Detection method and validation analysis based on the improved YOLOX-s apple blossom growth state

doi:10.13733/j.jcam.issn.2095-5553.2023.08.022

Abstract

Abstract: In order to realize intelligent blossom thinning in apple orchards, this paper proposes an improved YOLOX-s apple blossom growth state detection method. First, the apple blossom dataset was collected and established for the training and verification of the network model, and then the YOLOX-s model was built, and the backbone network was improved. The Convolutional Block Attention Module (CBAM) attention mechanism module was added to the two layers after the output features, EIOU was used as the regression function of the model and the Focal Loss function was introduced in the postprocessing stage to improve the models ability to detect clustered apple blossoms and improve the average accuracy of the model. The results showed that the accuracy of the improved YOLOX-s model was 91.75%, which was 0.5% higher in precision, 6.19% higher in recall rate, and 4.28% higher in average precision than before. This research provides technical support for the realization of intelligent detection of apple blossom growth status to guide the accurate decision of intelligent blossom thinning.

Key words: apple flower detection, YOLOX, intelligent orchard

摘要： 为实现苹果园智能疏花，提出一种基于改进YOLOX-s的苹果花生长状态检测方法。采集并建立苹果花数据集，以用于网络模型的训练和验证，搭建YOLOX-s模型，对主干网络进行改进，在输出特征后两层加入CBAM注意力机制模块，采用EIOU作为模型的回归函数，在后处理阶段引入Focal Loss损失函数，以提高模型对拥簇苹果花的检测能力，提高模型的平均精度。结果表明改进后的YOLOX-s模型精确度为91.75%，相比未改进前提升0.5%，召回率提升6.19%，平均精度提高4.28%。该研究为实现苹果花生长状态智能检测提供技术支持，指导智能化疏花精准决策。

关键词: 苹果花检测, YOLOX, 智能化果园

CLC Number:

高昂, 卢传兵, 任龙龙, 李玲, 沈向, 宋月鹏, . 基于改进YOLOX-s的苹果花生长状态检测方法及验证分析[J]. 中国农机化学报, 2023, 44(8): 162-167.

References

［1］　杨涛, 李晓晓. 机器视觉技术在现代农业生产中的研究进展［J］. 中国农机化学报, 2021, 42(3): 171-181.
Yang Tao, Li Xiaoxiao. Research progress of machine vision technology in modern agricultural production ［J］. Journal of Chinese Agricultural Mechanization, 2021, 42(3): 171-181.
［2］　陈国防, 陈兆英, 王玉亮, 等. 基于数据增强深度学习的苹果花检测方法研究［J］. 中国农机化学报, 2022, 43(5): 148-155.
Chen Guofang, Chen Zhaoying, Wang Yuliang, et al. Research on detection method of apple flower based on dataenhanced deep learning［J］. Journal of Chinese Agricultural Mechanization, 2022, 43(5): 148-155.
［3］　杨其晟, 李文宽, 杨晓峰, 等. 改进YOLOv5的苹果花生长状态检测方法［J］. 计算机工程与应用, 2022, 58(4): 237-246.
Yang Qisheng, Li Wenkuan, Yang Xiaofeng, et al. Improved YOLOv5 method for detecting growth status of apple flowers［J］. Computer Engineering and Applications, 2022, 58(4): 237-246.
［4］　Sun C H. Apple, peach, and pear flower detection using semantic segmentation network and shape constraint level set［J］. Computers and Electronics in Agriculture, 2021, 185(1).
［5］　尚钰莹, 张倩如, 宋怀波. 基于YOLOv5s的深度学习在自然场景苹果花朵检测中的应用［J］. 农业工程学报, 2022, 38(9): 222-229.
Shang Yuying, Zhang Qianru, Song Huaibo. Application of deep learning using YOLOv5s to apple flower detection in natural scenes ［J］. Transactions of the Chinese Society of Agricultural Engineering, 2022, 38(9): 222-229.
［6］　邱天衡, 王玲, 王鹏, 等. 基于改进YOLOv5的目标检测算法研究［J］. 计算机工程与应用, 2022, 58(13): 63-73.
Qiu Tianheng, Wang Ling, Wang Peng, et al. Research on object detection algorithm based on improved YOLOv5 ［J］. Computer Engineering and Applications, 2022, 58(13): 63-73.
［7］　李登峰, 高明, 叶文韬. 结合轻量级特征提取网络的舰船目标检测算法［J/OL］. 计算机工程与应用. https:∥kns.cnki.net/kcms/detail/11.2127.TP.20220909.1410.016.html.
Li Dengfeng, Gao Ming, Ye Wentao, et al. Ship target detection algorithm combined with lightweight feature extraction network［J/OL］. Computer Engineering and Applications. https:∥kns. cnki.net/kcms/detail/11.2127.TP.20220909.1410.016.html.
［8］　殷献博, 邓小玲, 兰玉彬, 等. 基于改进YOLOX-Nano算法的柑橘梢期长势智能识别［J］. 华南农业大学学报, 2023(1): 142-150.
Yin Xianbo, Deng Xiaoling, Lan Yubin, et al. Intelligent identification of citrus shoot growth based on improved YoloxNano algorithm ［J］. Journal of South China Agricultural University, 2023(1): 142-150.
［9］　张盼盼, 李其申, 杨词慧. 基于轻量级分组注意力模块的图像分类算法［J］. 计算机应用, 2020, 40(3): 645-650.
Zhang Panpan， Li Qishen， Yang Cihui. Image classification algorithm based on lightweight groupwise attention module［J］. Journal of Computer Applications, 2020, 40(3): 645-650.
［10］　Li R, Wang S, Wang Z, et al. Breast cancer Xray image staging: Based on efficient net with multiscale fusion and CBAM attention［J］. Journal of Physics: Conference Series. 2021, 2082(1): 012006.
［11］　江俊君, 李震宇, 刘贤明. 基于深度学习的单目深度估计方法综述［J］. 计算机学报, 2022, 45(6): 1276-1307.
Jiang Junjun, Li Zhenyu, Liu Xianming. Survey of monocular depth estimation methods based on deep learning ［J］. Chinese Journal of Computers, 2022, 45(6): 1276-1307.
［12］　Leng H, Tan M, Liu C, et al. PolyLoss: A polynomial expansion perspective of classification loss functions ［J］. In Proceedings of the Tenth International Conference on Learning Representations, Virtual, 25-29 April 2022.
［13］　Zheng Z, Wang P, Liu W, et al. DistanceIoU Loss: Faster and better learning for bounding box regression ［J］. Proceedings of the AAAI Conference on Artificial Intelligence, 2020, 34(7): 12993-13000.
［14］　Ren S, He K, Girshick, R, et al. Faster R-CNN: Towards realtime object detection with region proposal networks ［J］. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2017, 39(6): 1137-1149.
［15］　何颖, 陈丁号, 彭琳. 基于改进YOLOv5模型的经济林木虫害目标检测算法研究［J］. 中国农机化学报, 2022, 43(4): 106-115.
He Ying, Chen Dinghao, Peng Lin. Research on economic forest pest target detection algorithm based on improved YOLOv5 model［J］. Journal of Chinese Agricultural Mechanization, 2022, 43(4): 106-115.