Shixiu Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Shixiu Zhang has a total of 23 co-authors. Here are the top co-authors by number of shared publications:
- Xiuguo Zou
Nanjing Agricultural University
3 shared publications - Yan Qian
College of Artificial Intelligence, Nanjing Agricultural University, Nanjing 210031, China
2 shared publications - Yuanyuan Song
College of Engineering, Nanjing Agricultural University, Nanjing 210031, China
2 shared publications - Fei Gong
College of Artificial Intelligence, Nanjing Agricultural University, Nanjing 210031, China
1 shared publication - Hongfei Chen
College of Artificial Intelligence, Nanjing Agricultural University, Nanjing 210031, China
1 shared publication - Jiahong Wu
College of Artificial Intelligence, Nanjing Agricultural University, Nanjing 210031, China
1 shared publication
Latest Publications
- Open Access
- 12 July 2025
- Published by MDPI AG in Agriculture
- Vol. 15 (14) , 1504
DualHet-YOLO: A Dual-Backbone Heterogeneous YOLO Network for Inspection Robots to Recognize Yellow-Feathered Chicken Behavior in Floor-Raised House
The behavior of floor-raised chickens is closely linked to their health status and environmental comfort. As a type of broiler chicken with special behaviors, understanding the daily actions of yellow-feathered chickens is crucial for accurately checking their health and improving breeding practices. Addressing the challenges of high computational complexity and insufficient detection accuracy in existing floor-raised chicken behavior recognition models, a lightweight behavior recognition model was proposed for floor-raised yellow-feathered chickens, based on a Dual-Backbone Heterogeneous YOLO Network. Firstly, DualHet-YOLO enhances the feature extraction capability of floor-raised chicken images through a dual-path feature map extraction architecture and optimizes the localization and classification of multi-scale targets using a TriAxis Unified Detection Head. Secondly, a Proportional Scale IoU loss function is introduced that improves regression accuracy. Finally, a lightweight structure Eff-HetKConv was designed, significantly reducing model parameters and computational complexity. Experiments on a private floor-raised chicken behavior dataset show that, compared with the baseline YOLOv11 model, the DualHet-YOLO model increases the mAP for recognizing five behaviors—pecking, resting, walking, dead, and inactive—from 77.5% to 84.1%. Meanwhile, it reduces model parameters by 14.6% and computational complexity by 29.2%, achieving a synergistic optimization of accuracy and efficiency. This approach provides an effective solution for lightweight object detection in poultry behavior recognition.
- 12 July 2025
- Published by MDPI AG in Agriculture
- Vol. 15 (14) , 1504
Open AccessDualHet-YOLO: A Dual-Backbone Heterogeneous YOLO Network for Inspection Robots to Recognize Yellow-Feathered Chicken Behavior in Floor-Raised House
The behavior of floor-raised chickens is closely linked to their health status and environmental comfort. As a type of broiler chicken with special behaviors, understanding the daily actions of yellow-feathered chickens is crucial for accurately checking their health and improving breeding practices. Addressing the challenges of high computational complexity and insufficient detection accuracy in existing floor-raised chicken behavior recognition models, a lightweight behavior recognition model was proposed for floor-raised yellow-feathered chickens, based on a Dual-Backbone Heterogeneous YOLO Network. Firstly, DualHet-YOLO enhances the feature extraction capability of floor-raised chicken images through a dual-path feature map extraction architecture and optimizes the localization and classification of multi-scale targets using a TriAxis Unified Detection Head. Secondly, a Proportional Scale IoU loss function is introduced that improves regression accuracy. Finally, a lightweight structure Eff-HetKConv was designed, significantly reducing model parameters and computational complexity. Experiments on a private floor-raised chicken behavior dataset show that, compared with the baseline YOLOv11 model, the DualHet-YOLO model increases the mAP for recognizing five behaviors—pecking, resting, walking, dead, and inactive—from 77.5% to 84.1%. Meanwhile, it reduces model parameters by 14.6% and computational complexity by 29.2%, achieving a synergistic optimization of accuracy and efficiency. This approach provides an effective solution for lightweight object detection in poultry behavior recognition.
- Open Access
- 27 January 2022
- Published by MDPI AG in Agriculture
- Vol. 12 (2) , 182
Study of Ammonia Concentration Characteristics and Optimization in Broiler Chamber during Winter Based on Computational Fluid Dynamics
Poultry breeding is one of the most significant components of agriculture and an essential link of material exchange between humans and nature. Moreover, poultry breeding technology has a considerable impact on the life quality of human beings, and could even influence the survival of human beings. As one of the most popular poultry, broiler has a good economic benefit due to its excellent taste and fast growing cycle. This paper aims to improve the efficiency of raising broilers by understanding the impact of ammonia concentration distribution within a smart broiler breeding chamber, and the rationality of the system’s design. More specifically, we used computational fluid dynamics (CFD) technology to simulate the process of ammonia production and identify the characteristics of ammonia concentration. Based on the simulation results, the structure of the broiler chamber was reformed, and the ammonia uniformity was significantly improved after the structural modification of the broiler chamber and the ammonia concentration in the chamber had remained extremely low. In general, this study provides a reference for structural optimization of the design of broiler chambers and the environmental regulation of ammonia.
- 27 January 2022
- Published by MDPI AG in Agriculture
- Vol. 12 (2) , 182
Open AccessStudy of Ammonia Concentration Characteristics and Optimization in Broiler Chamber during Winter Based on Computational Fluid Dynamics
Poultry breeding is one of the most significant components of agriculture and an essential link of material exchange between humans and nature. Moreover, poultry breeding technology has a considerable impact on the life quality of human beings, and could even influence the survival of human beings. As one of the most popular poultry, broiler has a good economic benefit due to its excellent taste and fast growing cycle. This paper aims to improve the efficiency of raising broilers by understanding the impact of ammonia concentration distribution within a smart broiler breeding chamber, and the rationality of the system’s design. More specifically, we used computational fluid dynamics (CFD) technology to simulate the process of ammonia production and identify the characteristics of ammonia concentration. Based on the simulation results, the structure of the broiler chamber was reformed, and the ammonia uniformity was significantly improved after the structural modification of the broiler chamber and the ammonia concentration in the chamber had remained extremely low. In general, this study provides a reference for structural optimization of the design of broiler chambers and the environmental regulation of ammonia.
- Open Access
- 4 July 2021
- Published by MDPI AG in Atmosphere
- Vol. 12 (7) , 869
An Atmospheric Visibility Grading Method Based on Ensemble Learning and Stochastic Weight Average
In order to adequately characterize the visual characteristics of atmospheric visibility and overcome the disadvantages of the traditional atmospheric visibility measurement method with significant dependence on preset reference objects, high cost, and complicated steps, this paper proposed an ensemble learning method for atmospheric visibility grading based on deep neural network and stochastic weight averaging. An experiment was conducted using the scene of an expressway, and three visibility levels were set, i.e., Level 1, Level 2, and Level 3. Firstly, the EfficientNet was transferred to extract the abstract features of the images. Then, training and grading were performed on the feature sets through the SoftMax regression model. Subsequently, the feature sets were ensembled using the method of stochastic weight averaging to obtain the atmospheric visibility grading model. The obtained datasets were input into the grading model and tested. The grading model classified the results into three categories, with the grading accuracy being 95.00%, 89.45%, and 90.91%, respectively, and the average accuracy of 91.79%. The results obtained by the proposed method were compared with those obtained by the existing methods, and the proposed method showed better performance than those of other methods. This method can be used to classify the atmospheric visibility of traffic and reduce the incidence of traffic accidents caused by atmospheric visibility.
- 4 July 2021
- Published by MDPI AG in Atmosphere
- Vol. 12 (7) , 869
Open AccessAn Atmospheric Visibility Grading Method Based on Ensemble Learning and Stochastic Weight Average
In order to adequately characterize the visual characteristics of atmospheric visibility and overcome the disadvantages of the traditional atmospheric visibility measurement method with significant dependence on preset reference objects, high cost, and complicated steps, this paper proposed an ensemble learning method for atmospheric visibility grading based on deep neural network and stochastic weight averaging. An experiment was conducted using the scene of an expressway, and three visibility levels were set, i.e., Level 1, Level 2, and Level 3. Firstly, the EfficientNet was transferred to extract the abstract features of the images. Then, training and grading were performed on the feature sets through the SoftMax regression model. Subsequently, the feature sets were ensembled using the method of stochastic weight averaging to obtain the atmospheric visibility grading model. The obtained datasets were input into the grading model and tested. The grading model classified the results into three categories, with the grading accuracy being 95.00%, 89.45%, and 90.91%, respectively, and the average accuracy of 91.79%. The results obtained by the proposed method were compared with those obtained by the existing methods, and the proposed method showed better performance than those of other methods. This method can be used to classify the atmospheric visibility of traffic and reduce the incidence of traffic accidents caused by atmospheric visibility.