基于多光谱图像的土壤有机质含量检测系统与APP研究  

Detection System and APP Development of Soil Organic Matter ContentBased on Multispectral Images

作　　者：杨玮[1] 于滈 李浩[1] 曹永研 郝子源 YANG Wei;YU Hao;LI Hao;CAO Yongyan;HAO Ziyuan(Key Laboratory of Smart Agriculture Systems,Ministry of Education,China Agricultural University,Beijing 100083,China)

机构地区：[1]中国农业大学智慧农业系统集成研究教育部重点实验室,北京100083

出　　处：《农业机械学报》2023年第9期270-278,共9页Transactions of the Chinese Society for Agricultural Machinery

基　　金：浙江省科技计划项目(2021C02023)。

摘　　要：受到土壤种类、水分等客观因素的干扰,基于图像预测土壤有机质(Soil organic matter, SOM)含量与传统方法在检测精度上仍存在差距,限制了相关技术的推广和普及。为提升基于图像预测SOM含量的精度,本研究提出N_DenseNet网络模型,在DenseNet169基础上加入多尺度池化模块,通过获取更多的维度特征提升模型的性能,并结合Android端开发SOM实时检测应用程序(APP),通过内网透射实现PC端与手机端数据的及时传输。以黑龙江省友谊县、北京市昌平区、山东省泰安市3地的350份土样为基础,通过手机以及多光谱无人机获取原位土壤的高清图像,R波段、红边波段与近红外波段图像,以丰富数据信息,并通过室内胁迫的方式拍摄土壤样品在不同水分梯度下的图像缓解水分对图像造成的影响。对比不同深度学习模型,基于多光谱图像数据训练的N_DenseNet表现最好,整体表现优于DenseNet169,测试集R~2为0.833,RMSE为3.943 g/kg,R~2相比于可见光数据提升0.016,证明了训练过程加入R波段与红边和近红外波段图像后有助于提升模型的性能,证明了该方法的可行性。手机端APP与后台端数据相连实现数据实时传输,实现了田间土样SOM含量的实时预测,经田间试验验证,模型预测集R~2为0.805,检测时间为2.8 s,满足了田间SOM含量检测的需求,为SOM含量实时检测提供了新的思路。Predicting soil organic matter(SOM)content based on images has the advantages of convenience and low cost.Interfered by objective factors such as soil type and moisture,there is still a gap between the detection accuracy of image prediction SOM content and traditional methods,which limits the promotion and popularization of related technologies.In order to improve the accuracy of image prediction of SOM content,a N_DenseNet multi-scale pooling module was added to DenseNet169 to improve the performance of the model by obtaining more dimensional features,and combined the development of SOM real-time detection APP on the Android side to realize the timely transmission of server and mobile phone data through intranet projection.Based on 350 soil samples from Youyi County,Heilongjiang Province,Changping District,Beijing City and Tai'an City,Shandong Province,highdefinition images,R-band,red-edged band and near-infrared band images of in situ soil were obtained through mobile phones and multispectral drones to enrich data information,and image samples of soil samples under different moisture gradients were taken through indoor stress to alleviate the impact of moisture on the image.Compared with different deep learning models,the N_DenseNet trained based on multispectral image data performed the best,the overall performance was better than that of DenseNet169,the test set R2 was 0.833,RMSE was 3.943 g/kg,and R2 was improved by 0.011 compared with the visible light data,which proved that the addition of R-band and red-edged and nearinfrared images to the training process helped to improve the performance of the model,which proved the feasibility of the method.The mobile phone APP was connected to the background data to realize real-time data transmission,and realized the real-time detection of SOM content of soil samples in the field,and the model predicted R2 as 0.805 and the detection time was 2.8 s after field verification,which met the needs of SOM content detection in the field and provided an idea for real-time de

关 键 词：土壤有机质 检测系统 多光谱图像 深度学习 ANDROID APP 

分 类 号：S237[农业科学—农业机械化工程]