首页分享基于改进YoloV4的轻量化菠萝苗心检测算法

基于改进YoloV4的轻量化菠萝苗心检测算法

来源：花匠小妙招时间：2024-12-15 17:38

摘要: 当前菠萝催花作业以人工喷洒为主，生产效率低、劳动强度大。菠萝苗心位置的精准识别和定位是实现机械化、智能化菠萝催花的核心问题。该研究在YoloV4目标识别算法的基础上，选择GhostNet作为主干特征提取网络，构建了一种混合网络模型，并在颈部网络中融合深度可分离卷积与轻量级的注意力模块。改进后的模型相较于YoloV4模型的总参数量减少70%。与YoloV4、Faster R-CNN和CenterNet 3个模型进行检测对比试验，结果可得：改进模型在菠萝植株种植密集与稀疏的条件下识别精度分别为94.7%和95.5%，实时识别速度可达27帧/s，每张图像平均检测时间为72 ms，相比常规YoloV4模型用时缩短23%。总体性能表现均优于对比组的目标检测模型。总的来说，改进模型YoloV4-GHDW在一定程度上实现了检测速度、识别精度和模型体量三者之间平衡，能够在实际种植环境下对菠萝苗心有较好的识别效果。研究结果可为智能化菠萝精准催花设备研发提供视觉技术支持。

Abstract: Abstract: Spraying is one of the main physiological and biochemical processes for flower induction in pineapple production. However, manual spraying cannot fully meet large-scale production in recent years, due to the very high labor intensity. Fortunately, automatic spraying can be expected for the flower induction of pineapple during this time. Among them, it is a high demand to accurately detect the center position of pineapple plants during spraying. In this study, a hybrid network model was proposed to accurate recognize the pineapple plant center using an improved YoloV4 deep learning. The conventional YoloV4 model shared an outstanding detect speed more suitable for the agricultural robot. However, the large number of parameters occupied a huge amount of memory in the original model, leading to more computing power in the device. Therefore, it was necessary for a much lighter and more flexible network structure in the reasoning work, particularly with the shorter running time in the actual field operation. The lightweight was then realized using the network framework of the YoloV4 model. Firstly, the number of network layers and feature map channels in the backbone were reduced using the GhostNet network structure as the backbone to replace the CSP-Darknet53, which enhanced the efficiency of the feature extraction. The GhostNet presented a Ghost module for building efficient neural architectures, in order to reduce the computational costs of recent deep neural networks. As such, the Ghost bottlenecks performed better in the lightweight structures of the entire network. Secondly, the depthwise separable convolution was introduced into the Neck network. The convolution set was integrated to reduce the number of parameters with a high level of feature extraction. Besides, the lightweight attention mechanism (called CBAM) was added between the backbone and the neck. Attention weights were then put on the channel and spatial dimensions for a better connection in the channel and space, in order to extract the effective features of the target. Finally, the quantity of the improved model (named Yolov4-GHDW) parameters was reduced by 70%, compared with the original. In the dataset-making stage, some operations were selected to process the collected images for the better generalization ability of the training model, including the vertical flip, adding Gaussian blur, adding salt and pepper noise, and extracting green features. Among them, 2530 images were divided into 1962, 218, and 350 images for the training set, the validation set, and the test set. The collected image dataset was used for the model training at the workstation, where the loss value of the training set and test set were recorded. The model with the smallest loss value was then selected as the optimal solution after training. The results show that the Yolov4-GHDW model presented a much higher recognition speed, and an excellent extraction performance to the target, compared with the original YoloV4, Faster R-CNN, and CenterNet. Specifically, the average accuracies of target recognition under the dense and sparse planting of pineapple plants were 94.7% and 95.5%, respectively, with only 68.4 MB of model memory. The real-time recognition speed reached 27 frames/s, and the average inference time of each image was 72ms, which was 23% shorter than that of YoloV4. Over all, the improved model can realize the identification and detection of pineapple plant center under complex conditions. The performance can fully meet the location requirement of the equipment during operation. Moreover, the improved model was also deployed in an embedded device (named Jetson Nano) to verify the performance. As a result, 13 frames per second was achieved to fluently operate with low power consumption. In conclusion, the Yolov4-GHDW has a high accuracy and recognition speed with a small memory footprint, indicating the balance between speed and accuracy. The finding can also provide technical support for intelligent and accurate flowering equipment in pineapple production.

[1] 邓春梅，李玉萍，梁伟红，等. 我国菠萝产业发展现状及对策[J]. 山西农业科学，2018，46(6)：1031-1034.DENG Chunmei, LI Yuping, LIANG Weihong, et al. Present situation and countermeasures of pineapple industry in China[J]. Journal of Shanxi Agricultural Sciences, 2018, 388(6): 169-172. (in Chinese with English abstract) [2] 张红娜，刘胜辉，孙伟生，等. 菠萝对乙烯利诱花的敏感性差异研究[J]. 热带作物学报，2018，39(6)：1087-1094.ZHANG Hongna, LIU Shenghui, SU Weisheng, et al. Ethylene induced sensitivity of difference pineapple varieties[J]. Chinese Journal of Tropical Crops, 2018, 39(6): 1087-1094. (in Chinese with English abstract) [3] 云南农业大学. 一种菠萝催花装置：201410483231. 2[P]. 2017-08-25. [4] SONG Y, SUN H, LI M, et al. Technology application of smart spray in agriculture: A review[J]. Intelligent Automation & Soft Computing, 2015, 21(3): 319-333. [5] 何雄奎. 中国植保机械与施药技术研究进展[J]. 农药学学报，2019，21(Z1)：921-930.HE Xiongkui. Research and development of crop protection machinery and chemical application technology in China[J]. Chinese Journal of Pesticide Science, 2019, 21(Z1): 921-930. (in Chinese with English abstract) [6] MESHRAM A T, VANALKAR A V, KALAMBE K B, et al. Pesticide spraying robot for precision agriculture: A categorical literature review and future trends[J]. Journal of Field Robotics, 2022, 39(2): 153-171. [7] SHARMA S, BORSE R. Automatic agriculture spraying robot with smart decision making[C]//Intelligent Systems Technologies and Applications 2016. Springer International Publishing, India, 2016: 743-758. [8] 闫晓静，杨代斌，薛新宇，等. 中国农药应用工艺学20年的理论研究与技术概述[J]. 农药学学报，2019，21(Z1)：908-920.YAN Xiaojing, YANG Daibin, XUE Xinyu, et al. Overview in theories and technologies for pesticide application in China during the last two decades[J]. Chinese Journal of Pesticide Science, 2019, 21(Z1): 908-920. (in Chinese with English abstract) [9] 何雄奎. 高效植保机械与精准施药技术进展[J]. 植物保护学报，2022，49(1)：389-397.HE Xiongkui. Research and development of efficient plant protection equipment and precision spraying technology in China: A review[J]. Journal of Plant Protection, 2022, 49(1): 389-397. (in Chinese with English abstract) [10] 吕沐华，丁珠玉. 基于机器视觉的果园喷药除草机器人视觉系统设计[J]. 中国农机化学报，2021，42(5)：42-48.LV Muhua, DING Zhuyu. Design of visual system for orchard spraying weeding robot based on machine vision[J]. Journal of Chinese Agricultural Mechanization, 2021, 42(5): 42-48. (in Chinese with English abstract) [11] UTSTUMO T, URDAL F, BREVIK A, et al. Robotic in-row weed control in vegetables[J]. Computers and electronics in agriculture, 2018, 154: 36-45. [12] ZONG Z, LIU G, ZHAO S. Real-time localization approach for maize cores at seedling stage based on machine vision[J]. Agronomy, 2020, 10(4): 470. [13] LIU J, WANG X. Plant diseases and pests detection based on deep learning: a review[J]. Plant Methods, 2021, 17: 1-18. [14] 赵民，葛云露，丁宁. 基于CenterNet算法的复杂环境目标检测技术[J]. 中国电子科学研究院学报，2021，16(7)：654-660.ZHAO Min, GE Yunlu, DING Ning. Object detection technology in complex environment based on centerNet algorithm[J]. Journal of Chinese Academy of Electronic Sciences, 2021, 16(7): 654-660. (in Chinese with English abstract) [15] GIRSHICK R, DONAHUE J, DARRELL T, et al. Rich feature hierarchies for accurate object detection and semantic segmentation[C]// Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. IEEE Computer Society: Piscataway, 2014: 580-587. [16] REN S Q, HE K M, 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. [17] REDMON J, DIVVALA S, GIRSHICK R, et al. You only look once: Unified, real-time object detection[C] //Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. Las Vegas, USA: IEEE, 2016: 779-788 [18] REDMON J, FARHADI A. YOLO9000: better, faster, stronger[C]//Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. IEEE Computer Society: Piscataway, 2017: 7263-7271. [19] BOCHKOVSKIY A, WANG C Y, LIAO H Y M. YOLOv4: optimal speed and accuracy of object detection[C]// IEEE Conference on Computer Vision and Pattern Recognition, Seattle, USA: IEEE, 2020: 1-17. [20] LIU W, ANGUELOV D, ERHAN D, et al. Ssd: Single shot multibox detector[C]//European conference on computer vision. German: Springer, 2016: 21-37. [21] FERREIRA A, FREITAS D, SILVA G, et al. Weed detection in soybean crops using ConvNets[J]. Computers and Electronics in Agriculture, 2017, 143: 314-324. [22] JIANG H, ZHANG C, QIAO Y, et al. CNN feature based graph convolutional network for weed and crop recognition in smart farming[J]. Computers and Electronics in Agriculture, 2020, 174: 105450. [23] 彭明霞，夏俊芳，彭辉. 融合FPN的Faster R-CNN复杂背景下棉田杂草高效识别方法[J]. 农业工程学报，2019，35(20)：202-209. PENG Mingxia, XIA Junfang, PENG Hui. Efficient recognition of cotton and weed in field based on Faster R-CNN by integrating FPN[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2019, 35(20): 202-209. (in Chinese with English abstract) [24] LE V N T, AHDEROM S, ALAMEH K. Performances of the lbp based algorithm over cnn models for detecting crops and weeds with similar morphologies[J]. Sensors, 2020, 20(8): 2193. [25] 孙俊，何小飞，谭文军，等. 空洞卷积结合全局池化的卷积神经网络识别作物幼苗与杂草[J]. 农业工程学报，2018，34(11)：159-165. SUN Jun, HE Xiaofei, TAN Wenjun, et al. Recognition of crop seedling and weed recognition based on dilated convolution and global pooling in CNN[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2018, 34(11): 159-165. (in Chinese with English abstract) [26] CHOLLET F. Xception: Deep learning with depthwise separable convolutions[C]//Proceedings of the IEEE conference on computer vision and pattern recognition. 2017: 1251-1258. [27] 张兆国，张振东，李加念，等. 采用改进YoloV4模型检测复杂环境下马铃薯[J]. 农业工程学报，2021，37(22)：170-178. ZHANG ZhaoGuo, ZHANG Zhendong, LI Jianian, et al. Potato detection in complex environment based on improved YoloV4 model[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2021, 37(22): 170-178. (in Chinese with English abstract) [28] 王昱潭，薛君蕊. 改进SSD的灵武长枣图像轻量化目标检测方法[J]. 农业工程学报，2021，37(19)：173-182. WANG Yutan, XUE Junrui. Lightweight object detection method for Lingwu long jujube images based on improved SSD[J]. Transactions of the Chinese Society of Agricultural Engineering (Transactions of the CSAE), 2021, 37(19): 173-182. (in Chinese with English abstract) [29] HE K, ZHANG X, REN S, et al. Spatial pyramid pooling in deep convolutional networks for visual recognition[J]. IEEE transactions on pattern analysis and machine intelligence, 2015, 37(9): 1904-1916. [30] 孔维刚,李文婧,王秋艳，等.基于改进YOLOv4算法的轻量化网络设计与实现[J].计算机工程,2022,48(03):181-188. KONG Weigang, LI Wenjing, WANG Qiuyan, et al. Design and implementation of lightweight network based on YOLOv4 algorithm[J]. Computer Engineering, 2022, 48(3): 181-188. (in Chinese with English abstract) [31] HOWARD A G, ZHU M, CHEN B, et al. Mobilenets: Efficient convolutional neural networks for mobile vision applications[J]. arXiv preprint arXiv: 1704. 04861, 2017. [32] HAN K, WANG Y, TIAN Q, et al. Ghostnet: More features from cheap operations[C]//Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. Seattle, USA: IEEE, 2020: 1580-1589. [33] WOO S, PARK J, LEE J Y, et al. Cbam: Convolutional block attention module[C]//Proceedings of the European Conference on Computer Vision (ECCV). German, 2018: 3-19. [34] LIN T Y, GOYAL P, GIRSHICK R, et al. Focal loss for dense object detection[C]//Proceedings of the IEEE International Conference on Computer Vision. IEEE Computer Society: Piscataway, 2017: 2980-2988.