In recent years,Siamese-based trackers have achieved excellent performance,most of them usually calculate the similarity of each position on the search region to the object through the cross-correlation layer for trac...
详细信息
ISBN:
(数字)9789887581536
ISBN:
(纸本)9781665482561
In recent years,Siamese-based trackers have achieved excellent performance,most of them usually calculate the similarity of each position on the search region to the object through the cross-correlation layer for tracking obj *** solve the problem that the above method neglects the correspondence of the local information between the object and the search region and cannot adapt to the object deformation well,we propose a Siamese network-based tracker with position attention network(SiamPA).First,we use Siamese backbone network to extract template and search region ***,we adopt the boxguided object feature selection strategy to avoid similarity calculations for background *** addition,we introduce the position attention network instead of the cross-correlation layer to learn the part-level relationship between the object and the search region ***,the classification-regression sub-network is used to decode the similarity respond map obtained by the position attention network and predict the position of the *** contribution,one is to propose a box-guided method for refining object features,and the other is to introduce a position attention network for information *** on three challenging benchmarks including GOT-10 k,UAV123 and OTB-100 demonstrate that our SiamPA achieves excellent tracking performance with a real-time speed.
Due to the low efficiency and high cost of manual tea picking,as well as the development and application of machine vision and image recognition technology,the mechanization and intelligence of tea picking will become...
详细信息
ISBN:
(数字)9789887581536
ISBN:
(纸本)9781665482561
Due to the low efficiency and high cost of manual tea picking,as well as the development and application of machine vision and image recognition technology,the mechanization and intelligence of tea picking will become a *** of all,this paper analyzed the advantages and disadvantages of several common target detection methods by using Matlab ***,considering cost,accuracy and real-time performance,the method of combining K-means clustering and image morphology processing is finally selected to extract tea ***,the method is reproduced on STM32 single chip ***,the effect was verified in the actual tea garden,which laid a foundation for the subsequent intelligent picking.
Dynamics parameter identification is a key factor and a difficult area in the development of robot motion control technology. In order to obtain accurate dynamics parameters, an integral identification method combined...
详细信息
Although deep learning methods have been widely applied in slam visual odometry over the past decade with impressive improvements， the accuracy remains limited in complex dynamic environments. In this paper， a compo...
Although deep learning methods have been widely applied in slam visual odometry over the past decade with impressive improvements， the accuracy remains limited in complex dynamic environments. In this paper， a composite mask-based generative adversarial network is introduced to predict camera motion and binocular depth maps. Specifically， a perceptual generator is constructed to obtain the corresponding parallax map and optical flow from between two neighboring frames. Then， an iterative pose improvement strategy is proposed to improve the accuracy of pose estimation. Finally， a composite mask is embedded in the discriminator to sense structural deformation in the synthesized virtual image， thereby increasing the overall structural constraints of the network model， improving the accuracy of camera pose estimation， and reducing drift issues in the Visual Odometer. Detailed quantitative and qualitative evaluations on the KITTI dataset show that the proposed framework outperforms existing conventional， supervised learning and unsupervised depth VO methods， providing better results in both pose estimation and depth estimation.
BIG models or foundation models are rapidly emerging as a key force in advancing intelligent societies[1]–[3]Their significance stems not only from their exceptional ability to process complex data and simulate advan...
详细信息
BIG models or foundation models are rapidly emerging as a key force in advancing intelligent societies[1]–[3]Their significance stems not only from their exceptional ability to process complex data and simulate advanced cognitive functions,but also from their potential to drive innovation across various industries.
作者：
Jiao, RanranLi, BoChen, XiangyongJiang, XiaoweiSchool of Automation
China University of Geosciences Hubei Key Laboratory of Advanced Control Intelligent Automation of Complex Systems Engineering Research Center of Intelligent Technology for Geo-Exploration Ministry of Education Wuhan China School of Finance
Anhui University of Finance and Economics Bengbu China Key Laboratory of Complex systems and Intelligent Computing in Universities of Shandong Linyi University Linyi China
This paper addresses the problem of leader-following consensus in heterogeneous multi-agent systems (HMAS) with input saturation and communication delay. Given the practical differences between leader and follower dim...
详细信息
This paper studies a classical single pursuer and single evader pursuit-evasion *** pursuer attempts to capture the slower evader who aims to extend its lifetime during the *** simplify this question,requiring the eva...
详细信息
ISBN:
(数字)9789887581536
ISBN:
(纸本)9781665482561
This paper studies a classical single pursuer and single evader pursuit-evasion *** pursuer attempts to capture the slower evader who aims to extend its lifetime during the *** simplify this question,requiring the evader to take fixation strategy which is choosing the farthest point in its current dominant region as aimpoint and moving at a constant *** the pursuer is faster than the ***,the speed ratio is a *** instaneous state space will be partitioned into pursuer's dominant zone and evader's dominant zone by the generalized Apollonius *** pursuit strategy is based on minimizing the area of the evader's dominant ***,we propose a supermodular game for this ***,the existence of the Nash equilibrium is *** results based on Q-learning are presented to solve the problem,which shows the effectiveness of this method.
Recently,the formation control of multiple autonomous mobile robots(AMRs) have gained significant attention,and autonomous mobile robots(AMRs) have applied to all aspects of our ***-agent reinforcement learning is use...
详细信息
ISBN:
(数字)9789887581536
ISBN:
(纸本)9781665482561
Recently,the formation control of multiple autonomous mobile robots(AMRs) have gained significant attention,and autonomous mobile robots(AMRs) have applied to all aspects of our ***-agent reinforcement learning is used to solve the autonomously sequential decision-making problem of agents in a common environment with competition or ***,we present a utility function and a reward function to achieve formation control with collision avoidance for a rigid AMRs system,and build a simulation environment to meet environmental requirements based on MPE.
Rail surface defect inspection is an essential task for the railway system. However, due to the similarity of the background and defect foreground pixels, uneven textures, irregular shapes and multiple scales of the e...
详细信息
ISBN:
(数字)9798331510138
ISBN:
(纸本)9798331510145
Rail surface defect inspection is an essential task for the railway system. However, due to the similarity of the background and defect foreground pixels, uneven textures, irregular shapes and multiple scales of the existing defects, the inspection accuracy is still required to be further improved. In this paper, we first expand our dataset by GAN-based method for the scarcity of the defect samples. Then, we propose a diffusion-based model to perform the classification and segmentation task. The diffusion model predicts the mask from the Gaussian noise broken ground truth by successive designed transformer decoder. Meanwhile, to better distinguish the pixels in the boundary or other illegible regions, a boundary-frequency feature enhancement module (BFM) is proposed to make the model pay more attention to the defect-related features. Experiments on the rail surface defect dataset revealed that the proposed diffusion-based transformer decoder and BFM can work well together and efficiently improve the rail surface defect segmentation accuracy.
In renewable energy grids, solar irradiance prediction is significant for electric operation planning. The emphasis of our work is forecasting irradiance changes over multiple days ahead using hourly irradiance data i...
详细信息
ISBN:
(数字)9798350387780
ISBN:
(纸本)9798350387797
In renewable energy grids, solar irradiance prediction is significant for electric operation planning. The emphasis of our work is forecasting irradiance changes over multiple days ahead using hourly irradiance data in association with weather parameters such as temperature, humidity, barometric pressure, wind speed, and cloud cover. Specifically, a framework based on TimesNet was proposed for 4-day global horizontal irradiance (GHI) forecasting and a comparative experiment was conduct with three well-performing models in long-term forecasting tasks. The experimental results indicate that the suggested method outperformed the accuracy when compared to other models, with MSE 0.198, MAE 0.259, RMSE 0.445 and R
2
0.787.
暂无评论