Semi-Supervised Tree Species Classification for Multi-Source Remote Sensing Images Based on a Graph Convolutional Neural Network

Wang, Xueliang; Wang, Jian; Lian, Zuozheng; Yang, Nan

doi:10.3390/f14061211

Cited by 8 publications

(3 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Conversely, if the super-pixel size is too small, the segmentation efficiency is reduced. Additionally, compactness must also be balanced: the larger the compactness value, the worse the boundary connectivity [76]. After multiple experiments, this study has set the following five parameters for implementing SNIC: size is 5, compactness is 0, connectivity is 8, neighborhood size is 256, and seed is null.…”

Section: Snic Superpixel Segmentationmentioning

confidence: 99%

Eucalyptus Plantation Area Extraction Based on SLPSO-RFE Feature Selection and Multi-Temporal Sentinel-1/2 Data

Lin,

Ren,

et al. 2023

Forests

View full text Add to dashboard Cite

An accurate and efficient estimation of eucalyptus plantation areas is of paramount significance for forestry resource management and ecological environment monitoring. Currently, combining multidimensional optical and SAR images with machine learning has become an important method for eucalyptus plantation classification, but there are still some challenges in feature selection. This study proposes a feature selection method that combines multi-temporal Sentinel-1 and Sentinel-2 data with SLPSO (social learning particle swarm optimization) and RFE (Recursive Feature Elimination), which reduces the impact of information redundancy and improves classification accuracy. Specifically, this paper first fuses multi-temporal Sentinel-1 and Sentinel-2 data, and then carries out feature selection by combining SLPSO and RFE to mitigate the effects of information redundancy. Next, based on features such as the spectrum, red-edge indices, texture characteristics, vegetation indices, and backscatter coefficients, the study employs the Simple Non-Iterative Clustering (SNIC) object-oriented method and three different types of machine-learning models: Random Forest (RF), Classification and Regression Trees (CART), and Support Vector Machines (SVM) for the extraction of eucalyptus plantation areas. Each model uses a supervised-learning method, with labeled training data guiding the classification of eucalyptus plantation regions. Lastly, to validate the efficacy of selecting multi-temporal data and the performance of the SLPSO–RFE model in classification, a comparative analysis is undertaken against the classification results derived from single-temporal data and the ReliefF–RFE feature selection scheme. The findings reveal that employing SLPSO–RFE for feature selection significantly elevates the classification precision of eucalyptus plantations across all three classifiers. The overall accuracy rates were noted at 95.48% for SVM, 96% for CART, and 97.97% for RF. When contrasted with classification outcomes from multi-temporal data and ReliefF–RFE, the overall accuracy for the trio of models saw an increase of 10%, 8%, and 8.54%, respectively. The accuracy enhancement was even more pronounced when juxtaposed with results from single-temporal data and ReliefF-RFE, at increments of 15.25%, 13.58%, and 14.54% respectively. The insights from this research carry profound theoretical implications and practical applications, particularly in identifying and extracting eucalyptus plantations leveraging multi-temporal data and feature selection.

show abstract

Section: Snic Superpixel Segmentationmentioning

confidence: 99%

Eucalyptus Plantation Area Extraction Based on SLPSO-RFE Feature Selection and Multi-Temporal Sentinel-1/2 Data

Lin,

Ren,

et al. 2023

Forests

View full text Add to dashboard Cite

show abstract

“…OA is used to calculate the proportion of all correctly classified samples to the total samples from the overall point of view [34,35]. Kappa is the consistency measure of the actual and classification values, which can improve the OA shortcomings, ignoring the classification effect of categories with a small number of samples [36,37]. Combining the two evaluation indicators can obtain more reliable classification effect evaluation results.…”

Section: Evaluation Indicatorsmentioning

confidence: 99%

Improved SR-SSIM Band Selection Method Based on Band Subspace Partition

Ting-xing

Gao

et al. 2023

Remote Sensing

View full text Add to dashboard Cite

Scholars have performed much research on reducing the redundancy of hyperspectral data. As a measure of the similarity between hyperspectral bands, structural similarity is used in band selection methods. However, existing structural similarity methods calculate all the structural similarity between bands, which leads to excessively long runtimes for these methods. Aiming to address this problem, this paper proposes a band subspace partition method and combines it with the SR-SSIM band selection method to obtain an improved band selection method: E-SR-SSIM. E-SR-SSIM consists of two parts: band subspace partition and band subspace band selection. In the first part, the hyperspectral dataset is divided into subdatasets corresponding to a number of subspaces. In the second part, a modified SR-SSIM method is used for all subdatasets to select the most representative band in each subdataset. The Indian Pines, Salinas Kennedy Space Center and Wuhan unmanned aerial vehicle-borne hyperspectral image LongKou public datasets are used to implement the experiment. The experiment uses random forest as the supervised classifier: 10% of each category sample is randomly selected as training data, and the remaining 90% is used as test data. The evaluation indicators selected in the experiment are overall accuracy, average accuracy, kappa and recall. The experimental results show that E-SR-SSIM can effectively reduce the runtime while ensuring classification compared with SR-SSIM, and quantitative proof that the band subspace partition reduces the calculated amount of structural similarity is obtained through a mathematical analysis. The improved band subspace partition method could partition a dataset more reasonably than the original band subspace partition method.

show abstract

“…However, the majority of these computer vision approaches suffer from the key practical disadvantage of needing extensive image labelling [40], implying that response variables in all the pixels in every image have to be manually determined. Consequently, over recent years, manifold approaches for mitigating this disadvantage have appeared, leading to weakly supervised and semi-supervised methods [41][42][43], with applications often focused on classification rather than regression.…”

Section: Introductionmentioning

confidence: 99%

Mapping and analyzing the spatial drivers of Chinese subtropical forests features from Gaofen 1 satellite imagery

González-Rodríguez,

2023

Preprint

View full text Add to dashboard Cite

Understanding the spatial variations of forest features and their underlying drivers is crucial for managing and preserving forest ecosystems. Remote sensing techniques, due to their vast spatial coverage, are proving an essential tool for mapping and analyzing the complex spatial patterns of forests characteristics. In the current study, we use high-resolution Gaofen 1 satellite imagery to map features, such as biomass, biodiversity, and habitat type, of the subtropical forests in the Nanling area, a major biodiversity hot spot in the south of China; for subsequently analyzing the association between forest features and environmental drivers. We based our modeling approach on a semi-supervised fully convolutional neural network consisting of a ResNet architecture and a masking and aggregation module for computing predictions for forest survey plots of varied sizes. The fitted convolutional models yielded very high performance statistics, with test R2 values above 0.90 for basal area, aboveground biomass and Shannon diversity index regressions, slightly lower values for the regression of species richness (R2=0.83), and average test classification accuracy of forest types of 0.89, with lower performance for infrequent forest types, such as dwarf forests. The association analysis between predicted maps of forest features and environmental drivers revealed significant correlations being the strongest drivers the altitude, nitrogen content and climate variables, such as mean annual precipitation and temperature. Altitude was strongly associated with dwarf forests and high basal area and high aboveground biomass locations. Basal area and aboveground biomass were also positively influenced by annual rainfall and soil nitrogen and carbon, and negatively by temperature and radiation variables. Species richness was associated with high nitrogen and carbon contents, while the Shannon diversity index showed overall mild correlations with most drivers, with no remarkable patterns. We concluded that our semi-supervised convolutional approach using Gaofen 1 imagery proved useful for modeling the spatial distribution of the subtropical forests in Nanling; however, more research and data collection is necessary to improve the performance of species richness modeling and segmentation of infrequent forest types. Moreover, the strong associations found between climate, and predicted maps of biomass and forest types, provided key insights into the potential repercussions of climate change on Nanling forests.

show abstract

Semi-Supervised Tree Species Classification for Multi-Source Remote Sensing Images Based on a Graph Convolutional Neural Network

Cited by 8 publications

References 31 publications

Eucalyptus Plantation Area Extraction Based on SLPSO-RFE Feature Selection and Multi-Temporal Sentinel-1/2 Data

Eucalyptus Plantation Area Extraction Based on SLPSO-RFE Feature Selection and Multi-Temporal Sentinel-1/2 Data

Improved SR-SSIM Band Selection Method Based on Band Subspace Partition

Mapping and analyzing the spatial drivers of Chinese subtropical forests features from Gaofen 1 satellite imagery

Contact Info

Product

Resources

About