Gait recognition is an active research area that uses a walking theme to identify the subject *** Gait Recognition(HGR)is performed without any cooperation from the ***,in practice,it remains a challenging task under ...
详细信息
Gait recognition is an active research area that uses a walking theme to identify the subject *** Gait Recognition(HGR)is performed without any cooperation from the ***,in practice,it remains a challenging task under diverse walking sequences due to the covariant factors such as normal walking and walking with wearing a ***,over the years,have worked on successfully identifying subjects using different techniques,but there is still room for improvement in accuracy due to these covariant *** paper proposes an automated model-free framework for human gait recognition in this *** are a few critical steps in the proposed ***,optical flow-based motion region esti-mation and dynamic coordinates-based cropping are *** second step involves training a fine-tuned pre-trained MobileNetV2 model on both original and optical flow cropped frames;the training has been conducted using static *** third step proposed a fusion technique known as normal distribution serially *** the fourth step,a better optimization algorithm is applied to select the best features,which are then classified using a Bi-Layered neural *** publicly available datasets,CASIA A,CASIA B,and CASIA C,were used in the experimental process and obtained average accuracies of 99.6%,91.6%,and 95.02%,*** proposed framework has achieved improved accuracy compared to the other methods.
作者：
Xu, ZeliangKim, Dong InWoo, Simon S.
Department of Computer Science and Engineering Suwon16419 Korea Republic of
Department of Electrical and Computer Engineering Suwon16419 Korea Republic of
This paper proposes a novel cloud-edge collaborative distributed diffusion model for AI-generated content (AIGC) such as image generation, which integrates adaptive clustering techniques with dynamic step-size optimiz...
详细信息
The exchange of knowledge is widely recognized as a crucial aspect of effective knowledge management. When it comes to sharing knowledge within Prison settings, things get complicated due to various challenges such as...
详细信息
With flexibility in maneuverability and remarkable adaptability, airborne bistatic radar system can obtain excellent detection performance for high-speed target by employing coherent integration. However, range migrat...
详细信息
With flexibility in maneuverability and remarkable adaptability, airborne bistatic radar system can obtain excellent detection performance for high-speed target by employing coherent integration. However, range migration (RM) and Doppler frequency migration (DFM) could become serious issues due to the relative motion characteristics of airborne platforms and high-speed target. Meanwhile, various unpredictable factors such as atmospheric turbulence and mechanical issues, etc., resulting in additional motion errors, would have further negative impacts on motion state and flight trajectory of airborne platforms. This phenomenon would serious consequence on coherent integration and target detection. Thus, we make contributions to tackle these limitations and enhance coherent integration and detection performance. First, we establish signal model with high-speed target in three-dimensional (3-D) space for airborne bistatic radar system, along with motion error model which simultaneously includes translational error and rotational error. Next, we articulate range history's mathematical expression and further derive echo signal model. We then propose an improved generalized Radon Fourier transform (IGRFT) method. More specifically, the purpose of IGRFT is achieving joint search for the parameters of the target motion and the parameters of motion error, to ensure high precision parameter estimation and high gain integration. However, the computational complexity surges due to the increasing of search dimensionality. To devise computationally feasible methods for practical applications, we split the high-dimensional maximization process into two disjoint problems by sequentially searching motion parameters and then motion error parameters, and this method is named GRT (generalized Radon transform)-IGRFT. Numerical simulations show that the proposed algorithms can correctly estimate parameters and achieve signal integration and target detection. Finally, we present performanc
The increasing dependence on smartphones with advanced sensors has highlighted the imperative of precise transportation mode classification, pivotal for domains like health monitoring and urban planning. This research...
详细信息
The increasing dependence on smartphones with advanced sensors has highlighted the imperative of precise transportation mode classification, pivotal for domains like health monitoring and urban planning. This research is motivated by the pressing demand to enhance transportation mode classification, leveraging the potential of smartphone sensors, notably the accelerometer, magnetometer, and gyroscope. In response to this challenge, we present a novel automated classification model rooted in deep reinforcement learning. Our model stands out for its innovative approach of harnessing enhanced features through artificial neural networks (ANNs) and visualizing the classification task as a structured series of decision-making events. Our model adopts an improved differential evolution (DE) algorithm for initializing weights, coupled with a specialized agent-environment relationship. Every correct classification earns the agent a reward, with additional emphasis on the accurate categorization of less frequent modes through a distinct reward strategy. The Upper Confidence Bound (UCB) technique is used for action selection, promoting deep-seated knowledge, and minimizing reliance on chance. A notable innovation in our work is the introduction of a cluster-centric mutation operation within the DE algorithm. This operation strategically identifies optimal clusters in the current DE population and forges potential solutions using a pioneering update mechanism. When assessed on the extensive HTC dataset, which includes 8311 hours of data gathered from 224 participants over two years. Noteworthy results spotlight an accuracy of 0.88±0.03 and an F-measure of 0.87±0.02, underscoring the efficacy of our approach for large-scale transportation mode classification tasks. This work introduces an innovative strategy in the realm of transportation mode classification, emphasizing both precision and reliability, addressing the pressing need for enhanced classification mechanisms in an eve
The evolution of wireless networks necessitates so-phisticated optimization strategies to address the challenges posed by heterogeneous traffic arising from various domains. Digital Twin (DT) concept has emerged as an...
详细信息
The time-sensitive Internet of Things (IoT) applications within 5G and edge computing environments presents unique challenges in network resource management. Current systems struggle with efficiently managing the high...
详细信息
Recent advances in data-driven imitation learning and offline reinforcement learning have highlighted the use of expert data for skill acquisition and the development of hierarchical policies based on these skills. Ho...
Human activity recognition (HAR) plays a crucial role in assisting the elderly and individuals with vascular dementia by providing support and monitoring for their daily activities. This paper presents a deep learning...
详细信息
Autonomous Vehicle System (AVS) is rapidly advancing and is expected to completely transform the transportation industry, bringing about a new era of mobility. As digital data proliferation strains network resources, ...
详细信息
暂无评论