基于无人机图像和改进LSC-CNN模型的密集苗木检测和计数方法  

Dense Nursery Stock Detecting and Counting Based on UAV Aerial Images and Improved LSC-CNN

作　　者：彭小丹 陈锋军[1,2,3,4] 朱学岩 才嘉伟 顾梦梦 PENG Xiaodan;CHEN Fengjun;ZHU Xueyan;CAI Jiawei;GU Mengmeng(School of Technology,Beijing Forestry University,Beijing 100083,China;National Key Laboratory of Efficient Production of Forest Resources,Beijing 100083,China;Beijing Laboratory of Urban and Rural Ecological Environment,Beijing Municipal Education Commission,Beijing 100083,China;Key Laboratory of State Forestry Administration for Forestry Equipment and Automation,Beijing 100083,China;Architecture of Horticultural and Landscape,Colorado State University,Fort Collins CO 80523,USA)

机构地区：[1]北京林业大学工学院,北京100083 [2]林木资源高效生产全国重点实验室,北京100083 [3]城乡生态环境北京实验室,北京100083 [4]林业装备与自动化国家林业和草原局重点实验室,北京100083 [5]科罗拉多州立大学园艺与景观建筑系,美国柯林斯堡80523

出　　处：《智慧农业（中英文）》2024年第5期88-97,共10页Smart Agriculture

基　　金：国家重点研发计划项目(2019YFD1002401);北京林业大学科技创新计划项目(2021ZY74);北京市共建项目专项。

摘　　要：[目的/意义]快速、准确地统计密集种植的苗木数量对苗木经营管理具有重要意义。为解决无人机航拍的密集种植苗木图像中苗木粘连、尺度差异大的问题,提出以点标签数据为监督信号的改进密集检测计数模型(Locate,Size and Count,LSC-CNN),同时实现苗木的检测和计数。[方法]改进的LSC-CNN模型通过将LSC-CNN模型特征提取网络的最后一层卷积替换为扩张卷积(Dilated Convolutions,DConv),实现在保留苗木细节特征的同时扩大感受野,帮助模型更好地理解上下文信息以区分粘连苗木。此外,在多个尺度分支前引入注意力机制(Convolutional Block Attention Module,CBAM)使模型聚焦于有助于苗木检测和计数的关键特征,以更好地适应不同尺度的苗木。为解决类别不平衡问题,提高模型的泛化能力,将损失函数替换为标签平滑交叉熵损失函数。[结果和讨论]经测试,改进LSC-CNN模型在456幅苗木图像的测试集上的平均绝对误差(Mean Absolute Error,MAE)、均方根误差(Root Mean Square Error,RMSE)和平均计数准确率(Mean Counting Accurate,MCA)分别为14.24株、22.22株和91.23%,三项指标均优于IntegrateNet、PSGCNet、CANet、CSRNet、CLTR和LSC-CNN模型。[结论]改进LSC-CNN模型能够准确实现密集种植苗木的检测和计数,适用于多种树木的检测和计数工作。[Objective]The number,location,and crown spread of nursery stock are important foundations data for their scientific management.Traditional approach of conducting nursery stock inventories through on-site individual plant surveys is labor-intensive and time-con suming.Low-cost and convenient unmanned aerial vehicles(UAVs)for on-site collection of nursery stock data are beginning to be utilized,and the statistical analysis of nursery stock information through technical means such as image processing achieved.During the data collection process,as the flight altitude of the UAV increases,the number of trees in a single image also increases.Although the anchor box can cover more information about the trees,the cost of annotation is enormous in the case of a large number of densely populated tree images.To tackle the challenges of tree adhesion and scale variance in images captured by UAVs over nursery stock,and to reduce the annotation costs,using point-labeled data as supervisory signals,an improved dense detection and counting model was proposed to accurately obtain the location,size,and quantity of the targets.[Method]To enhance the diversity of nursery stock samples,the spruce dataset,the Yosemite,and the KCL-London publicly available tree datasets were selected to construct a dense nursery stock dataset.A total of 1520 nursery stock images were acquired and divided into training and testing sets at a ratio of 7:3.To enhance the model's adaptability to tree data of different scales and variations in lighting,data augmentation methods such as adjusting the contrast and resizing the images were applied to the images in the training set.After enhancement,the training set consists of 3192 images,and the testing set contains 456 images.Considering the large number of trees contained in each image,to reduce the cost of annotation,the method of selecting the center point of the trees was used for labeling.The LSC-CNN model was selected as the base model.This model can detect the quantity,location,and size of trees throu

关 键 词：无人机 密集种植 计数 多尺度 LSC-CNN 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]