Sijia Zhou
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Sijia Zhou has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Chein-I Chang
Center for Hyperspectral Imaging in Remote Sensing (CHIRS), Information and Technology College, Dalian Maritime University, Dalian, China
2 shared publications - Chunyan Yu
Information and Technology College, Dalian Maritime University, Dalian, China
2 shared publications Meiping Song
Center of Hyperspectral Imaging in Remote Sensing (CHIRS), Information and Technology College, Dalian Maritime University, Dalian, China
2 shared publicationsBaoyu Gong
Center of Hyperspectral Imaging in Remote Sensing (CHIRS), Information and Technology College, Dalian Maritime University, Dalian, China
1 shared publicationDezhou Zhang
Wenzhou University
1 shared publication- Enyu Zhao
Information Science and Technology College, Dalian Maritime University, Dalian 116026, China
1 shared publication
Latest Publications
- Open Access
- 22 May 2023
- Published by Springer Nature in Research Square
A lightweight high-resolution Human pose estimation method with an integrated attention mechanism
Deep learning has significant advantages in the field of human posture estimation. However, traditional methods often increase the parameter size and complexity of posture estimation networks by widening and deepening them to improve performance. To address this issue, this paper proposes a lightweight human posture estimation network, ADNet, based on the high-resolution network(HRNet). ADNet replaces Basic Blocks and downsampling in the high-resolution network with GABlocks and GDBlocks that incorporate lightweight convolution and attention mechanisms to reduce the number of parameters and computational complexity in the network. Additionally, ADNet solves the problem of feature redundancy by fusing relationships between different channels to make effective feature channels fully utilized. Experimental results show that ADNet achieves high-precision detection of human keypoint information even with fewer parameters and less computational complexity than commonly used human posture estimation networks, such as Cascaded Pyramid Network (CPN), Stacked Hourglass Network(Hourglass), and HRNet. Results also indicate that, compared to HRNet, ADNet reduces the model parameter size by 69.1% and computational complexity by 58.9% under the same image resolution and environment configuration. Furthermore, in the visualization of feature maps, the network using ADNet exhibits stronger feature extraction performance and less feature redundancy.
- 22 May 2023
- Published by Springer Nature in Research Square
Open AccessA lightweight high-resolution Human pose estimation method with an integrated attention mechanism
Deep learning has significant advantages in the field of human posture estimation. However, traditional methods often increase the parameter size and complexity of posture estimation networks by widening and deepening them to improve performance. To address this issue, this paper proposes a lightweight human posture estimation network, ADNet, based on the high-resolution network(HRNet). ADNet replaces Basic Blocks and downsampling in the high-resolution network with GABlocks and GDBlocks that incorporate lightweight convolution and attention mechanisms to reduce the number of parameters and computational complexity in the network. Additionally, ADNet solves the problem of feature redundancy by fusing relationships between different channels to make effective feature channels fully utilized. Experimental results show that ADNet achieves high-precision detection of human keypoint information even with fewer parameters and less computational complexity than commonly used human posture estimation networks, such as Cascaded Pyramid Network (CPN), Stacked Hourglass Network(Hourglass), and HRNet. Results also indicate that, compared to HRNet, ADNet reduces the model parameter size by 69.1% and computational complexity by 58.9% under the same image resolution and environment configuration. Furthermore, in the visualization of feature maps, the network using ADNet exhibits stronger feature extraction performance and less feature redundancy.
- 1 June 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 60 (01962892) , 1-15
Unsupervised Hyperspectral Band Selection via Hybrid Graph Convolutional Network
Hyperspectral image (HSI) provided with a substantial number of correlated bands causes calculation consumption and an undesirable “dimension disaster” problem for the classification. Band selection (BS) is an effective measure to reduce the information redundancy with the physics spectrum preserved for HSI. Although the existing BS methods have achieved noticeable progress, the correlation between neighbor bands still needs to be mined deeply for an effective selection criterion. This article proposes a BS approach to collecting the discriminative band subset for HSI classification (HSIC), which adopts the self-supervised learning paradigm to implement the BS by the auxiliary spectrum rebuilding (SR) task. In specific, we utilized a convolutional neural network (CNN) and a graph convolutional network (GCN) for the spectral–spatial feature extraction. Next, GCN and CNN are developed for the refinement of the band correlation sequentially. Afterward, the selected bands in terms of the acquired correlation are fed into the presented self-supervised SR network for spectral reconstruction. Simultaneously, the proposed architecture completed the selection with the optimization of the band reconstruction by a defined loss function. In this way, we supply substitution for selection criterion and path searching through the end-to-end framework. The extensive experimental results and analysis demonstrated that the proposed hybrid architecture provided a competitive band subset for the classification, and the accuracies with different types of classifiers are more effective than the compared BS methods.
- 1 June 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Geoscience and Remote Sensing
- Vol. 60 (01962892) , 1-15
Unsupervised Hyperspectral Band Selection via Hybrid Graph Convolutional Network
Hyperspectral image (HSI) provided with a substantial number of correlated bands causes calculation consumption and an undesirable “dimension disaster” problem for the classification. Band selection (BS) is an effective measure to reduce the information redundancy with the physics spectrum preserved for HSI. Although the existing BS methods have achieved noticeable progress, the correlation between neighbor bands still needs to be mined deeply for an effective selection criterion. This article proposes a BS approach to collecting the discriminative band subset for HSI classification (HSIC), which adopts the self-supervised learning paradigm to implement the BS by the auxiliary spectrum rebuilding (SR) task. In specific, we utilized a convolutional neural network (CNN) and a graph convolutional network (GCN) for the spectral–spatial feature extraction. Next, GCN and CNN are developed for the refinement of the band correlation sequentially. Afterward, the selected bands in terms of the acquired correlation are fed into the presented self-supervised SR network for spectral reconstruction. Simultaneously, the proposed architecture completed the selection with the optimization of the band reconstruction by a defined loss function. In this way, we supply substitution for selection criterion and path searching through the end-to-end framework. The extensive experimental results and analysis demonstrated that the proposed hybrid architecture provided a competitive band subset for the classification, and the accuracies with different types of classifiers are more effective than the compared BS methods.
- 20 January 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Letters
- Vol. 19 (1545598X) , 1-5
Semisupervised Hyperspectral Band Selection Based on Dual-Constrained Low-Rank Representation
Band selection (BS) aims to choose a salient subset implied sufficient information from the numerous bands, which supplies a significantly efficient way to alleviate the barrier of dimensionality disaster for hyperspectral image classification (HSIC). This letter develops a semisupervised BS approach based on dual-constrained low-rank representation BS (DCLRR-BS) with two regularizations for HSIC. To be specific, a low-rank representation model is first proposed with super-pixel and imbalanced class-wise constraints, which are explicitly integrated to improve the performance of the band description. Next, the clusters are built adaptively based on graph theory in an unsupervised manner to rapid selection efficiency. A selection criterion is last designed to highlight the prominent band of each subset cluster to fulfill the BS procedure. Experimental results conducted on four types of classifiers with two real hyperspectral image (HSI) data sets demonstrate that the proposed DCLRR-BS method performs well in the imbalanced HSIC area.
- 20 January 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Geoscience and Remote Sensing Letters
- Vol. 19 (1545598X) , 1-5
Semisupervised Hyperspectral Band Selection Based on Dual-Constrained Low-Rank Representation
Band selection (BS) aims to choose a salient subset implied sufficient information from the numerous bands, which supplies a significantly efficient way to alleviate the barrier of dimensionality disaster for hyperspectral image classification (HSIC). This letter develops a semisupervised BS approach based on dual-constrained low-rank representation BS (DCLRR-BS) with two regularizations for HSIC. To be specific, a low-rank representation model is first proposed with super-pixel and imbalanced class-wise constraints, which are explicitly integrated to improve the performance of the band description. Next, the clusters are built adaptively based on graph theory in an unsupervised manner to rapid selection efficiency. A selection criterion is last designed to highlight the prominent band of each subset cluster to fulfill the BS procedure. Experimental results conducted on four types of classifiers with two real hyperspectral image (HSI) data sets demonstrate that the proposed DCLRR-BS method performs well in the imbalanced HSIC area.
- 1 January 2012
- Published by Springer Nature
- p. 389-393
Study on the Comparison of Driving Configuration for Hybrid SUVs
- 1 January 2012
- Published by Springer Nature
- p. 389-393
Study on the Comparison of Driving Configuration for Hybrid SUVs