Search

In intelligent traffic monitoring systems, the significant distance between cameras and their non-overlapping fields of view leads to several issues. These include incomplete tracking results from individual cameras, difficulty in matching targets across multiple cameras, and the complexity of inferring the global trajectory of a target. In response to the challenges above, a deep learning-based vehicle tracking algorithm called FairMOT-MCVT is proposed. This algorithm con-siders the vehicles' characteristics as rigid targets from a roadside perspective. Firstly, a Block-Efficient module is designed to enhance the network's ability to capture and characterize image features across different layers by integrating a multi-branch structure and depth-separable convolutions. Secondly, the Multi-scale Dilated Attention (MSDA) module is introduced to improve the feature extraction capability and computational efficiency by combining multi-scale feature fusion and attention mechanisms. Finally, a joint loss function is crafted to better distinguish between vehicles with similar appearances by combining the trajectory smoothing loss and velocity consistency loss, thereby considering both position and velocity continuity during the optimization process. The proposed method was evaluated on the public UA-DETRAC dataset, which comprises 1210 video sequences and over 140,000 frames captured under various weather and lighting conditions. The experimental results demonstrate that the FairMOT-MCVT algorithm significantly enhances multi-target tracking accuracy (MOTA) to 79.0, IDF1 to 84.5, and FPS to 29.03, surpassing the performance of previous algorithms. Additionally, this algorithm expands the detection range and reduces the deployment cost of roadside equipment, effectively meeting the practical application requirements. [ABSTRACT FROM AUTHOR]

Changes in land use patterns, types, and intensities significantly impact ecosystem services. This study follows the time series logic from history to the expected future to investigate the spatial and temporal characteristics of land use changes in Ezhou and their potential impacts on the ecosystem services value (ESV). The results show that the Markov–FLUS model has strong applicability in predicting the spatial pattern of land use, with a Kappa coefficient of 0.9433 and a FoM value of 0.1080. Between 2000 and 2020, construction land expanded continuously, while water area remained relatively stable, and other land types experienced varying degrees of contraction. Notably, the area of construction land expanded significantly compared to 2000, and it expanded by 70.99% in 2020. Moreover, the watershed area expanded by 9.30% from 2000 to 2010, but there was very little change in the following 10 years. Under the three scenarios, significant differences in land use changes were observed in Ezhou City, driven by human activities, particularly the strong expansion of construction land. In the inertial development scenario, construction land expanded to 313.39 km2 by 2030, representing a 38.30% increase from 2020. Conversely, under the farmland protection scenario, construction land increased to 237.66 km2, a 4.89% rise from 2020. However, in the ecological priority development scenario, the construction land area expanded to 253.59 km2, a 10.13% increase from 2020. Compared to 2020, the ESV losses in the inertia development and farmland protection scenarios were USD 4497.71 and USD 1072.23, respectively, by 2030. Conversely, the ESV under the ecological protection scenario increased by USD 2749.09, emphasizing the importance of prioritizing ecological protection in Ezhou City's development. This study may provide new clues for the formulation of regional strategies for sustainable land use and ecosystem restoration. [ABSTRACT FROM AUTHOR]

Developing complex hyperspectral image (HSI) sensors that capture high-resolution spatial information and voluminous (hundreds) spectral bands of the earth's surface has made HSI pixel-wise classification a reality. The 3D-CNN has become the preferred HSI pixel-wise classification approach because of its ability to extract discriminative spectral and spatial information while maintaining data integrity. However, HSI datasets are characterized by high nonlinearity, voluminous spectral features, and limited training sample data. Therefore, developing deep HSI classification methods that purely utilize 3D-CNNs in their network structure often results in computationally expensive models prone to overfitting when the model depth increases. In this regard, this paper proposes an integrated deep multi-scale 3D/2D convolutional network block (MiCB) for simultaneous low-level spectral and high-level spatial feature extraction, which can optimally train on limited sample data. The strength of the proposed MiCB model solely lies in the innovative arrangement of convolution layers, giving the network the ability (i) to simultaneously convolve the low-level spectral with high-level spatial features; (ii) to use multiscale kernels to extract abundant contextual information; (iii) to apply residual connections to solve the degradation problem when the model depth increases beyond the threshold; and (iv) to utilize depthwise separable convolutions in its network structure to address the computational cost of the proposed MiCB model. We evaluate the efficacy of our proposed MiCB model using three publicly accessible HSI benchmarking datasets: Salinas Scene (SA), Indian Pines (IP), and the University of Pavia (UP). When trained on small amounts of training sample data, MiCB is better at classifying than the state-of-the-art methods used for comparison. For instance, the MiCB achieves a high overall classification accuracy of 97.35%, 98.29%, and 99.20% when trained on 5% IP, 1% UP, and 1% SA data, respectively. [ABSTRACT FROM AUTHOR]

In this paper, we propose a robust and effective framework to largely improve the performance of human action recognition using depth maps. The key contribution is the proposition of the Sub-action Motion History Image (SMHI) and Static History Image (SHI) in a depth sequence. We evenly subdivide the normalized motion energy into a set of segments which corresponding frame indices are used to partition a video into different sub-actions segments. The Local Binary Patterns (LBP) descriptor is then computed from the SMHI and SHI for the representation of an action. We evaluate the proposed framework on MSR Action3D dataset. Experimental results indicate that the proposed approach outperforms the most of the art methods and demonstrate the effectiveness of the proposed approaches.