Fengshuang Liu

Prof. Dr. Fengshuang Liu

orcid

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 March 2020 in Signal Processing
Verified
Journal Article
Closed access
Fengshuang LiuFengshuang Liu
Fengshuang Liu
QWQiang Wang
Qiang Wang

9 May 2025 in Biosystems Engineering
Verified
Journal Article
Closed access
JZJialiang Zhang
Jialiang Zhang
Fengshuang LiuFengshuang Liu
Fengshuang Liu
JFJun Fu
Jun Fu

31 October 2024 in Agriculture
Verified
Journal Article
Open access
PDF file
JLJindai Liu
Jindai Liu
Fengshuang LiuFengshuang Liu
Fengshuang Liu
JFJun Fu
Jun Fu
Maize is susceptible to pest disease, and the production of maize would suffer a significant decline without precise early detection. Hyperspectral imaging is well-suited for the precise detection of diseases due to its ability to capture the internal chemical characteristics of vegetation. However, the abundance of redundant information in hyperspectral data poses challenges in extracting significant features. To overcome the above problems, in this study we proposed an attention-based spatial-spectral joint network model for hyperspectral detection of pest-infected maize. The model contains 3D and 2D convolutional layers that extract features from both spatial and spectral domains to improve the identification capability of hyperspectral images. Moreover, the model is embedded with an attention mechanism that improves feature representation by focusing on important spatial and spectral-wise information and enhances the feature extraction ability of the model. Experimental results demonstrate the effectiveness of the proposed model across different field scenarios, achieving overall accuracies (OAs) of 99.24% and 97.4% on close-up hyperspectral images and middle-shot hyperspectral images, respectively. Even under the condition of a lack of training data, the proposed model performs a superior performance relative to other models and achieves OAs of 98.29% and 92.18%. These results proved the validity of the proposed model, and it is accomplished efficiently for pest-infected maize detection. The proposed model is believed to have the potential to be applied to mobile devices such as field robots in order to monitor and detect infected maize automatically.

12 February 2026 in Computers and Electronics in Agriculture
Verified
Journal Article
Closed access
JLJindai Liu
Jindai Liu
Fengshuang LiuFengshuang Liu
Fengshuang Liu
JFJun Fu
Jun Fu

25 February 2026 in Journal of Food Process Engineering
Verified
Journal Article
Closed access
YXYanpeng Xu
Yanpeng Xu
Fengshuang LiuFengshuang Liu
Fengshuang Liu
JFJun Fu
Jun Fu
CZChengyuan Zhang
Chengyuan Zhang
SLShujia Li
Shujia Li

9 October 2025 in Research Square
Verified
Preprint
Open access
PDF file
BCBen Che
Ben Che
JFJun Fu
Jun Fu
Fengshuang LiuFengshuang Liu
Fengshuang Liu
ZXZhao Xue
Zhao Xue
YZYiqiang Zheng
Yiqiang Zheng
JLJindai Liu
Jindai Liu
Efficient, precise measurement of chopped maize-stalk length is vital for biomass processing and sustainable residue use. Yet fragmented segments—with irregular shapes, curvature, overlap, and low contrast—challenge lightweight detectors and degrade length estimation. We propose a lightweight, two-stage hybrid framework: an optimized YOLOv8n first localizes stalk and calibration regions, and a calibrated geometric module then refines these ROIs to segment, skeletonize, and convert pixel arc length to millimeters. The detector integrates RepELAN for richer spatial features, Coordinate Attention for direction-aware sensitivity, an LSCD head for robust multi-scale prediction, and WIoU for precise box regression. On our dataset, the optimized YOLOv8n attains 96.90% detection accuracy and surpasses representative lightweight baselines. The ROI-based measurement stage applies contour enhancement and skeletonization, followed by washer-based pixel-to-length conversion, yielding a mean absolute error of 2.16 mm (3.63% relative error). Together, this integrated detection–measurement pipeline offers a practical solution for automated residue management, improving straw-utilization efficiency and supporting sustainable agricultural production.