Due to illumination variation, view changes, and complex road conditions, automatic lane detection is crucial for path finding and planning for autonomous car-like robots. In this paper, a robust lane detection method is proposed. Firstly, in order to extract the edges of lanes in traffic scenarios, we adopt an adaptive thresholding strategy to binarize a gradient image and trace edges by using their local gradient information. Moreover, by integrating gradient constraints and introducing a reverse voting strategy to the standard Hough transform, we greatly improved speed and stability of line extraction. Further, through inverse perspective mapping the endpoints of extracted lines to the world coordination, we can combine the extracted lines from different cameras. Finally, the lane could be detected by matching two points instead of two parallel lines in parameter space. Extensive experiments and comparisons show the efficiency of the proposed method.
Voting strategy is very useful in pattern recognition. Many methods, like Boosting and Bagging, are proposed and are successfully used in some applications using this strategy. However, these methods are infeasible or...
详细信息
Voting strategy is very useful in pattern recognition. Many methods, like Boosting and Bagging, are proposed and are successfully used in some applications using this strategy. However, these methods are infeasible or unsuitable for handwritten Chinese character recognition because of the problem's characteristics. In this paper, a self-generation voting method is proposed for further improving the recognition rate in handwritten Chinese character recognition. This method learns a set of parameters first for generating a set of samples from the test sample, and then classify these generated samples using a base-line classifier. At last, it gives the final recognition result by voting. Experimental results on two databases show that the proposed method is effective and useful in handwritten Chinese character recognition systems.
A streamlined parallel traffic management system (PtMS) is outlined that works alongside a redesigned intelligent transportation system in Qingdao, China. The PtMS"s structure provides enhanced control and manage...
详细信息
A streamlined parallel traffic management system (PtMS) is outlined that works alongside a redesigned intelligent transportation system in Qingdao, China. The PtMS"s structure provides enhanced control and management support, with increased versatility for use in real-world scenarios.
A biomimetic underwater vehicle propelled by two undulating long-fins is introduced in this paper. The concerned vehicle is propelled by two symmetrical undulating long-fins installed on both sides. Ten servo motors a...
详细信息
A biomimetic underwater vehicle propelled by two undulating long-fins is introduced in this paper. The concerned vehicle is propelled by two symmetrical undulating long-fins installed on both sides. Ten servo motors are employed to drive the long-fins and cosine wave function is employed for motor control. A real-time control system is designed for controlling the long-fins by adjusting its oscillating frequency and oscillating amplitude. An inertial measurement unit is installed to collect the accelerations and angular velocity. To obtain the relationship between oscillating frequency/amplitude and swimming performance, kinematic analysis and hydromechanic analysis are given. By dividing the long-fin into many small elements and computing the hydrodynamic force acting on each element, the instantaneous thrust generated by the long-fin is obtained. Then the average thrust of the long-fin is obtained by summing up the forces acted on the elements in one undulating period. Then swimming experiments are carried out to validate the vehicle design and kinematic analysis and hydromechanic analysis. And two swimming motion modes including marching and rotating locomotion are chosen. Finally, discussions between the swimming performance and the oscillating parameters are given.
Despite the growing number of videos featuring electronic cigarettes, there has been no investigation of the portrayal of these videos. This paper presents the first surveillance data of electronic cigarette videos on...
详细信息
In traffic video surveillance systems, vehicles with various distances from the camera have different sizes, resolutions, and angles in traffic images. The common multi-scale method, which scales one vehicle template ...
详细信息
In traffic video surveillance systems, vehicles with various distances from the camera have different sizes, resolutions, and angles in traffic images. The common multi-scale method, which scales one vehicle template or the input image for detecting vehicles with different sizes, may fail to detect vehicles with various distances from the camera due to the change of the resolution and angle. To deal with this problem, we have proposed a multi-scale model including multiple templates with different scales and features. Our method includes two steps: constructing the multi-scale model and its probability model, and detecting vehicles from traffic images. In the first step, the multi-scale model is constructed by using three templates T 1 , T 2 , T 3 which represent vehicles with the short, medium, and long distance from the camera respectively. Each template contains one or some combination of sketch, texture, flatness, and color. In the second step, the three templates are applied for vehicle detection by using the template matching with local maximization operations. The main innovation of this paper is that the combination of multi-template and multi-scale method is applied to detect vehicles with various distances from the camera. To test our method, we have done several experiments on various traffic conditions. The experimental results show that our method effectively copes with vehicles with various distances from the camera and provides the detailed vehicle information after vehicle detection. Moreover, our method adapts to various weather conditions, slight pose variance, and slight occlusion.
暂无评论