Sun, Sashuang and Liang, Ning and Zuo, Zhiyu and Parsons, David and Morel, Julien and Shi, Jiang and Wang, Zhao and Luo, Letan and Zhao, Lin and Fang, Hui and He, Yong and Zhou, Zhenjiang
(2021).
Estimation of Botanical Composition in Mixed Clover-Grass Fields Using Machine Learning-Based Image Analysis.
Frontiers in Plant Science. 12
, 622429
[Research article]
![]() |
PDF
4MB |
Abstract
This study aims to provide an effective image analysis method for clover detection and botanical composition (BC) estimation in clover-grass mixture fields. Three transfer learning methods, namely, fine-tuned DeepLab V3+, SegNet, and fully convolutional network-8s (FCN-8s), were utilized to detect clover fractions (on an area basis). The detected clover fraction (CFdetected), together with auxiliary variables, viz., measured clover height (H-clover) and grass height (H-grass), were used to build multiple linear regression (MLR) and back propagation neural network (BPNN) models for BC estimation. A total of 347 clover-grass images were used to build the estimation model on clover fraction and BC. Of the 347 samples, 226 images were augmented to 904 images for training, 25 were selected for validation, and the remaining 96 samples were used as an independent dataset for testing. Testing results showed that the intersection-over-union (IoU) values based on the DeepLab V3+, SegNet, and FCN-8s were 0.73, 0.57, and 0.60, respectively. The root mean square error (RMSE) values for the three transfer learning methods were 8.5, 10.6, and 10.0%. Subsequently, models based on BPNN and MLR were built to estimate BC, by using either CFdetected only or CFdetected, grass height, and clover height all together. Results showed that BPNN was generally superior to MLR in terms of estimating BC. The BPNN model only using CFdetected had a RMSE of 8.7%. In contrast, the BPNN model using all three variables (CFdetected, H-clover, and H-grass) as inputs had an RMSE of 6.6%, implying that DeepLab V3+ together with BPNN can provide good estimation of BC and can offer a promising method for improving forage management.
Authors/Creators: | Sun, Sashuang and Liang, Ning and Zuo, Zhiyu and Parsons, David and Morel, Julien and Shi, Jiang and Wang, Zhao and Luo, Letan and Zhao, Lin and Fang, Hui and He, Yong and Zhou, Zhenjiang | ||||||
---|---|---|---|---|---|---|---|
Title: | Estimation of Botanical Composition in Mixed Clover-Grass Fields Using Machine Learning-Based Image Analysis | ||||||
Series Name/Journal: | Frontiers in Plant Science | ||||||
Year of publishing : | 2021 | ||||||
Volume: | 12 | ||||||
Article number: | 622429 | ||||||
Number of Pages: | 12 | ||||||
Publisher: | FRONTIERS MEDIA SA | ||||||
ISSN: | 1664-462X | ||||||
Language: | English | ||||||
Publication Type: | Research article | ||||||
Article category: | Scientific peer reviewed | ||||||
Version: | Published version | ||||||
Copyright: | Creative Commons: Attribution 4.0 | ||||||
Full Text Status: | Public | ||||||
Subjects: | (A) Swedish standard research categories 2011 > 1 Natural sciences > 106 Biological Sciences (Medical to be 3 and Agricultural to be 4) > Botany | ||||||
Keywords: | crop species classification, forage crop, transfer learning, DeepLab V3+, back propagation neural network | ||||||
URN:NBN: | urn:nbn:se:slu:epsilon-p-111124 | ||||||
Permanent URL: | http://urn.kb.se/resolve?urn=urn:nbn:se:slu:epsilon-p-111124 | ||||||
Additional ID: |
| ||||||
ID Code: | 22811 | ||||||
Faculty: | NJ - Fakulteten för naturresurser och jordbruksvetenskap | ||||||
Department: | (NL, NJ) > Dept. of Agricultural Research for Northern Sweden (VH) > Dept. of Agricultural Research for Northern Sweden | ||||||
Deposited By: | SLUpub Connector | ||||||
Deposited On: | 22 Mar 2021 07:43 | ||||||
Metadata Last Modified: | 22 Mar 2021 07:51 |
Repository Staff Only: item control page