Search

Efficient and precise forest surveys are crucial for in-depth understanding of the present state of forest resources and conducting scientific forest management. Close-range photogrammetry (CRP) technology enables the convenient and fast collection of highly overlapping sequential images, facilitating the reconstruction of 3D models of forest scenes, which significantly improves the efficiency of forest surveys and holds great potential for forestry visualization management. However, in practical forestry applications, CRP technology still presents challenges, such as low image quality and low reconstruction rates when dealing with complex undergrowth vegetation or forest terrain scenes. In this study, we utilized an iPad Pro device equipped with high-resolution cameras to collect sequential images of four plots in Gaofeng Forest Farm in Guangxi and Genhe Nature Reserve in Inner Mongolia, China. First, we compared the image enhancement effects of two algorithms: histogram equalization (HE) and median–Gaussian filtering (MG). Then, we proposed a deep learning network model called SA-Pmnet based on self-attention mechanisms for 3D reconstruction of forest scenes. The performance of the SA-Pmnet model was compared with that of the traditional SfM+MVS algorithm and the Patchmatchnet network model. The results show that histogram equalization significantly increases the number of matched feature points in the images and improves the uneven distribution of lighting. The deep learning networks demonstrate better performance in complex environmental forest scenes. The SA-Pmnet network, which employs self-attention mechanisms, improves the 3D reconstruction rate in the four plots to 94%, 92%, 94%, and 96% by capturing more details and achieves higher extraction accuracy of diameter at breast height (DBH) with values of 91.8%, 94.1%, 94.7%, and 91.2% respectively. These findings demonstrate the potential of combining of the image enhancement algorithm with deep learning models based on self-attention mechanisms for 3D reconstruction of forests, providing effective support for forest resource surveys and visualization management. [ABSTRACT FROM AUTHOR]

An efficient and accurate forest sample plot survey is of great significance to understand the current status of forest resources at the stand or regional scale and the basis of scientific forest management. Close-range photogrammetry (CRP) technology can easily and quickly collect sequence images with high overlapping to reconstruct the 3D model of forest scenes and extract the individual tree parameters automatically and, therefore, can greatly improve the efficiency of forest investigation and has great application potential in forestry visualization management. However, it has some issues in practical forestry applications. First, the imaging quality is affected by the illumination in the forest, resulting in difficulty in feature matching and low accuracy of parameter extraction. Second, the efficiency of 3D forest model reconstruction is limited under complex understory vegetation or the topographic situation in the forest. In addition, the density of point clouds by dense matching directly affects the accuracy of individual tree parameter extraction. This research collected the sequence images of sample plots of four tree species by smartphones in Gaofeng Forest Farm in Guangxi and Wangyedian Forest Farm in Mongolia to analyze the effects of image enhancement, feature detection and dense point cloud algorithms on the efficiency of 3D forest reconstruction and accuracy of individual tree parameter extraction, then proposed a strategy of 3D reconstruction and parameter extraction suitable for different forest scenes. First, we compared the image enhancement effects of median–Gaussian (MG) filtering, single-scale retinex (SSR) and multi-scale retinex (MSR) filtering algorithms. Then, an improved algorithm combining Harris corner detection with speeded-up robust features (SURF) feature detection (Harris+SURF) is proposed, and the feature matching effect is compared with that of a scale invariant feature transform (SIFT) operator. Third, according to the morphological characteristics of the trees in the sequence images, we used the iterative interpolation algorithm of a planar triangulation network based on geometric constraints (GC-based IIPTN) to increase the density of point clouds and reconstruct the 3D forest model, and then extract the position and DBH of the individual trees. The results show that MSR image enhancement can significantly increase the number of matched point pairs. The improved Harris+SURF method can reduce the reconstruction time of the 3D forest model, and the GC-based IIPTN algorithm can improve the accuracy of individual tree parameter extraction. The extracted position of the individual tree is the same as the measured position with the bias within 0.2 m. The accuracy of extracted DBH of Eucalyptus grandis, Taxus chinensis, Larix gmelinii and Pinus tabuliformis is 94%, 95%, 96% and 90%, respectively, which proves that the proposed 3D model reconstruction method based on image enhancement has great potential for tree position and DBH extraction, and also provides effective support for forest resource investigation and visualization management in the future. [ABSTRACT FROM AUTHOR]

Tree species classification is important for the management and sustainable development of forest resources. Traditional object-oriented tree species classification methods, such as support vector machines, require manual feature selection and generally low accuracy, whereas deep learning technology can automatically extract image features to achieve end-to-end classification. Therefore, a tree classification method based on deep learning is proposed in this study. This method combines the semantic segmentation network U-Net and the feature extraction network ResNet into an improved Res-UNet network, where the convolutional layer of the U-Net network is represented by the residual unit of ResNet, and linear interpolation is used instead of deconvolution in each upsampling layer. At the output of the network, conditional random fields are used for post-processing. This network model is used to perform classification experiments on airborne orthophotos of Nanning Gaofeng Forest Farm in Guangxi, China. The results are then compared with those of U-Net and ResNet networks. The proposed method exhibits higher classification accuracy with an overall classification accuracy of 87%. Thus, the proposed model can effectively implement forest tree species classification and provide new opportunities for tree species classification in southern China. [ABSTRACT FROM AUTHOR]

It is essential to classify tree species accurately for the sustainable management of forest resources and effective monitoring of species diversity. Airborne hyperspectral images have high spatial and spectral resolution, and consequently, the large quantity of information on spectral and spatial structures is effective for tree species classification. In this research, Gaofeng Forest Farm in Nanning, Guangxi Province, China, was used as the study site, and the airborne hyperspectral images were used as the data source. The spectral and textural information extracted by wavelet analysis and edge information extracted by mathematical morphological analysis composed a feature set. The feature set was filtered through a random forest, and object-oriented methods were used to classify tree species through a support vector classifier. The results showed that spectral features extracted by wavelet analysis were highly effective in classifying tree species that had the greatest spectral separability. Horizontal and vertical textures had no positive effect on the classification accuracy, while diagonal textures improved the classification accuracy of tree species. Texture features were not sensitive to stands with small areas and broken distributions, while the edge structure features extracted from mathematical morphology were sensitive to the complex forests. The overall accuracy of tree species classification by combining spectral, textural, and edge structural features was 96.54%, with a Kappa coefficient of 0.96. In the comparative test, the first-derivative and second-derivative of the hyperspectral image and texture features composed a feature set. Using the same classification methods, the OA was 80.91% and Kappa was 0.7711. Therefore, the wavelet analysis and mathematical morphology can significantly improve the tree species classification accuracy of hyperspectral images. Accurate tree species classification can provide basic scientific data for forest resource monitoring and management measures. • Wavelet analysis and mathematical morphological analysis were evaluated. • Wavelet analysis could classify tree species with high spectral separability. • Diagonal (not horizontal or vertical) textures improved classification accuracy. • Complex forests were sensitive to edge structure features of mathematical morphology. • This approach facilitates the sustainable management of forest resources and effective monitoring of species diversity. [ABSTRACT FROM AUTHOR]

Airborne hyperspectral remote sensing data with both rich spectral and spatial features can effectively improve the classification accuracy of vegetation species. However, the spectral data of hundreds of bands brings about problems such as dimensional explosion, which poses a huge challenge for hyperspectral remote sensing classification based on classical parameters models. Deep learning methods have been used for remotely sensed images classification in recent years, but the popular HSI datasets including Kennedy Space Center, Indian Pines, Pavia University scene and Salinas scene, have low spatial resolution, significant differences between categories, and regular boundaries. When applied to the classification of forestry tree species, the accuracy often decreases because the spectral response of different plants of the same family and genus are very similar, especially under the fragmented species distribution, complex topography and the occluded canopy. So we collect new data sets, selected Gaofeng State Owned Forest Farm in Guangxi province in south China as the research area and adopted the airborne hyperspectral data obtained by the LiCHy system of the Chinese Academy of Forestry to explore an improved three-dimensional convolutional neural network(3D-CNN) model for tree species classification. The proposed model uses raw data as input without dimension reduction or feature screening, and simultaneously extracts spectral and spatial features. After the 3D convolutional layer, the captured high-level semantic concept is a joint spatial spectral feature representation, so we can turn it into a one-dimensional feature as a new input to learn a more abstract level of expression. The widely used earlystop method is also used to prevent overfitting. The proposed model is a lightweight, generalized, and fast convergence classification model, by which the short-time and large-area of multiple tree species classification with high-precision can be realized. The result shows that the 3D-1D CNN model can shorten the training time of the 3D CNN model by 60% and achieve a classification accuracy of 93.14% within 50 ha in 6.37 min, which provides a basis for the classification of tree species, the mapping of forest form and the inventory of forest resources. • The 3D CNN is applied to tree species classification of actual hyperspectral images. • The improved model accelerates the computational efficiency while ensuring accuracy. • The model can be used for classifications with only a very small number of samples. • Using feature map and t-SNE to provide the possibility to explore the CNN black box. [ABSTRACT FROM AUTHOR]

• Proposing a solution for tree height estimation based on trunk curve. • Estimating the trunk volume combining segmentation measurement and cylinder fitting. • Achieving automatic measurement of the individual tree parameters using CRP data. The forest plots survey's primary measurement factors are the diameter at breast height (DBH), tree height, and the individual tree volume. The reconstruction of a 3D forest scene based on dense images can contribute to the automatic extraction of individual tree parameters. Close-range photogrammetry (CRP) can achieve the reconstruction of a 3D scene from the sequence images at a highly efficient and low cost. However, the reconstructed point clouds often miss the trees' upper part, resulting in the tree height and volume not being measured directly. This paper proposes an integrating extraction framework and methods of the DBH, tree height, and individual tree volume for achieving automatic measurement of the individual tree parameters with high accuracy based on CRP. Four plantation plots in Wangyedian Forest Farm, Inner Mongolia, and Gaofeng Forest Farm, Guangxi Zhuang, are selected to obtain sequence close-up photographs and construct 3D scenes. The DBH, tree height, and trunk volume of the individual tree are extracted using circle fitting, stem curve fitting, growth equation, and segmentation measurement. The extracted individual tree parameters from Terrestrial laser scanning (TLS) are used as validation data to evaluate the accuracy. It shows a good performance in two Larix gmelinii plots, one Pinus tabulaeformis plot, and one Eucalyptus grandis plot with the accuracy of extracted DBH above 90%, which is 90.18%, 96.54%, 90.17%, and 95.00%, respectively. The accuracy of extracted tree height of the four plots is all above 85%, which is 93.34%, 91.71%, 90.32%, and 85.96%, respectively, and the accuracy of trunk volume is 91.55%, 91.94%, 89.86%, and 89.6%, respectively. The results show that our proposed integrated parameter extraction frame and methods can achieve high accuracy economically, efficiently, and conveniently, and be used for automatic measurements of individual tree factors in forest plot surveys. [ABSTRACT FROM AUTHOR]