Individual Tree Species Identification Based on a Combination of Deep Learning and Traditional Features
Accurate identification of individual tree species (ITS) is crucial to forest management. However, current ITS identification methods are mainly based on traditional image features or deep learning. Traditional image features are more interpretative, but the generalization and robustness of such met...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2023-04-01
|
Series: | Remote Sensing |
Subjects: | |
Online Access: | https://www.mdpi.com/2072-4292/15/9/2301 |
_version_ | 1797601828128423936 |
---|---|
author | Caiyan Chen Linhai Jing Hui Li Yunwei Tang Fulong Chen |
author_facet | Caiyan Chen Linhai Jing Hui Li Yunwei Tang Fulong Chen |
author_sort | Caiyan Chen |
collection | DOAJ |
description | Accurate identification of individual tree species (ITS) is crucial to forest management. However, current ITS identification methods are mainly based on traditional image features or deep learning. Traditional image features are more interpretative, but the generalization and robustness of such methods are inferior. In contrast, deep learning based approaches are more generalizable, but the extracted features are not interpreted; moreover, the methods can hardly be applied to limited sample sets. In this study, to further improve ITS identification, typical spectral and texture image features were weighted to assist deep learning models for ITS identification. To validate the hybrid models, two experiments were conducted; one on the dense forests of the Huangshan Mountains, Anhui Province and one on the Gaofeng forest farm, Guangxi Province, China. The experimental results demonstrated that with the addition of image features, different deep learning ITS identification models, such as DenseNet, AlexNet, U-Net, and LeNet, with different limited sample sizes (480, 420, 360), were all enhanced in both study areas. For example, the accuracy of DenseNet model with a sample size of 480 were improved to 87.67% from 85.41% in Huangshan. This hybrid model can effectively improve ITS identification accuracy, especially for UAV aerial imagery or limited sample sets, providing the possibility to classify ITS accurately in sample-poor areas. |
first_indexed | 2024-03-11T04:08:11Z |
format | Article |
id | doaj.art-a0cad3f51e95492ea72e80fcc472c71d |
institution | Directory Open Access Journal |
issn | 2072-4292 |
language | English |
last_indexed | 2024-03-11T04:08:11Z |
publishDate | 2023-04-01 |
publisher | MDPI AG |
record_format | Article |
series | Remote Sensing |
spelling | doaj.art-a0cad3f51e95492ea72e80fcc472c71d2023-11-17T23:38:16ZengMDPI AGRemote Sensing2072-42922023-04-01159230110.3390/rs15092301Individual Tree Species Identification Based on a Combination of Deep Learning and Traditional FeaturesCaiyan Chen0Linhai Jing1Hui Li2Yunwei Tang3Fulong Chen4Key Laboratory of Digital Earth Science, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaInternational Research Center of Big Data for Sustainable Development Goals, Beijing 100094, ChinaKey Laboratory of Digital Earth Science, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaInternational Research Center of Big Data for Sustainable Development Goals, Beijing 100094, ChinaKey Laboratory of Digital Earth Science, Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, ChinaAccurate identification of individual tree species (ITS) is crucial to forest management. However, current ITS identification methods are mainly based on traditional image features or deep learning. Traditional image features are more interpretative, but the generalization and robustness of such methods are inferior. In contrast, deep learning based approaches are more generalizable, but the extracted features are not interpreted; moreover, the methods can hardly be applied to limited sample sets. In this study, to further improve ITS identification, typical spectral and texture image features were weighted to assist deep learning models for ITS identification. To validate the hybrid models, two experiments were conducted; one on the dense forests of the Huangshan Mountains, Anhui Province and one on the Gaofeng forest farm, Guangxi Province, China. The experimental results demonstrated that with the addition of image features, different deep learning ITS identification models, such as DenseNet, AlexNet, U-Net, and LeNet, with different limited sample sizes (480, 420, 360), were all enhanced in both study areas. For example, the accuracy of DenseNet model with a sample size of 480 were improved to 87.67% from 85.41% in Huangshan. This hybrid model can effectively improve ITS identification accuracy, especially for UAV aerial imagery or limited sample sets, providing the possibility to classify ITS accurately in sample-poor areas.https://www.mdpi.com/2072-4292/15/9/2301remote sensingindividual tree species identificationdeep learningspectral featuretexture feature |
spellingShingle | Caiyan Chen Linhai Jing Hui Li Yunwei Tang Fulong Chen Individual Tree Species Identification Based on a Combination of Deep Learning and Traditional Features Remote Sensing remote sensing individual tree species identification deep learning spectral feature texture feature |
title | Individual Tree Species Identification Based on a Combination of Deep Learning and Traditional Features |
title_full | Individual Tree Species Identification Based on a Combination of Deep Learning and Traditional Features |
title_fullStr | Individual Tree Species Identification Based on a Combination of Deep Learning and Traditional Features |
title_full_unstemmed | Individual Tree Species Identification Based on a Combination of Deep Learning and Traditional Features |
title_short | Individual Tree Species Identification Based on a Combination of Deep Learning and Traditional Features |
title_sort | individual tree species identification based on a combination of deep learning and traditional features |
topic | remote sensing individual tree species identification deep learning spectral feature texture feature |
url | https://www.mdpi.com/2072-4292/15/9/2301 |
work_keys_str_mv | AT caiyanchen individualtreespeciesidentificationbasedonacombinationofdeeplearningandtraditionalfeatures AT linhaijing individualtreespeciesidentificationbasedonacombinationofdeeplearningandtraditionalfeatures AT huili individualtreespeciesidentificationbasedonacombinationofdeeplearningandtraditionalfeatures AT yunweitang individualtreespeciesidentificationbasedonacombinationofdeeplearningandtraditionalfeatures AT fulongchen individualtreespeciesidentificationbasedonacombinationofdeeplearningandtraditionalfeatures |