Forest Species Classification Research Articles

A novel framework combining band selection algorithm and improved 3D prototypical network for tree species classification using airborne hyperspectral images

Fine-grained identification of forest types and tree species represents a critical aspect of forest resource inventory and monitoring. The use of airborne hyperspectral remote sensing imagery stands out for its ability to finely differentiate among tree species, leveraging its exceptional spatial resolution and rich spectral details. However, this approach is limited by several challenges (e.g., high spectral correlation and information redundancy). In accordance, the adoption of a lightweight deep learning approach in the form of a few-shot learning model can effectively resolve the challenges of multi-forest tree species classification. Therefore, integrating a data dimensionality reduction algorithm with a few-shot classification model presents a promising avenue for resolving the fine-grained classification of forest tree species. In this study, we propose the innovative classification framework FAST 3D-CNN P-Net. This framework utilizes CNN for band selection, enhances the fine-grained identification process in hyperspectral data, and integrates an optimized FAST 3D-CNN into the P-Net classifier (a few-shot classifier). First, a CNN-based band selection method is employed to learn the nonlinear dependencies between spectral bands, assign weights to rank the bands, and reconstruct the global spectral information using the most informative bands. It then constructs a novel classification model, designated FAST 3D-CNN P-Net, through the integration of an optimal 3D-CNN with a prototypical network. To enhance classification performance, the FAST 3D-CNN P-Net utilizes reconstructed hyperspectral images derived from the band selection results as input. The effectiveness of the proposed framework was assessed with the airborne GFF dataset and the widely accessible medium-resolution hyperspectral datasets, Indian Pines (IP) and Kennedy Space Center (KSC). The overall classification accuracy reached 98.33 % for the GFF dataset and 97.21 % and 99.43 % for the IP and KSC, respectively, exhibiting performance superiority compared to the standalone 3D-CNN classification network. This classification framework demonstrates efficiency in selecting a subset of hyperspectral bands with minimal redundancy, empowering the rapid and accurate classification and mapping of tree species in complicated, multi-species forest stands, even with a limited quantity of labeled samples.

Mapping Forest Tree Species Using Sentinel-2 Time Series by Taking into Account Tree Age

Accurate classification of forest tree species holds great significance in the context of forest biodiversity assessment and the management of forest resources. In this study, we utilized Sentinel-2 time series data with high temporal and spatial resolution for tree species classification. To address potential classification errors stemming from spectral differences due to tree age variations, we implemented the Continuous Change Detection and Classification (CCDC) algorithm to estimate tree ages, which were integrated as additional features into our classification models. Four different combinations of classification features were created for both the random forest (RF) algorithm and extreme gradient boosting (XGB) algorithm: spectral band (Spec), spectral band combined with tree age feature (SpecAge), spectral band combined with spectral index (SpecVI), and spectral band combined with spectral index and tree age feature (SpecVIAge). The results demonstrated that the XGB-based models outperformed the RF-based ones, with the SpecVIAge model achieving the highest accuracy at 78.8%. The incorporation of tree age as a classification feature led to an improvement in accuracy by 2% to 3%. The improvement effect on classification accuracy varies across tree species, due to the varying uniformity of tree age among different tree species. These results also showed it is feasible to accurately map regional tree species based on a time-series multi-feature tree species classification model which takes into account tree age.

HPAC: a forest tree species recognition network based on multi- scale spatial enhancement in remote sensing images

ABSTRACT Forest tree species recognition is a pivotal subject in the field of remote sensing. To address this, deep learning has been extensively applied. Thus far, most classification methods have generally relied on learning certain global features, yet often overlook the characteristics of specific regions, thereby struggling to adequately handle the similarity between classes. Furthermore, due to the singular nature of features, effectively representing the attributes of tree species images becomes challenging, consequently impacting classification performance. To tackle these issues, a novel approach for forest tree species classification in remote sensing images is proposed, based on the Hollow Pyramid Attention Combination (HPAC) network. Initially, a Shallow Multi-scale Hollow Fusion (SMHF) module is introduced before the 7 × 7 convolution in the ResNet-50 network and the first residual block’s first layer. This module employs dilated convolutions to achieve varying receptive fields. Moreover, it incorporates positional feature information, significantly enhancing the shallow-level feature extraction capabilities, resulting in a richer feature representation. Subsequently, to minimize network parameters and computational workload while bolstering the capacity to recognize deep-level features, the last residual block of the ResNet-50 differentiation is substituted with a Maxpool Avgpool Fusion (MAF) module. This replacement serves to enhance classification accuracy. The classification process is ultimately concluded with a Softmax classifier. Experimental results underscore the effectiveness of the proposed method, achieving a classification accuracy of 95.89% on the PCANDVI dataset of forest tree species data (FTSD). In summary, the introduced HPAC network proves to be both feasible and effective.

A Multi-Scale Convolution and Multi-Layer Fusion Network for Remote Sensing Forest Tree Species Recognition

Forest tree species identification in the field of remote sensing has become an important research topic. Currently, few research methods combine global and local features, making it challenging to accurately handle the similarity between different categories. Moreover, using a single deep layer for feature extraction overlooks the unique feature information at intermediate levels. This paper proposes a remote sensing image forest tree species classification method based on the Multi-Scale Convolution and Multi-Level Fusion Network (MCMFN) architecture. In the MCMFN network, the Shallow Multi-Scale Convolution Attention Combination (SMCAC) module replaces the original 7 × 7 convolution at the first layer of ResNet-50. This module uses multi-scale convolution to capture different receptive fields, and combines it with the attention mechanism to effectively enhance the ability of shallow features and obtain richer feature information. Additionally, to make efficient use of intermediate and deep-level feature information, the Multi-layer Selection Feature Fusion (MSFF) module is employed to improve classification accuracy. Experimental results on the Aerial forest dataset demonstrate a classification accuracy of 91.03%. The comprehensive experiments indicate the feasibility and effectiveness of the proposed MCMFN network.

Discriminative feature constraints via supervised contrastive learning for few-shot forest tree species classification using airborne hyperspectral images

In scenarios where sample collection is limited, studying few-shot learning algorithms such as prototypical networks (P-Net) is a keynote topic for supervised multiple tree species classification. In a previous study, we improved the P-Net by combining the feature enhancement algorithm based on the convolutional block attention module and several popular data augmentation methods in the computer vision domain, the classification accuracy can be significantly increased, and the degree of model overfitting can be reduced. However, there is a clear boundary between the data augmentations and the feature enhancement algorithm, which is manifested in that data augmentations are only used to enrich the diversity of the learned samples, but cannot directly affect the construction of the objective function, thus limiting the ability of data augmentations. In fact, in the supervised contrastive learning research, data augmentation methods are often used to generate positive samples of an anchor image to construct the objective function, i.e. supervised contrastive loss. The core idea for solving such a boundary problem is to use contrastive learning to make the anchor image close to its positive samples and the negative samples away from each other. Inspired by this, we introduced supervised contrastive learning in the P-Net, i.e., SCL-P-Net, which takes the discriminative feature representations as the constraints of the prototype clustering algorithm. In SCL-P-Net, data augmentation methods can not only extend the sample distribution, but also be used to construct the supervised contrastive loss directly. The study involves four airborne hyperspectral image datasets related to tree species classification, including the GFF-A and GFF-B datasets collected from Gaofeng Forest Farm in Nanning City, Guangxi Province, South China, the Xiongan dataset from Matiwan Village in Xiongan New Area, Hebei Province, North China, and the Tea Farm dataset from Fanglu Tea Farm in Changzhou City, Jiangsu Province, East China. The highest overall accuracy (OA) for the four datasets is 99.23% for GFF-A, 98.39% for GFF-B, 99.30% for Xiongan, and 99.54% for Tea Farm. It is concluded that the proposed SCL-P-Net classification framework can achieve multiple tree species classification with high-precision. Without changing the basic classification framework of P-Net, the introduction of supervised contrastive learning makes the combination of the data augmentations and the feature enhancement algorithm and plays a positive role in improving the distinguishability between samples.

Tree Species Classification in UAV Remote Sensing Images Based on Super-Resolution Reconstruction and Deep Learning

We studied the use of self-attention mechanism networks (SAN) and convolutional neural networks (CNNs) for forest tree species classification using unmanned aerial vehicle (UAV) remote sensing imagery in Dongtai Forest Farm, Jiangsu Province, China. We trained and validated representative CNN models, such as ResNet and ConvNeXt, as well as the SAN model, which incorporates Transformer models such as Swin Transformer and Vision Transformer (ViT). Our goal was to compare and evaluate the performance and accuracy of these networks when used in parallel. Due to various factors, such as noise, motion blur, and atmospheric scattering, the quality of low-altitude aerial images may be compromised, resulting in indistinct tree crown edges and deficient texture. To address these issues, we adopted Real-ESRGAN technology for image super-resolution reconstruction. Our results showed that the image dataset after reconstruction improved classification accuracy for both the CNN and Transformer models. The final classification accuracies, validated by ResNet, ConvNeXt, ViT, and Swin Transformer, were 96.71%, 98.70%, 97.88%, and 98.59%, respectively, with corresponding improvements of 1.39%, 1.53%, 0.47%, and 1.18%. Our study highlights the potential benefits of Transformer and CNN for forest tree species classification and the importance of addressing the image quality degradation issues in low-altitude aerial images.

Benchmark for Building Segmentation on Up-Scaled Sentinel-2 Imagery

Currently, we can solve a wide range of tasks using computer vision algorithms, which reduce manual labor and enable rapid analysis of the environment. The remote sensing domain provides vast amounts of satellite data, but it also poses challenges associated with processing this data. Baseline solutions with intermediate results are available for various tasks, such as forest species classification, infrastructure recognition, and emergency situation analysis using satellite data. Despite these advances, two major issues with high-performing artificial intelligence algorithms remain in the current decade. The first issue relates to the availability of data. To train a robust algorithm, a reasonable amount of well-annotated training data is required. The second issue is the availability of satellite data, which is another concern. Even though there are a number of data providers, high-resolution and up-to-date imagery is extremely expensive. This paper aims to address these challenges by proposing an effective pipeline for building segmentation that utilizes freely available Sentinel-2 data with 10 m spatial resolution. The approach we use combines a super-resolution (SR) component with a semantic segmentation component. As a result, we simultaneously consider and analyze SR and building segmentation tasks to improve the quality of the infrastructure analysis through medium-resolution satellite data. Additionally, we collected and made available a unique dataset for the Russian Federation covering area of 1091.2 square kilometers. The dataset provides Sentinel-2 imagery adjusted to the spatial resolution of 2.5 m and is accompanied by semantic segmentation masks. The building footprints were created using OpenStreetMap data that was manually checked and verified. Several experiments were conducted for the SR task, using advanced image SR methods such as the diffusion-based SR3 model, RCAN, SRGAN, and MCGR. The MCGR network produced the best result, with a PSNR of 27.54 and SSIM of 0.79. The obtained SR images were then used to tackle the building segmentation task with different neural network models, including DeepLabV3 with different encoders, SWIN, and Twins transformers. The SWIN transformer achieved the best results, with an F1-score of 79.60.

Identification of Tree Species in Forest Communities at Different Altitudes Based on Multi-Source Aerial Remote Sensing Data

The accurate identification of forest tree species is important for forest resource management and investigation. Using single remote sensing data for tree species identification cannot quantify both vertical and horizontal structural characteristics of tree species, so the classification accuracy is limited. Therefore, this study explores the application value of combining airborne high-resolution multispectral imagery and LiDAR data to classify tree species in study areas of different altitudes. Three study areas with different altitudes in Muyu Town, Shennongjia Forest Area were selected. Based on the object-oriented method for image segmentation, multi-source remote sensing feature extraction was performed. The recursive feature elimination algorithm was used to filter out the feature variables that were optimal for classifying tree species in each altitude study area. Four machine learning algorithms, SVM, KNN, RF, and XGBoost, were combined to classify tree species at each altitude and evaluate the accuracy. The results show that the diversity of tree layers decreased with the altitude in the different study areas. The texture features and height features extracted from LiDAR data responded better to the forest community structure in the different study areas. Coniferous species showed better classification than broad-leaved species within the same study areas. The XGBoost classification algorithm showed the highest accuracy of 87.63% (kappa coefficient of 0.85), 88.24% (kappa coefficient of 0.86), and 84.03% (kappa coefficient of 0.81) for the three altitude study areas, respectively. The combination of multi-source remote sensing numbers with the feature filtering algorithm and the XGBoost algorithm enabled accurate forest tree species classification.

Use of Machine Learning Algorithms in the Classification of Forest Species

Optimization in the process of managing forest resources seeks alternatives that make data collection possible. One of them alternatives is spectroradiometry, which consists of measuring the spectral response, having as product the response of the target in relation to the incident radiation along the electromagnetic spectrum, and that, using machine learning, with pre-selected models, makes it possible to identify. Given the above, the study aimed to use machine learning algorithms to classify species by vegetation indices from reflectance data. The study was developed at the Federal University from Santa Maria, working with the species Ficus benjamina, Inga marginata, Handroanthus chrysotrichus, Psidium cattleianum, Salix humboldtiana, Corymbia citriodora and Myrcianthes pungens, and spectral readings of the leaves were taken using the FieldSpec®3 spectroradiometer connected to RTS-3ZC3 integrating sphere. The reflectance values with wavelength ranged in amplitude from 350 ƞm to 2,500 ƞm and spectral resolution of 1 ƞm. Vegetation indices were calculated using the software R Studio, being: NDVI, SAVI, RVI, GNDVI, NDWI, NDWI2, GEMI, DVI, TVI, RVI, MSAVI, WDVI. The algorithms used to develop machine learning were: Random Forest (RF), k-Nearest Neighbors (K-NN), Naive Bayes (NB) and Support Vector Machine (SVM). RF proves to be the most appropriate for data validation, with 85% global accuracy, followed by SVM, with 71%, K-NN with 64% and NB with 35%. The indices with the best performance to point the species were NDWI and SAVI.

Deep Learning in Forest Tree Species Classification Using Sentinel-2 on Google Earth Engine: A Case Study of Qingyuan County

Forest tree species information plays an important role in ecology and forest management, and deep learning has been used widely for remote sensing image classification in recent years. However, forest tree species classification using remote sensing images is still a difficult task. Since there is no benchmark dataset for forest tree species, a forest tree species dataset (FTSD) was built in this paper to fill the gap based on the Sentinel-2 images. The FTSD contained nine kinds of forest tree species in Qingyuan County with 8,815 images, each with a resolution of 64 × 64 pixels. The images were produced by combining forest management inventory data and Sentinel-2 images, which were acquired with less than 20% clouds from 1 April to 31 October, including the years 2017, 2018, 2019, 2020, and 2021. Then, the images were preprocessed and downloaded from Google Earth Engine (GEE). Four different band combinations were compared in the paper. Moreover, a Principal Component Analysis (PCA) and Normalized Difference Vegetation Index (NDVI) were also calculated using the GEE. Deep learning algorithms including DenseNet, EfficientNet, MobileNet, ResNet, and ShuffleNet were trained and validated in the FTSD. RGB images with red, green, and blue (PC1, PC2, and NDVI) obtained the highest validation accuracy in four band combinations. ResNet obtained the highest validation accuracy in all algorithms after 500 epochs were trained in the FTSD, which reached 84.91%. As a famous and widely used remote sensing classification satellite imagery dataset, NWPU RESISC-45 was also trained and validated in the paper. ResNet achieved a high validation accuracy of 87.90% after training 100 epochs in NWPU RESISC-45. The paper shows in forest tree species classification based on remote sensing images and deep learning that (1) PCA and NDVI can be combined to improve the accuracy of classification; (2) ResNet is more suitable than other deep learning algorithms including DenseNet, EfficientNet, MobileNet, and ShuffleNet in remote sensing classification; and (3) being too shallow or deep in ResNet does not perform better in the FTSD, that is, 50 layers are better than 34 and 101 layers.

Forest plantation species classification using Full-Pol-Time-Averaged SAR scattering powers

Forest plantation species identification is essential for biodiversity conservation, sustainable management, and carbon sequestration. Fully Polarimetric Synthetic Aperture Radar (Full-Pol-SAR) remote sensing data aids as an alternative to time-consuming field inventory and cloud-covered optical data for forest plantation species identification. This study tried to identify forest plantation species using Fully Polarimetric Time-series Averaged SAR (Full-Pol-Time-Averaged SAR) datasets. Due to the availability of time series quad polarization data of TerraSAR-X from December 20, 2014 to March 07, 2015 and near-real-time field with us for the Haldwani forest area, we have utilized these data sets for the proposed work. The temporal effect of phenological changes can be characterized using the Full-Pol-SAR 6-component Scattering power Decomposition (6SD) and is utilized to identify different forest species. Misclassification problem exists in forests with single-date 6SD images, an improved forest species classification approach based on time-averaged 6SD and a supervised Random Forest (RF) classifier is proposed. This study reveals that Teak, Eucalyptus, Poplar, mixed, and non-forest could be distinctly identified and classified in time-averaged 6SD images. The overall classification accuracy of the proposed approach (∼83.72%) was, higher than the single-date Full-Pol-SAR results (mean overall accuracy with standard deviation 68.949 ± 2.096%).

A Study on the Difference of LULC Classification Results Based on Landsat 8 and Landsat 9 Data

Landsat 9 enhances the radiation resolution of the operational land imager from the 12 bits of Landsat 8 to 14 bits. The higher radiation resolution improves the sensitivity of the sensor to detect many subtler differences, especially in the case of dense forests or water. However, it remains unclear whether the difference in radiation resolution between Landsat 8 and Landsat 9 actually affects the classification results of water and tree species. Accordingly, the spectral reflectance and vegetation indices were extracted in this study, based on Landsat 8 and Landsat 9 images. Then, the classification models of land use and land cover (LULC) and tree species were developed by using a gradient tree boosting algorithm. Subsequently, the results were analyzed to further investigate how the differences in radiation resolution affect the classification results of LULC and tree species. It is shown that the LULC classification results of Landsat 8 and Landsat 9 are relatively favorable in most cases. However, the LULC classification results are relatively poor in test areas with a lower classification accuracy of water. Further analysis, in the case of test areas with poor classification results, indicates that there are significant differences in the water classification results between the two datasets. In other words, Landsat 9 produces better water classification results than Landsat 8 in most test areas. However, a temperature close to zero may lead to inverse water classification results. In addition, it indicates that the difference in forest classification results between the two datasets is small, but the results of forest tree species classification based on Landsat 9 are superior to those based on Landsat 8, with an improvement in overall accuracy of 6.01%. The results demonstrate that the difference in radiation resolution between Landsat 8 and Landsat 9 has little impact on the results of LULC classification in most cases. Nevertheless, in the case of some test areas, Landsat 9 is better suited for enhancing the classification accuracy of water and tree species.

Forest Tree Species Classification Based on Sentinel-2 Images and Auxiliary Data

Most research on forest tree species classification based on optical image data uses information such as spectral reflectance, vegetation index, texture, and phenology data. However, owing to the limited spectral resolution of multispectral images and the high cost of hyperspectral data, there is room for improvement in the classification of tree species in large areas based on optical images. The combined application of multispectral images and other auxiliary data can provide a new method for improving tree species classification accuracy. Hence, Sentinel-2 images were used to extract spectral reflectance, spectral index, texture, and phenological information. Data for topography, precipitation, air temperature, ultraviolet aerosol index, NO2 concentration, and other variables were included as auxiliary data. Models for forest tree species classification were constructed through feature combination and feature optimization using the random forest (RF), gradient tree boost (GTB), support vector machine (SVM), and classification and regression tree (CART) algorithms. The classification results of 16 feature combinations with the 4 classification methods were compared, and the contributions of different features to the classification models of forest tree species were evaluated. Finally, the optimal classification model was selected to identify the spatial distribution of forest tree species in the study area. The model based on feature optimization gave the best results among the 16 feature combination models. The overall accuracy and kappa coefficient were increased by 18% and 0.21, respectively, compared with the spectral classification model, and by 17% and 0.20, respectively, compared with the spectral and spectral index classification model. By analyzing the feature optimization model, it was found that terrain, ultraviolet aerosol index, and phenological information ranked as the top three features in terms of importance. Although the importance of spectral reflectance and spectral index features was lower, the number of feature variables accounted for a large proportion of the total. The importance of commonly used texture features was limited, and these features were not present in the feature optimization model. The RF algorithm had the highest classification accuracy, with an overall accuracy of 82.69% and a kappa coefficient of 0.80, among the four classification algorithms. The results of GTB were close to those of RF, and the difference in overall classification accuracy was only 0.14%. However, the results of the SVM and CART algorithms were relatively weaker, with overall classification accuracies of about 70%. It can be concluded that the combined application of Sentinel-2 images and auxiliary data can improve forest tree species classification accuracy. The model based on feature optimization achieved the highest classification accuracy among the 16 feature combination models. The spectral reflectance and spectral index data extracted from optical images are useful for tree species classification, but the effect of texture features was very limited. Auxiliary data, such as topographic features, ultraviolet aerosol index, phenological features, NO2 concentration features, topographic diversity features, precipitation features, temperature features, and multi-scale topographic location index data, can effectively improve forest tree species classification accuracy. The RF algorithm had the highest accuracy, and it can be used for tree species classification space distribution identification. The combined application of Sentinel-2 images and auxiliary data can improve classification accuracy, but the highest accuracy of the model was only 82.69%, which leaves room for improvement. Thus, more effective auxiliary data and the vertical structural parameters extracted from satellite LiDAR can be combined with multispectral images to improve forest tree species classification accuracy in future research.

Combining Spectral, Spatial-Contextual, and Structural Information in Multispectral UAV Data for Spruce Crown Delineation

Precise delineation of individual tree crowns is critical for accurate forest biophysical parameter estimation, species classification, and ecosystem modelling. Multispectral optical remote sensors mounted on low-flying unmanned aerial vehicles (UAVs) can rapidly collect very-high-resolution (VHR) photogrammetric optical data that contain the spectral, spatial, and structural information of trees. State-of-the-art tree crown delineation approaches rely mostly on spectral information and underexploit the spatial-contextual and structural information in VHR photogrammetric multispectral data, resulting in crown delineation errors. Here, we propose the spectral, spatial-contextual, and structural information-based individual tree crown delineation (S3-ITD) method, which accurately delineates individual spruce crowns by minimizing the undesirable effects due to intracrown spectral variance and nonuniform illumination/shadowing in VHR multispectral data. We evaluate the performance of the S3-ITD crown delineation method over a white spruce forest in Quebec, Canada. The highest mean intersection over union (IoU) index of 0.83, and the lowest mean crown-area difference (CAD) of 0.14 m2, proves the superior crown delineation performance of the S3-ITD method over state-of-the-art methods. The reduction in error by 2.4 cm and 1.0 cm for the allometrically derived diameter at breast height (DBH) estimates compared with those from the WS-ITD and the BF-ITD approaches, respectively, demonstrates the effectiveness of the S3-ITD method to accurately estimate biophysical parameters.

Effect of feature standardization on reducing the requirements of field samples for individual tree species classification using ALS data

The objective of this study was to evaluate the effectiveness of three standardization approaches for airborne laser scanning (ALS) feature values used for individual tree species classification. This study is the first effort to assess the transferability of forest tree species classification models derived using monospectral and multispectral ALS data. Three research questions were asked; (1) How do the ALS features differ for the same species in different though comparable ecological regions? (2) How to train a model with one sub-population and apply it in another sub-population? (3) How to fuse models for two areas into a global model? To answer these questions, both 3D and intensity features were extracted from the ALS data from Canadian boreal forests. The ALS feature values were standardized in two different scenarios, disjoint areas, and partially overlapping areas, across three study areas. Feature standardization approaches were used: histogram matching, median-based standardization, and linear regression-based standardization. A linear discriminant analysis (LDA) and random forest (RF) algorithms were employed to classify the study area’s major tree species. The Bhattacharyya distance and overall accuracy (OA) were used to assess the classification model performance before and after the feature standardization. Three major conclusions were drawn. First, the Bhattacharyya distance confirmed that intensity features varied across study areas and among tree species, while 3D features were relatively less variable. Second, for the disjoint areas (York Regional Forest (YRF)) and Petawawa Research Forest (PRF)), the feature standardization procedure consistently improved the OA classification for both local model and global model approaches. The feature standardization improved the OA from 16% to 54% using LDA, and from 20% to 55% using RF in the local model. The improvement was from 58% to 66% using LDA, and from 63% to 70% using RF in the global model. It can be concluded that intensity features (at YRF and PRF) were most prone to differ between areas because of scanners and acquisition settings. If ALS data were available from both areas, intensity features need to be normalized so that the local model can be transferred. Finally, for the partially overlapping areas (the northern and southern parts of Black Brook Forest), this study suggests that normalization of ALS data is not needed because they were captured using quite similar ALS settings.

Data Augmentation in Prototypical Networks for Forest Tree Species Classification Using Airborne Hyperspectral Images

Accurate and fine multiple tree species supervised classification based on few-shot learning has attracted close attention from researchers, because the sample collection is often hindered in forests. Prototypical networks (P-Nets), as a simple but efficient few-shot learning method, have significant advantages in forest tree species classification. Nevertheless, the overfitting phenomenon caused by the lack of training samples is still prevalent in few-shot classifiers, which brings challenges to training accurate classification models. In this study, we proposed a novel Proto-MaxUp (PM) framework to minimize the issue of overfitting from the perspective of data augmentation and a feature extraction backbone for tree species classification. Taking Gaofeng Forest Farm (GFF) in Nanning City, Guangxi Province, as the study area, nine tree species, cutting site, and road were classified. First, by analyzing the effects of a series of popular data augmentation methods and their combinations in different parts of the P-Net, several effective data augmentation pools were established. Then, the pools aforementioned were combined with PM to obtain the best classification performance. To verify the robustness and validity of the proposed strategy, we applied PM to the other four popular public hyperspectral datasets and achieved excellent results. Finally, this efficient data augmentation method was used in different feature extraction backbones. The results show that the classification accuracy was greatly improved with the optimal backbone (overall accuracy (OA) and Kappa, are 98.08% and 0.9789, respectively), and the difference between training accuracy and test accuracy is less than 2%. It is concluded that the accurate and fine classification for multiple tree species can be realized by the PM data augmentation strategy and backbone proposed in this article.

Dual-Concentrated Network With Morphological Features for Tree Species Classification Using Hyperspectral Image

At present, deep learning is a hot topic in the field of the classification of hyperspectral image (HSI), and it has aroused wide attention. However, in fine-grained classification tasks, such as tree species classification, the uncertain spectrum remains the major factor restraining the classification performance. In order to solve the dilemma of forest tree species classification, a Dual-concentrated Network with Morphological Features (DNMF) is proposed. Firstly, mathematical morphology is used to extract the morphological features of HSI. Then, coarse-grained information is extracted from the original hyperspectral data, and fine-grained information is extracted from morphological features. After that, both morphological representations and spectral inputs are fed into DNMF, and the overall evaluation index and visual image are obtained. The advantage of DNMF is that it decouples the spatial and spectral information, and a multi-source information fusion process is then simulated. Accordingly, DNMF obtains high tree species classification accuracy. In order to verify the superiority of DNMF, we choose Gaofeng State-owned Forest Farm (GSFF) in Guangxi Province and the Belgium dataset which was collected near the western part of Belgium as the research area. Related experiments demonstrate that the DNMF model achieves clearly better classification performance over other competitive baselines.

An Approach Based on Deep Learning for Tree Species Classification in LiDAR Data Acquired in Mixed Forest

This letter proposes a novel method based on Deep Learning (DL) to forest species classification in airborne Light Detection and Ranging (LiDAR) data. Differently from the state-of- the-art approaches, the proposed method: (1) does not assume any prior knowledge either on the forest to be classified or on the sensor used to acquire the LiDAR data, and (2) can be applied to heterogeneous forest characterized by mixed species. First, the 3D point cloud of each individual tree is decomposed into 8 angular sectors to generate a multi-slices representation of the vertical structure of the tree. This representation models the foliage, the stem and the branches of the tree crown as well as depicts the internal and external crown properties. Then, a Multi-View CNN (MVCNN) DL automatically extracts features used to discriminate the different tree species. This network is pre-trained on the massive ImageNet database, thus guaranteeing fast convergence with a relatively small number of ground reference data. Experiments were carried out on high density airborne LiDAR data collected over a multi-layer multi-age forest characterized by four conifers and three broadleaf species. The proposed method outperformed the state-of-the-art approaches increasing the Overall Accuracy (OA) up to 16% and 18.9% compared to a DL and a shallow tree species classification methods, respectively. When applied to coniferous or broadlaef forests, the proposed method showed an increase of OA 10.1% and 15.9% (for conifers), and 9.5% and 21.6% (for broadleafs) compared to the DL and shallow methods, respectively.

Classification of Typical Tree Species in Laser Point Cloud Based on Deep Learning

We propose the Point Cloud Tree Species Classification Network (PCTSCN) to overcome challenges in classifying tree species from laser data with deep learning methods. The network is mainly composed of two parts: a sampling component in the early stage and a feature extraction component in the later stage. We used geometric sampling to extract regions with local features from the tree contours since these tend to be species-specific. Then we used an improved Farthest Point Sampling method to extract the features from a global perspective. We input the intensity of the tree point cloud as a dimensional feature and spatial information into the neural network and mapped it to higher dimensions for feature extraction. We used the data obtained by Terrestrial Laser Scanning (TLS) and Unmanned Aerial Vehicle Laser Scanning (UAVLS) to conduct tree species classification experiments of white birch and larch. The experimental results showed that in both the TLS and UAVLS datasets, the input tree point cloud density and the highest feature dimensionality of the mapping had an impact on the classification accuracy of the tree species. When the single tree sample obtained by TLS consisted of 1024 points and the highest dimension of the network mapping was 512, the classification accuracy of the trained model reached 96%. For the individual tree samples obtained by UAVLS, which consisted of 2048 points and had the highest dimension of the network mapping of 1024, the classification accuracy of the trained model reached 92%. TLS data tree species classification accuracy of PCTSCN was improved by 2–9% compared with other models using the same point density, amount of data and highest feature dimension. The classification accuracy of tree species obtained by UAVLS was up to 8% higher. We propose PCTSCN to provide a new strategy for the intelligent classification of forest tree species.

A New CBAM-P-Net Model for Few-Shot Forest Species Classification Using Airborne Hyperspectral Images

High-precision automatic identification and mapping of forest tree species composition is an important content of forest resource survey and monitoring. The airborne hyperspectral image contains rich spectral and spatial information, which provides the possibility of high-precision classification and mapping of forest tree species. Few-shot learning, as an application of deep learning, has become an effective method of image classification. Prototypical networks (P-Net) is a simple and practical deep learning network, which has significant advantages in solving few-shot classification problems. Considering the high band correlation and large data volume associated with airborne hyperspectral images, how to fully extract effective features, filter or reduce redundant features is the key to improving the classification accuracy of P-Net, in order to extract effective features in hyperspectral images and obtain a high-precision forest tree species classification model with limited samples. In this research, we embedded the convolutional block attention module (CBAM) between the convolution blocks of P-Net, the CBAM-P-Net was constructed, and a method to improve the feature extraction efficiency of the P-Net was proposed, although this method makes the network more complex and increases the computational cost to a certain extent. The results show that the combination strategy using Channel First for CBAM greatly improves the feature extraction efficiency of the model. In different sample windows, CBAM-P-Net has an average increase of 1.17% and 0.0129 in testing overall accuracy (OA) and kappa coefficient (Kappa). The optimal classification window is 17 × 17, the OA reaches 97.28%, and Kappa reaches 0.97, which is an increase of 1.95% and 0.0214 along with just 49 s of training time expended, respectively, compared with P-Net. Therefore, using a suitable sample window and applying the proposed CBAM-P-Net to classify airborne hyperspectral images can achieve high-precision classification and mapping of forest tree species.