As a unique property of the object surface, the spectral reflectance plays an important role in computer vision applications and in realistic material reproduction. To determine the influence of the light source on th...
详细信息
As a unique property of the object surface, the spectral reflectance plays an important role in computer vision applications and in realistic material reproduction. To determine the influence of the light source on the spectral reflectance reconstruction accuracy, the pseudo-inverse method is adopted to reconstruct the spectral reflectance. Further, the genetic algorithm is used to optimize the light source for improving the reconstruction accuracy. The experiment results show that the spectral reflectance reconstruction accuracy is highly affected by the light source spectra, and the light source spectra optimized by the proposed strategy significantly outperforms several commonly used illumination source.
We propose a position and posture measurement method based on active binocular vision to improve localization accuracy and stability of bolts in substation fittings. Firstly, we obtain the color image and the depth im...
详细信息
We propose a position and posture measurement method based on active binocular vision to improve localization accuracy and stability of bolts in substation fittings. Firstly, we obtain the color image and the depth image of bolts and use shape-based template matching method to obtain the image coordinates of the bolts. Secondly, we obtain the 3 D coordinates of the bolts based on the binocular camera stereo vision model. Finally, based on the point-cloud information of bolts plane, we obtain the posture information of the bolts by using the least square method. The experimental results show that the binocular system designed in this paper can accurately identify the bolts and has good stability. Within 500 mm measurement range, the position measurement error rate of bolts is less than 0.8 %, and the angle measurement error of the fitting plane is less than 1°.
Calligraphy is an important humanistic symbol of Chinese civilization. However, most of the calligraphy is incomplete, which has only a small number of Chinese characters circulating in the world. How to use these sam...
详细信息
Calligraphy is an important humanistic symbol of Chinese civilization. However, most of the calligraphy is incomplete, which has only a small number of Chinese characters circulating in the world. How to use these samples to efficiently restore the remaining calligraphy characters has always been considered as a difficult task. In this work, we propose Densenetpix2 pix model based style transfer method to solve this problem. By training some samples to learn the rules of transferring the printed font images to the calligraphy characters images, Densenet-pix2 pix can predict the remaining calligraphy characters. Our method modify the generation network and optimization strategies of style transfer, which improves the generation quality of the calligraphy characters and the stability of the model. In addition, we use pre-trained feature extraction models to extract content information and style information, and scientifically evaluate the quality of our generated calligraphy characters from these two aspects. We compared our method with several other baseline methods. The experimental results show that our method can effectively restore the remaining calligraphy characters, and the generated Chinese characters are more delicate.
Protein complexes are key molecular entities that play an integral role in human life activities. Systematic identification of protein complexes is an important application of data mining in the biological sciences. E...
详细信息
Protein complexes are key molecular entities that play an integral role in human life activities. Systematic identification of protein complexes is an important application of data mining in the biological sciences. Existing multi-label learning algorithms can effectively label nodes belonging to different complexes in protein-protein interaction network to identify overlapping complexes. However, the protein complexes formed by the stochastic strategy may have unstable results and insufficient community quality. To solve these problems, this paper proposes a novel protein complex identification method based on multisource fused data and the multi-label learning algorithm. The descending order of the potential influence of the nodes is used as the node selection order to solve the problem of unstable partitioning of the composite results. The comprehensive similarity obtained by the link correlation and the similarity of the gene annotations is used as the node label update strategy to improve the quality of the composite. The experimental results show that the new proposed method is much more effective and feasible,and has higher precision and biological significance.
In this paper, the problem of finite-time fault detection(FD) is investigated for networked discrete-time singularly perturbed systems under the stochastic communication protocol(SCP). To alleviate the undesired data ...
详细信息
In this paper, the problem of finite-time fault detection(FD) is investigated for networked discrete-time singularly perturbed systems under the stochastic communication protocol(SCP). To alleviate the undesired data collisions caused by largescale data transmissions via the bandwidth-limited communication network, the SCP modeled by a Markov chain is *** to practical factors such as observation error, the Markov modes of the SCP are assumed to be available to the fault detection filter(FDF) according to a hidden Markov process. Our aim is to design a hidden-Markov-model-based FDF as the residual generator such that the resulting augmented system is stochastically H∞ finite-time bounded. A sufficient condition in terms of linear matrix inequalities(LMIs) is derived which ensures the existence of such an FDF. The parameters of the desired FDF are explicitly given when there exist feasible solutions to these LMIs. Finally, the effectiveness of the FD method is demonstrated by a numerical example.
Recently, compressed sensing(CS) has been widely used in ground penetrating radar(GPR) imaging due to its low sampling time, high quality performance and high quality imaging results in underground imaging. However, f...
详细信息
Recently, compressed sensing(CS) has been widely used in ground penetrating radar(GPR) imaging due to its low sampling time, high quality performance and high quality imaging results in underground imaging. However, for densely distributed underground multi-target, the error of GPR imaging based on CS is large. So, this paper proposes a filter matrix,which can effectively reduce the energy of B-scan hyperbolic arms, increase the hyperbolic apex energy, and reduce the error of GPR imaging based on CS in densely distributed multi-target environment. The algorithm also takes into account the distance between the antenna and the ground. At first, a method of the normalized is used to calculate the echo data. Then the filter matrix designed in this paper is used. Experimental results present that the filter matrix can effectively reduce the imaging error in multi-target environment.
Aiming at the phenomenon that the target particle tracking algorithm is affected by illumination changes and occlusion in the target tracking process, the goal is lost. A target tracking algorithm based on convolution...
详细信息
Aiming at the phenomenon that the target particle tracking algorithm is affected by illumination changes and occlusion in the target tracking process, the goal is lost. A target tracking algorithm based on convolutional neural network and particle filtering is proposed. The algorithm uses the convolutional neural network to automatically learn the depth features of the target,and extracts the more abstract semantic information of the target. The semantic information makes the algorithm robust to the apparent changes of the target, which can alleviate the drift problem to some extent. The algorithm can effectively combine the target apparent model based on convolutional neural network with the particle filter framework. The experimental results show that compared with the other five algorithms in the particle filter framework, the algorithm can track the moving targets with partial occlusion and morphological changes more robustly.
In the process of urban sewage treating, reducing the energy consumption and improving the quality of the effluent are significantly meaningful. According to the activated sludge method, the key factors affecting the ...
详细信息
In the process of urban sewage treating, reducing the energy consumption and improving the quality of the effluent are significantly meaningful. According to the activated sludge method, the key factors affecting the energy consumption and water quality of wastewater treatment are determined. In order to minimize the energy consumption of the activated sludge process and maximize the quality of the effluent, four different objective functions are modeled [i.e., the airflow rate, the carbonaceous biochemical oxygen demand(CBOD) of the effluent, the total phosphorus(TP) of the effluent, and the ammonia nitrogen of the effluent(NH4-N)]. These models are developed using a back propagation(BP) neural network based on industrial data, and dissolved oxygen(DO) is the controlled variable. A multi-objective model was evaluated by six evaluation indicators. Based on the analysis of the model and the mechanism of activated sludge process, the multi-objective particle swarm optimization(MOPSO)algorithm was used to optimize the energy consumption and water quality of the activated sludge process. The experimental results show that eventually reduce aeration energy consumption by 17%.
This paper is concerned with the optimal trajectory planning for a 6 degree-of-freedom(DOF) manipulator, where the energy minimization is considered under the constraint of power. Firstly, based on the establishment o...
详细信息
This paper is concerned with the optimal trajectory planning for a 6 degree-of-freedom(DOF) manipulator, where the energy minimization is considered under the constraint of power. Firstly, based on the establishment of kinematics model, the influence of main factors on power is analyzed by using Jacobian matrix and Lagrange equation. Then the energy consumption is calculated as the objective function by the power consumption of each joint, and in order to obtain the desired trajectory for the arm, each joint locus is approximated by a 3-5-3 spline interpolation curve. To this end, in order to obtain the most reliable solutions for trajectory planning, the improved adaptive genetic algorithm(IAGA) is used to optimize the energy under the constraint of power, speed and acceleration, and the crossover rate and mutation rate are effectively adjusted with ***, we conduct the simulation by using the dynamics analysis software-ADAMS. The results demonstrate the effectiveness of minimizing the optimized energy in the case of constrained power mainly.
This paper investigates the problem of fault detection(FD) for networked singularly perturbed systems under the dynamic event-triggered communication. To save the network resources, a dynamic event-triggered communica...
详细信息
This paper investigates the problem of fault detection(FD) for networked singularly perturbed systems under the dynamic event-triggered communication. To save the network resources, a dynamic event-triggered communication mechanism(DETCM) is proposed, which contains some existing triggering mechanisms as special cases. Our aim is to design a fault detection filter(FDF) under such a DETCM, which ensures that the resulting filtering error dynamics of FD is asymptotically stable and satisfies an H∞ performance requirement. By constructing a Lyapunov function dependent on both the singular perturbation parameter(SPP) and flexible variable of the DETCM, a sufficient condition is obtained in terms of linear matrix inequalities(LMIs), which ensures the existence of the desired FDF. When these LMIs have feasible solutions, the parameters of the FDF are explicitly given and the admissible upper and lower bounds of the SPP are evaluated. A numerical example is provided to demonstrate the effectiveness of the design method of the FDF.
暂无评论