Crop Recognition Based on Improved YOLOv9

DOI: https://doi.org/10.62517/jike.202504110

Author(s)

Li Xiang, Jiao Xiangjian, Jia Maotang

Affiliation(s)

School of Intelligence Science and Technology, Xinjiang University, China

Abstract

To overcome challenges associated with crop species identification in unmanned aerial vehicle (UAV) imagery—specifically missed detections of small-sized targets, disturbances caused by intricate environmental elements, and suboptimal computational efficiency of detection models—this study introduces a streamlined detection framework that integrates YOLOv9 with Convolutional Block Attention Module (CBAM) and Spatial Channel Reconstruction Convolution (SCConv). Initial modifications involve embedding CBAM within the YOLOv9 backbone architecture. This enhancement exploits the collaborative functioning of channel-wise and spatial attention mechanisms to amplify the model's sensitivity toward critical localized crop characteristics, such as panicle morphological structures and foliar texture patterns, while concurrently mitigating noise interference from soil surfaces and weed vegetation. Subsequent improvements involve substituting conventional convolutional layers with Spatial Channel Reconstruction Convolution. This substitution capitalizes on the module's adaptive feature reorganization capabilities to achieve parameter reduction without compromising feature representation capacity, thereby substantially enhancing operational efficiency during edge device deployment. Empirical evaluations conducted on a proprietary dataset comprising 1,000 UAV-captured images representing seven distinct crop types demonstrate that the optimized model attains a mean Average Precision (mAP) of 94.5% when applying a 50% Intersection over Union threshold—a performance gain of 2.2 percentage points compared to the baseline YOLOv9 architecture. These results confirm the system's capability to meet real-time recognition demands within complex agricultural landscapes. The proposed methodology presents a cost-effective, high-accuracy monitoring solution for UAV-based precision agriculture applications, holding significant practical value for advancing intelligent management systems in modern agricultural practices.

Keywords

Unmanned Aerial Vehicle; Crop; Yolov9; Attention Module; Spatial Channel Reconstruction

References

[1].N. Falco et al., "Remote Sensing to Uav-Based Digital Farmland," IGARSS 2018 - 2018 IEEE International Geoscience and Remote Sensing Symposium, Valencia, Spain, 2018, pp. 5936-5939, [2].Y. Pratama, I. Iskandar and P. T. P. Giawa, "Implementation of Convolutional Neural Network on Farming Robots for Detecting Broccoli," 2022 International Conference on ICT for Smart Society (ICISS), Bandung, Indonesia, 2022, pp. 01-07, [3].Chen Jinlin, Wu Yiquan, Yuan Yubin. Research progress on YOLO series algorithms for target detection from UAV perspective [J/OL]. Journal of Beihang University, 1-33[2025-03-24]. [4].Sun Daming, Xing Lulu, Zhao Wei, et al. Application of YOLO series object detection algorithm in agricultural production [J]. Information and Computer (Theoretical Edition), 2024,36(12):23-26. [5].D. Zhu, L. Dai and P. Du, "CCE-YOLOv5s: An Improved YOLOv5 Model for UAV Small Target Detection," 2023 IEEE 5th International Conference on Civil Aviation Safety and Information Technology (ICCASIT), Dali, China, 2023, pp. 824-829, [6].Wang Ru, Lin Haojie, Huang Wei, et al. Detection method of ancient building blue brick damage with integrated attention mechanism [J]. Journal of Safety and Environment, 2024,24(11):4244-4252. [7].Wenxia B, Xinghua Y, Dong L, et al.Lightweight convolutional neural network model for field wheat ear disease identification [J]. Computers and Electronics in Agriculture, 2021,189 (Lightweight convolution) [8].K. Sangaiah, F. -N. Yu, Y. -B. Lin, W. -C. Shen and A. Sharma, "UAV T-YOLO-Rice: An Enhanced Tiny Yolo Networks for Rice Leaves Diseases Detection in Paddy Agronomy," in IEEE Transactions on Network Science and Engineering, vol. 11, no. 6, pp. 5201-5216,Nov.-Dec. 2024 (Agricultural drones) [9].Yuchen L ,Fuzheng L ,Mingshun J .Research on YOLO algorithm for lightweight PCB defect detection based on MobileViT[J/OL].Optoelectronics Letters,1-8[2025-03-25]. [10].Guo Xin, Ma Xing, Mu Chunyang, et al. Lightweight steel wire rope defect detection method based on YOLOv8-SLA [J]. Combined Machine Tool and Automation Processing Technology, 2025, (03):140-144. [11].Cao Jiayu, Qiao Guifang, Chen Mengyuan, et al. Improvement strategies of YOLO algorithm for small target detection from high altitude perspective [J]. Computer Applications, 2024,44(S2):280-285. [12].K. Kiropoulos, F. Dimaraki, D. C. Tsouros and S. Bibi, "UAV saffran monitoring process," 2022 11th International Conference on Modern Circuits and Systems Technologies (MOCAST), Bremen, Germany, 2022, pp. 1-4 (drone field monitoring) [13].Gong Xulu, Zhang Shujian. A light quantitative detection method for small target diseases in apple leaves based on improved YOLOv5s [J]. Journal of Agricultural Engineering, 2023,39(19):175-184. [14].Jia Xueying, Zhao Chunjiang, Zhou Juan, et al. Online detection of citrus surface defects based on improved YOLOv7 model [J]. Journal of Agricultural Engineering, 2023,39(23):142-151. [15].Yena, Wu Chenwen, Jiang Jialin. A lung sound classification model with spatial-channel reconstruction convolution module [J]. Journal of Southern Medical University, 2024,44(09):1720-1728. [16].Hu Chengcai, Chen Yuyang, Du Xinlin, et al. A significant object detection method based on composite attention mechanism [J/OL]. Software Guide, 1-11[2025-03-24].