摘要
为实现自动导航农机的避障,解决搭载在农机顶部的全景相机获取其周围360°的图像信息并精确实时快速检测出障碍物的问题,提出了一种改进YOLO v3-tiny目标检测模型,实现了田间行人和其他农机的检测与识别。为了提高全景图像中小目标的检测效果,以检测速度快、轻量级的网络模型YOLO v3-tiny为基础框架,通过融合浅层特征与第二YOLO预测层之前的拼接层作为第三预测层,增加小目标的检测效果;为了进一步增加网络模型对目标特征的提取能力,借鉴残差网络的思想,在YOLO v3-tiny主干网络上引入残差模块,增加网络深度和学习能力,从而能够较好地提高网络的检测能力。为了验证模型的性能,建立了农田环境下1100幅行人与农机两类障碍物图像原始数据集,经数据扩增后得到2200幅图像数据集,按8∶1∶1将数据集划分为训练集、验证集和测试集,在Pytorch 1.8深度学习框架下进行模型训练,模型训练完后用220幅测试集图像对不同模型进行测试。试验结果表明,基于改进YOLO v3-tiny的农田障碍物检测模型,平均准确率和召回率分别为95.5%和93.7%,相比于原网络模型,分别提高了5.6、5.2个百分点;单幅全景图像检测耗时为6.3 ms,视频流检测平均帧率为84.2 f/s,模型内存为64 MB。改进后的模型,在保证检测精度较高的同时,能够满足农机在运动状态下实时障碍物检测需求。
In order to realize the obstacle avoidance of automatic navigation agricultural machinery and solve the problem that the panoramic camera mounted on the top of the agricultural machinery needs to accurately and quickly detect obstacles in real time to obtain the 360°image information around it,an improved YOLO v3-tiny target detection model was proposed,which can realize the detection and identification of pedestrians and other agricultural machinery in the field.In order to improve the detection effect of small targets in panoramic images,the fast detection speed and lightweight network model YOLO v3-tiny was used as the basic framework,and the splicing layer before the second YOLO prediction layer was used as the third prediction layer by fusing the shallow features with the second YOLO prediction layer to increase the detection effect of small targets;in order to further increase the network model’s ability to extract target features,borrowing the idea of residual network,the residual module was introduced on the YOLO v3-tiny backbone network to increase the depth and learning ability of the network,so that it can better improve the detection capabilities of the network.In order to verify the performance of the model,totally 1100 original data sets of pedestrian and agricultural machinery obstacles in the farmland environment were established,after data amplification,totally 2200 images data sets were obtained,the data sets were divided into training set,verification and test set according to 8∶1∶1,and the model was trained under the Pytorch 1.8 deep learning framework.After the model was trained,totally 220 images of test set were used to test different models.The test results showed that the farmland obstacle detection model based on improved YOLO v3-tiny had an average accuracy rate and recall rate of 95.5%and 93.7%,respectively,which were 5.6 percentage points and 5.2 percentage points higher than that of the original network model.Single panoramic image detection took 6.3 ms,the average frame rate of video stream detection was 84.2 f/s,and the model memory was 64 MB.The improved model can meet the real-time obstacle detection requirements of agricultural machinery in motion while ensuring high detection accuracy.
作者
陈斌
张漫
徐弘祯
李寒
尹彦鑫
CHEN Bin;ZHANG Man;XU Hongzhen;LI Han;YIN Yanxin(Key Laboratory of Modern Precision Agriculture System Integration Research,Ministry of Education,China Agricultural University,Beijing 100083,China;Key Laboratory of Agricultural Information Acquisition Technology,Ministry of Agriculture and Rural Affairs,China Agricultural University,Beijing 100083,China;National Research Center of Intelligent Equipment for Agriculture,Beijing 100097,China)
出处
《农业机械学报》
EI
CAS
CSCD
北大核心
2021年第S01期58-65,共8页
Transactions of the Chinese Society for Agricultural Machinery
基金
国家重点研发计划项目(2019YFB1312300-2019YFB1312305)
中国农业大学建设世界一流大学(学科)和特色发展引领专项资金项目(2021AC006)
作者简介
陈斌(1996—),男,硕士生,主要从事机器视觉在农机自动导航中的应用研究,E-mail:SY20203081544@cau.edu.cn;通讯作者:张漫(1975—),女,教授,博士生导师,主要从事农业电气化与自动化研究,E-mail:cauzm@cau.edu.cn
