Scene text recognition has inspired great interests from the computer vision community in recent years. In this paper, we propose a novel scene text recognition method using part-based tree-structured character detect...
详细信息
ISBN:
(纸本)9780769549897
Scene text recognition has inspired great interests from the computer vision community in recent years. In this paper, we propose a novel scene text recognition method using part-based tree-structured character detection. Different from conventional multi-scale sliding window character detection strategy, which does not make use of the character-specific structure information, we use part-based tree-structure to model each type of character so as to detect and recognize the characters at the same time. While for word recognition, we build a Conditional Random Field model on the potential character locations to incorporate the detection scores, spatial constraints and linguistic knowledge into one framework. The final word recognition result is obtained by minimizing the cost function defined on the random field. Experimental results on a range of challenging public datasets (ICDAR 2003, ICDAR 2011, SVT) demonstrate that the proposed method outperforms state-of-the-art methods significantly both for character detection and word recognition.
We make a thorough kinematic comparison of forward and backward swimming and maneuvering on a self-propelled robot platform that uses sub-carangifbrm swimming as the primary propulsor. An improved Central Pattern Gene...
详细信息
We make a thorough kinematic comparison of forward and backward swimming and maneuvering on a self-propelled robot platform that uses sub-carangifbrm swimming as the primary propulsor. An improved Central Pattern Generator （CPG） model allowing free adjustment of phase relationship and directional bias is employed to achieve flexible swimming and smooth transition. Considering the characteristics of forward swimming in carangiform fish and backward swimming in anguilliform fish, various backward swimming patterns for the sub-carangiform robotic fish are suitably created by reversing the direction of propagating propulsive waves. Through a combined use of the CPG control and closed-loop swimming direction control strategy, flexible and precise turning maneuvers in both forward and backward swimming are implemented and compared. By contrast with forward swimming, backward swimming requires a higher frequency or an increased lateral displacement to reach the same relative swimming speed. Noticeably, the phase difference shows a greater impact on forward swimming than on backward swimming. Our observations also indicate that the robotic fish achieves a larger turning rate in forward maneuvering than in backward maneuvering, yet these two maneuvers display comparable turning precision.
A new visual measurement method is proposed to estimate three-dimensional （3D） position of the object on the floor based on a single camera. The camera fixed on a robot is in an inclined position with respect to the...
详细信息
A new visual measurement method is proposed to estimate three-dimensional （3D） position of the object on the floor based on a single camera. The camera fixed on a robot is in an inclined position with respect to the floor. A measurement model with the camera＇s extrinsic parameters such as the height and pitch angle is described. Single image of a chessboard pattern placed on the floor is enough to calibrate the camera＇s extrinsic parameters after the camera＇s intrinsic parameters are calibrated. Then the position of object on the floor can be computed with the measurement model. Furthermore, the height of object can be calculated with the paired-points in the vertical line sharing the same position on the floor. Compared to the conventional method used to estimate the positions on the plane, this method can obtain the 3D positions. The indoor experiment testifies the accuracy and validity of the proposed method.
Pneumatic artificial muscles(PAMs)usually exhibit strong hysteresis nonlinearity and time-varying features that bring PAMs modeling and control *** characterize the hysteresis relation between PAMs’displacement and f...
详细信息
Pneumatic artificial muscles(PAMs)usually exhibit strong hysteresis nonlinearity and time-varying features that bring PAMs modeling and control *** characterize the hysteresis relation between PAMs’displacement and fluid pressure,a long short term memory(LSTM)neural network model and an adaptive Takagi-Sugeno(T-S)fuzzy model are *** show that both models perform well under the load free conditions,and the adaptive T-S Fuzzy model can furtherly adapt to the change of load with the online adaptation *** the concise expression and satisfactory performance of the adaptive T-S Fuzzy model,a model predictive controller is designed and *** show that the model predictive controller has a good performance on tracking the given references.
Fish＇s outstanding motion and coordination performance make it an excellent source of inspiration for scientists and engineers aiming to design and control next-generation autonomous underwater vehicles within the fr...
详细信息
Fish＇s outstanding motion and coordination performance make it an excellent source of inspiration for scientists and engineers aiming to design and control next-generation autonomous underwater vehicles within the framework of bionics. This paper offers a general review of the current status of bionic robotic fish, with particular emphasis on the hydrodynamic modeling and testing, kinematic modeling and control, learning and optimization, as well as motion coordination control. Among these aspects, representative studies based on ideas and concepts inspired from fish motion and coordination are discussed. At last, the major challenges and the future research directions are addressed in the context of integration of various research streams from ichthyologic, hydrodynamic, mechanical, electronic, control, and artificial intelligence. Further development of bionic robotic fish can be utilized to execute some specific missions in complex underwater environments, where operations are unsafe or impractical for divers or conventional underwater vehicles.
Hand gesture recognition has become a vital subject in the fields of human-computer interaction and rehabilitation *** paper presents a multi-modal fusion for hand gesture recognition(MFHG)model,which uses two heterog...
详细信息
Hand gesture recognition has become a vital subject in the fields of human-computer interaction and rehabilitation *** paper presents a multi-modal fusion for hand gesture recognition(MFHG)model,which uses two heterogeneous networks to extract and fuse the features of the vision-based motion signals and the surface electromyography(s EMG)signals,*** extract the features of the vision-based motion signals,a graph neural network,named the cumulation graph attention(CGAT)model,is first proposed to characterize the prior knowledge of motion coupling between finger *** CGAT model uses the cumulation mechanism to combine the early and late extracted features to improve motion-based hand gesture *** the s EMG signals,a time-frequency convolutional neural network model,named TF-CNN,is proposed to extract both the signals'time-domain and frequency-domain *** improve the performance of hand gesture recognition,the deep features from multiple modes are merged with an average layer,and then the regularization items containing center loss and the mutual information loss are employed to enhance the robustness of this multi-modal ***,a data set containing the multi-modal signals from seven subjects on different days is built to verify the performance of the multi-modal *** experimental results indicate that the MFHG can reach 99.96%and 92.46%accuracy on hand gesture recognition in the cases of within-session and cross-day,respectively.
Dear Editor,Modeling is the first and essential step for control and automation,and large models,from current Chat GPT or large language models(LLMs)to future large knowledge models of knowledge automation,would be th...
详细信息
Dear Editor,Modeling is the first and essential step for control and automation,and large models,from current Chat GPT or large language models(LLMs)to future large knowledge models of knowledge automation,would be the foundation model and infrastructure intelligence for coming intelligent industries and smart societies.
Unmanned Aerial Vehicles (UAVs) are increasingly important in dynamic environments such as logistics transportation and disaster response. However, current tasks often rely on human operators to monitor aerial videos ...
详细信息
This paper estimates an off-policy integral reinforcement learning（IRL） algorithm to obtain the optimal tracking control of unknown chaotic systems. Off-policy IRL can learn the solution of the HJB equation from the...
详细信息
This paper estimates an off-policy integral reinforcement learning（IRL） algorithm to obtain the optimal tracking control of unknown chaotic systems. Off-policy IRL can learn the solution of the HJB equation from the system data generated by an arbitrary control. Moreover, off-policy IRL can be regarded as a direct learning method, which avoids the identification of system dynamics. In this paper, the performance index function is first given based on the system tracking error and control error. For solving the Hamilton–Jacobi–Bellman（HJB） equation, an off-policy IRL algorithm is *** is proven that the iterative control makes the tracking error system asymptotically stable, and the iterative performance index function is convergent. Simulation study demonstrates the effectiveness of the developed tracking control method.
In recent years,the introduction of Siamese network has brought new vitality to the object tracking ***,high-performance Siamese trackers cannot run at a real-time speed on mobile devices due to their complex and huge...
详细信息
ISBN:
(数字)9789887581536
ISBN:
(纸本)9781665482561
In recent years,the introduction of Siamese network has brought new vitality to the object tracking ***,high-performance Siamese trackers cannot run at a real-time speed on mobile devices due to their complex and huge *** distillation is a common and effective model compression method,but it is difficult to be applied to the challenging task like object *** find out the fundamental cause is that the imbalance between the foreground and background in the object tracking task,which aggravates the problem of insufficient feature extraction ability of small ***,we propose the attention mask distillation(AMD) to help the student tracker focus on the foreground area faster and more *** attention mask can be easily obtained from the feature maps and brings fine-granularity to the traditional binary *** experimental results on OTB100 and VOT2018 show that our method enables the student tracker perform as well as the teacher *** the same time,it's able to run on the CPU at a hyper-real-time of 66 fps and achieves nearly 9 times model compression *** low computational and storage costs make it possible to deploy high-performance trackers on resource-constrained platforms.
暂无评论