Ge Song
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Ge Song has a total of 25 co-authors. Here are the top co-authors by number of shared publications:
- Jiaxin Chen
College of Mathematics and Informatics, College of Software Engineering, South China Agricultural University, Guangzhou, China
2 shared publications - Xiaofei Yang
School of Electronics and Communication Engineering, Guangzhou University, Guangzhou 510006, China
2 shared publications - Xiaoqi Luo
College of Mathematics and Informatics, College of Software Engineering, South China Agricultural University, Guangzhou, China
2 shared publications - Yuqiao Deng
School of Statistics and Mathematics, Guangdong University of Finance and Economics, Guangzhou 510120, China
2 shared publications - Zhihong Zhang
College of Mathematics and Informatics, College of Software Engineering, South China Agricultural University, Guangzhou, China
2 shared publications - Beibei Du
School of Artificial Intelligence and Robotics, Hunan University, Changsha, China
1 shared publication
Latest Publications
- Open Access
- 23 October 2026
- Published by Springer Nature in Journal on Image and Video Processing
CFN: a cross-fusion network for hyperspectral image classification
- 23 October 2026
- Published by Springer Nature in Journal on Image and Video Processing
Open AccessCFN: a cross-fusion network for hyperspectral image classification
- Open Access
AWGLFuser: A Global–Local Feature Fusion Network with Adaptive Wavelet Filter for Lithium-Ion Battery State-of-Health Estimation
Accurate estimation of the State of Health (SOH) of lithium-ion batteries is critical to ensuring the safety and stability of energy storage and power systems. However, existing SOH evaluation methods fall short in adaptive denoising, multi-scale feature extraction, and effective fusion of global and local information. To overcome these limitations, this paper proposes a global–local feature fusion network with an adaptive wavelet filter (AWGLFuser) for end-to-end SOH estimation. The proposed model consists of three modules: an adaptive wavelet filter (AWF) module to suppress high-frequency noise and highlight key information in the frequency domain; a global–local feature extraction (GLFE) module to capture both global and local temporal dependencies at multiple scales; and a bidirectional cross-attention fusion (BCAF) module to enable deep interaction between global and local features, thereby facilitating their effective fusion. Comparison experiments on the NASA and XJTU datasets demonstrate that the proposed model yields lower estimation errors than the eight competing models. The average reductions in MAE, MAPE, and RMSE are 44.158%, 44.209%, and 38.014%, respectively, and this improvement is statistically significant against every comparison model. Furthermore, ablation studies clarify what each component contributes, with the modules proving mutually reinforcing when combined. AWGLFuser also attains a compact parameter count and storage footprint with competitive inference latency, despite comparatively higher FLOPs.
Open AccessAWGLFuser: A Global–Local Feature Fusion Network with Adaptive Wavelet Filter for Lithium-Ion Battery State-of-Health Estimation
Accurate estimation of the State of Health (SOH) of lithium-ion batteries is critical to ensuring the safety and stability of energy storage and power systems. However, existing SOH evaluation methods fall short in adaptive denoising, multi-scale feature extraction, and effective fusion of global and local information. To overcome these limitations, this paper proposes a global–local feature fusion network with an adaptive wavelet filter (AWGLFuser) for end-to-end SOH estimation. The proposed model consists of three modules: an adaptive wavelet filter (AWF) module to suppress high-frequency noise and highlight key information in the frequency domain; a global–local feature extraction (GLFE) module to capture both global and local temporal dependencies at multiple scales; and a bidirectional cross-attention fusion (BCAF) module to enable deep interaction between global and local features, thereby facilitating their effective fusion. Comparison experiments on the NASA and XJTU datasets demonstrate that the proposed model yields lower estimation errors than the eight competing models. The average reductions in MAE, MAPE, and RMSE are 44.158%, 44.209%, and 38.014%, respectively, and this improvement is statistically significant against every comparison model. Furthermore, ablation studies clarify what each component contributes, with the modules proving mutually reinforcing when combined. AWGLFuser also attains a compact parameter count and storage footprint with competitive inference latency, despite comparatively higher FLOPs.
- Top Cited PapersOpen Access
- 16 January 2026
- Published by MDPI AG in Agriculture
- Vol. 16 (2) , 234
DRPU-YOLO11: A Multi-Scale Model for Detecting Rice Panicles in UAV Images with Complex Infield Background
In the field of precision agriculture, accurately detecting rice panicles is crucial for monitoring rice growth and managing rice production. To address the challenges posed by complex field backgrounds, including variety differences, variations across growth stages, background interference, and occlusion due to dense distribution, this study develops an improved YOLO11-based rice panicle detection model, termed DRPU-YOLO11. The model incorporates a task-oriented CSP-PGMA module in the backbone to enhance multi-scale feature extraction and provide richer representations for downstream detection. In the neck network, DySample and CGDown are adopted to strengthen global contextual feature aggregation and suppress background interference for small targets. Furthermore, fine-grained P2 level information is integrated with higher-level features through a cross-scale fusion module (CSP-ONMK) to improve detection robustness in dense and occluded scenes. In addition, the PowerTAL strategy adapts quality-aware label assignment to emphasize high-quality predictions during training. The experimental results based on a self-constructed dataset demonstrate that DRPU-YOLO11 significantly outperforms baseline models in rice panicle detection under complex field environments, achieving an accuracy of 82.5%. Compared with the baseline model YOLO11 and RT-DETR, the mAP50 increases by 2.4% and 5.0%, respectively. These results indicate that the proposed task-driven design provides a practical and high-precision solution for rice panicle detection, with potential applications in rice growth monitoring and yield estimation.
- 16 January 2026
- Published by MDPI AG in Agriculture
- Vol. 16 (2) , 234
Top Cited PapersOpen AccessDRPU-YOLO11: A Multi-Scale Model for Detecting Rice Panicles in UAV Images with Complex Infield Background
In the field of precision agriculture, accurately detecting rice panicles is crucial for monitoring rice growth and managing rice production. To address the challenges posed by complex field backgrounds, including variety differences, variations across growth stages, background interference, and occlusion due to dense distribution, this study develops an improved YOLO11-based rice panicle detection model, termed DRPU-YOLO11. The model incorporates a task-oriented CSP-PGMA module in the backbone to enhance multi-scale feature extraction and provide richer representations for downstream detection. In the neck network, DySample and CGDown are adopted to strengthen global contextual feature aggregation and suppress background interference for small targets. Furthermore, fine-grained P2 level information is integrated with higher-level features through a cross-scale fusion module (CSP-ONMK) to improve detection robustness in dense and occluded scenes. In addition, the PowerTAL strategy adapts quality-aware label assignment to emphasize high-quality predictions during training. The experimental results based on a self-constructed dataset demonstrate that DRPU-YOLO11 significantly outperforms baseline models in rice panicle detection under complex field environments, achieving an accuracy of 82.5%. Compared with the baseline model YOLO11 and RT-DETR, the mAP50 increases by 2.4% and 5.0%, respectively. These results indicate that the proposed task-driven design provides a practical and high-precision solution for rice panicle detection, with potential applications in rice growth monitoring and yield estimation.
- Open Access
- 7 April 2025
- Published by MDPI AG in Electronics
- Vol. 14 (7) , 1477
MDFFN: Multi-Scale Dual-Aggregated Feature Fusion Network for Hyperspectral Image Classification
Employing the multi-scale strategy in hyperspectral image (HSI) classification enables the exploration of complex land-cover structures with diverse shapes. However, existing multi-scale methods still have limitations for fine feature extraction and deep feature fusion, which hinder the further improvement of classification performance. In this paper, we propose a multi-scale dual-aggregated feature fusion network (MDFFN) for both balanced and imbalanced environments. The network comprises two main core modules: a multi-scale convolutional information embedding (MCIE) module and a dual aggregated cross-attention (DACA) module. The proposed MCIE module introduces a multi-scale pooling operation to aggregate local features, which efficiently highlights discriminative spectral–spatial information and especially learns key features in small target samples in the imbalanced environment. Furthermore, the proposed DACA module employs a cross-scale interaction strategy to realize the deep fusion of multi-scale features and designs a dual aggregation mechanism to mitigate the loss of information, which facilitates further spatial–spectral feature enhancement. The experimental results demonstrate that the proposed method outperforms state-of-the-art methods on three classical HSI datasets, proving the superiority of the proposed MDFFN.
- 7 April 2025
- Published by MDPI AG in Electronics
- Vol. 14 (7) , 1477
Open AccessMDFFN: Multi-Scale Dual-Aggregated Feature Fusion Network for Hyperspectral Image Classification
Employing the multi-scale strategy in hyperspectral image (HSI) classification enables the exploration of complex land-cover structures with diverse shapes. However, existing multi-scale methods still have limitations for fine feature extraction and deep feature fusion, which hinder the further improvement of classification performance. In this paper, we propose a multi-scale dual-aggregated feature fusion network (MDFFN) for both balanced and imbalanced environments. The network comprises two main core modules: a multi-scale convolutional information embedding (MCIE) module and a dual aggregated cross-attention (DACA) module. The proposed MCIE module introduces a multi-scale pooling operation to aggregate local features, which efficiently highlights discriminative spectral–spatial information and especially learns key features in small target samples in the imbalanced environment. Furthermore, the proposed DACA module employs a cross-scale interaction strategy to realize the deep fusion of multi-scale features and designs a dual aggregation mechanism to mitigate the loss of information, which facilitates further spatial–spectral feature enhancement. The experimental results demonstrate that the proposed method outperforms state-of-the-art methods on three classical HSI datasets, proving the superiority of the proposed MDFFN.
- 7 March 2024
- Published by Springer Nature in Cluster Computing
- Vol. 27 (5) , 6773-6786
A coupled generative graph convolution network by capturing dynamic relationship of regional flow for traffic prediction
- 7 March 2024
- Published by Springer Nature in Cluster Computing
- Vol. 27 (5) , 6773-6786
A coupled generative graph convolution network by capturing dynamic relationship of regional flow for traffic prediction
- Open Access
- 1 January 2023
- Published by Tech Science Press in Computers, Materials & Continua
- Vol. 77 (3) , 4055-4072
DNEF: A New Ensemble Framework Based on Deep Network Structure
Deep neural networks have achieved tremendous success in various fields, and the structure of these networks is a key factor in their success. In this paper, we focus on the research of ensemble learning based on deep network structure and propose a new deep network ensemble framework (DNEF). Unlike other ensemble learning models, DNEF is an ensemble learning architecture of network structures, with serial iteration between the hidden layers, while base classifiers are trained in parallel within these hidden layers. Specifically, DNEF uses randomly sampled data as input and implements serial iteration based on the weighting strategy between hidden layers. In the hidden layers, each node represents a base classifier, and multiple nodes generate training data for the next hidden layer according to the transfer strategy. The DNEF operates based on two strategies: (1) The weighting strategy calculates the training instance weights of the nodes according to their weaknesses in the previous layer. (2) The transfer strategy adaptively selects each node’s instances with weights as transfer instances and transfer weights, which are combined with the training data of nodes as input for the next hidden layer. These two strategies improve the accuracy and generalization of DNEF. This research integrates the ensemble of all nodes as the final output of DNEF. The experimental results reveal that the DNEF framework surpasses the traditional ensemble models and functions with high accuracy and innovative deep ensemble methods.
- 1 January 2023
- Published by Tech Science Press in Computers, Materials & Continua
- Vol. 77 (3) , 4055-4072
Open AccessDNEF: A New Ensemble Framework Based on Deep Network Structure
Deep neural networks have achieved tremendous success in various fields, and the structure of these networks is a key factor in their success. In this paper, we focus on the research of ensemble learning based on deep network structure and propose a new deep network ensemble framework (DNEF). Unlike other ensemble learning models, DNEF is an ensemble learning architecture of network structures, with serial iteration between the hidden layers, while base classifiers are trained in parallel within these hidden layers. Specifically, DNEF uses randomly sampled data as input and implements serial iteration based on the weighting strategy between hidden layers. In the hidden layers, each node represents a base classifier, and multiple nodes generate training data for the next hidden layer according to the transfer strategy. The DNEF operates based on two strategies: (1) The weighting strategy calculates the training instance weights of the nodes according to their weaknesses in the previous layer. (2) The transfer strategy adaptively selects each node’s instances with weights as transfer instances and transfer weights, which are combined with the training data of nodes as input for the next hidden layer. These two strategies improve the accuracy and generalization of DNEF. This research integrates the ensemble of all nodes as the final output of DNEF. The experimental results reveal that the DNEF framework surpasses the traditional ensemble models and functions with high accuracy and innovative deep ensemble methods.