Object-Based Wetland Vegetation Classification Using Multi-Feature Selection of Unoccupied Aerial Vehicle RGB Imagery

Mangrove Species Classification from Unmanned Aerial Vehicle Hyperspectral Images Using Object-Oriented Methods Based on Feature Combination and Optimization

Accurate and timely acquisition of the spatial distribution of mangrove species is essential for conserving ecological diversity. Hyperspectral imaging sensors are recognized as effective tools for monitoring mangroves. However, the spatial complexity of mangrove forests and the spectral redundancy of hyperspectral images pose challenges to fine classification. Moreover, finely classifying mangrove species using only spectral information is difficult due to spectral similarities among species. To address these issues, this study proposes an object-oriented multi-feature combination method for fine classification. Specifically, hyperspectral images were segmented using multi-scale segmentation techniques to obtain different species of objects. Then, a variety of features were extracted, including spectral, vegetation indices, fractional order differential, texture, and geometric features, and a genetic algorithm was used for feature selection. Additionally, ten feature combination schemes were designed to compare the effects on mangrove species classification. In terms of classification algorithms, the classification capabilities of four machine learning classifiers were evaluated, including K-nearest neighbor (KNN), support vector machines (SVM), random forests (RF), and artificial neural networks (ANN) methods. The results indicate that SVM based on texture features achieved the highest classification accuracy among single-feature variables, with an overall accuracy of 97.04%. Among feature combination variables, ANN based on raw spectra, first-order differential spectra, texture features, vegetation indices, and geometric features achieved the highest classification accuracy, with an overall accuracy of 98.03%. Texture features and fractional order differentiation are identified as important variables, while vegetation index and geometric features can further improve classification accuracy. Object-based classification, compared to pixel-based classification, can avoid the salt-and-pepper phenomenon and significantly enhance the accuracy and efficiency of mangrove species classification. Overall, the multi-feature combination method and object-based classification strategy proposed in this study provide strong technical support for the fine classification of mangrove species and are expected to play an important role in mangrove restoration and management.

Aboveground biomass estimation of wetland vegetation at the species level using unoccupied aerial vehicle RGB imagery

Wetland vegetation biomass is an essential indicator of wetland health, and its estimation has become an active area of research. Zizania latifolia (Z. latifolia) is the dominant species of emergent vegetation in Honghu Wetland, and monitoring its aboveground biomass (AGB) can provide a scientific basis for the protection and restoration of this and other wetlands along the Yangtze River. This study aimed to develop a method for the AGB estimation of Z. latifolia in Honghu Wetland using high-resolution RGB imagery acquired from an unoccupied aerial vehicle (UAV). The spatial distribution of Z. latifolia was first extracted through an object-based classification method using the field survey data and UAV RGB imagery. Linear, quadratic, exponential and back propagation neural network (BPNN) models were constructed based on 17 vegetation indices calculated from RGB images to invert the AGB. The results showed that: (1) The visible vegetation indices were significantly correlated with the AGB of Z. latifolia. The absolute value of the correlation coefficient between the AGB and CIVE was 0.87, followed by ExG (0.866) and COM2 (0.837). (2) Among the linear, quadratic, and exponential models, the quadric model based on CIVE had the highest inversion accuracy, with a validation R² of 0.37, RMSE and MAE of 853.76 g/m² and 671.28 g/m², respectively. (3) The BPNN model constructed with eight factors correlated with the AGB had the best inversion effect, with a validation R² of 0.68, RMSE and MAE of 732.88 g/m² and 583.18 g/m², respectively. ​Compared to the quadratic model constructed by CIVE, the BPNN model achieved better results, with a reduction of 120.88 g/m² in RMSE and 88.10 g/m² in MAE. This study indicates that using UAV-based RGB images and the BPNN model provides an effective and accurate technique for the AGB estimation of dominant wetland species, making it possible to efficiently and dynamically monitor wetland vegetation cost-effectively.

Fine Classification of UAV Urban Nighttime Light Images Based on Object-Oriented Approach

Fine classification of urban nighttime lighting is a key prerequisite step for small-scale nighttime urban research. In order to fill the gap of high-resolution urban nighttime light image classification and recognition research, this paper is based on a small rotary-wing UAV platform, taking the nighttime static monocular tilted light images of communities near Meixi Lake in Changsha City as research data. Using an object-oriented classification method to fully extract the spectral, textural and geometric features of urban nighttime lights, we build four types of classification models based on random forest (RF), support vector machine (SVM), K-nearest neighbor (KNN) and decision tree (DT), respectively, to finely extract five types of nighttime lights: window light, neon light, road reflective light, building reflective light and background. The main conclusions are as follows: (i) The equal division of the image into three regions according to the visual direction can alleviate the variable scale problem of monocular tilted images, and the multiresolution segmentation results combined with Canny edge detection are more suitable for urban nighttime lighting images; (ii) RF has the highest classification accuracy among the four classification algorithms, with an overall classification accuracy of 95.36% and a kappa coefficient of 0.9381 in the far view region, followed by SVM, KNN and DT as the worst; (iii) Among the fine classification results of urban light types, window light and background have the highest classification accuracy, with both UA and PA above 93% in the RF classification model, while road reflective light has the lowest accuracy; (iv) Among the selected classification features, the spectral features have the highest contribution rates, which are above 59% in all three regions, followed by the textural features and the geometric features with the smallest contribution rates. This paper demonstrates the feasibility of nighttime UAV static monocular tilt image data for fine classification of urban light types based on an object-oriented classification approach, provides data and technical support for small-scale urban nighttime research such as community building identification and nighttime human activity perception.

Karst vegetation coverage detection using UAV multispectral vegetation indices and machine learning algorithm

BACKGROUND

Karst vegetation is of great significance for ecological restoration in karst areas. Vegetation Indices (VIs) are mainly related to plant yield which is helpful to understand the status of ecological restoration in karst areas. Recently, karst vegetation surveys have gradually shifted from field surveys to remote sensing-based methods. Coupled with the machine learning methods, the Unmanned Aerial Vehicle (UAV) multispectral remote sensing data can effectively improve the detection accuracy of vegetation and extract the important spectrum features.

RESULTS

In this study, UAV multispectral image data at flight altitudes of 100 m, 200 m, and 400 m were collected to be applied for vegetation detection in a karst area. The resulting ground resolutions of the 100 m, 200 m, and 400 m data are 5.29, 10.58, and 21.16 cm/pixel, respectively. Four machine learning models, including Random Forest (RF), Support Vector Machine (SVM), Gradient Boosting Machine (GBM), and Deep Learning (DL), were compared to test the performance of vegetation coverage detection. 5 spectral values (Red, Green, Blue, NIR, Red edge) and 16 VIs were selected to perform variable importance analysis on the best detection models. The results show that the best model for each flight altitude has the highest accuracy in detecting its training data (over 90%), and the GBM model constructed based on all data at all flight altitudes yields the best detection performance covering all data, with an overall accuracy of 95.66%. The variables that were significantly correlated and not correlated with the best model were the Modified Soil Adjusted Vegetation Index (MSAVI) and the Modified Anthocyanin Content Index (MACI), respectively. Finally, the best model was used to invert the complete UAV images at different flight altitudes.

CONCLUSIONS

In general, the GBM_all model constructed based on UAV imaging with all flight altitudes was feasible to accurately detect karst vegetation coverage. The prediction models constructed based on data from different flight altitudes had a certain similarity in the distribution of vegetation index importance. Combined with the method of visual interpretation, the karst green vegetation predicted by the best model was in good agreement with the ground truth, and other land types including hay, rock, and soil were well predicted. This study provided a methodological reference for the detection of karst vegetation coverage in eastern China.

Fusion Classification of HSI and MSI Using a Spatial-Spectral Vision Transformer for Wetland Biodiversity Estimation

The rapid development of remote sensing technology provides wealthy data for earth observation. Land-cover mapping indirectly achieves biodiversity estimation at a coarse scale. Therefore, accurate land-cover mapping is the precondition of biodiversity estimation. However, the environment of the wetlands is complex, and the vegetation is mixed and patchy, so the land-cover recognition based on remote sensing is full of challenges. This paper constructs a systematic framework for multisource remote sensing image processing. Firstly, the hyperspectral image (HSI) and multispectral image (MSI) are fused by the CNN-based method to obtain the fused image with high spatial-spectral resolution. Secondly, considering the sequentiality of spatial distribution and spectral response, the spatial-spectral vision transformer (SSViT) is designed to extract sequential relationships from the fused images. After that, an external attention module is utilized for feature integration, and then the pixel-wise prediction is achieved for land-cover mapping. Finally, land-cover mapping and benthos data at the sites are analyzed consistently to reveal the distribution rule of benthos. Experiments on ZiYuan1-02D data of the Yellow River estuary wetland are conducted to demonstrate the effectiveness of the proposed framework compared with several related methods.