Search

Locust plagues inflict severe agricultural damage. Climate change-induced extreme events like rainfall and droughts have expanded locust habitats. These new areas, often beyond routine monitoring, could become potential high-risk locust areas (PHRLA). Quantitatively understanding the outbreak mechanism driving drought and flood dynamics is crucial for identifying PHRLA, but such studies are scarce. To address this gap, we conducted a case study on locust outbreaks in Xiashan Reservoir, the largest reservoir in Shandong Province, China, in 2017 and 2018. Using time series satellite imagery and meteorological products, we quantitatively analyzed how drought–flood dynamics and temperature affect locust habitats, reproduction, and aggregation. Employing an object-oriented random forest classifier, we generated locust habitat classification maps with 93.77% average overall accuracy and Kappa coefficient of 0.90. Combined with meteorological analysis, we found that three consecutive drought years from 2014 to 2016 reduced the water surface area by 75%, expanding suitable habitats (primarily reeds and weeds) to cover 60% of the reservoir. Warm winters and high temperatures during locust key growth periods, coupled with expanding suitable habitats, promoted multi-generational locust reproduction. However, substantial flooding events in 2017 and 2018, driven by plentiful rainfall during key growth periods, reduced suitable habitats by approximately 54% and 29%, respectively. This compression led to high locust density, causing the locust plague and high-density spots of locusts (HDSL). Our study elucidates locust plague outbreak mechanisms under dynamic drought and flood environments. Based on this, we propose an approach to identify PHRLA by monitoring changes in drought and flood patterns around water bodies and variations in suitable habitat size and distribution, as well as surrounding topography. These findings hold significant implications for enhancing locust monitoring and early warning capabilities, reducing pesticide usage, and ensuring food and ecological security and sustainable agriculture. [ABSTRACT FROM AUTHOR]

Featured Application: Change detection for remote sensing images. Aiming at the problem of misdetection caused by the traditional texture characteristic extraction model, which does not describe the correlation among multiple bands, an object-oriented remote sensing image change detection method based on a color co-occurrence matrix is proposed. First, the image is divided into multi-scale objects by graph-based superpixel segmentation, and the optimal scale is determined by the overall goodness F-measure (OGF). Then, except for the extraction of the spectral features, the multi-channel texture features based on the color co-occurrence matrix (CCM) are extracted to consider the correlation among multiple bands. To accurately find the representative features to overcome the impact of feature redundancy, a cumulative backward search strategy (CBSS) is further designed. Finally, the change detection is completed by inputting the difference image of dual time points to the trained random forest model. Taking Shenzhen and Dapeng as the study areas, with Sentinel-2 and Skysat images under different spatial resolutions, and the forest–bareland change type as an example, the effectiveness of the proposed algorithm is verified by qualitative and quantitative analyses. They show that the proposed algorithm can obtain higher detection accuracy than the texture features without band correlation. [ABSTRACT FROM AUTHOR]

Linear spectral mixture analysis (LSMA) and regression analysis are the two most conventionally used methods to estimate impervious surfaces at the subpixel scale in an urban area. However, LSMA lacks the sensitivity to pixel brightness, which leads to inter variability of endmembers and affects the ability to distinguish features with a similar spectral signature. This research aims to develop LSMA aided by a regression analysis model to estimate impervious surfaces with higher accuracy. A spectral angle mapping (SAM) based regression analysis model is introduced to reduce errors. Based on high-resolution images and field survey data, the SAM-based regression analysis can estimate non-impervious surface and high-impervious surface densities with high accuracy, while less accurate in impervious surfaces with low/medium density. In contrast, LSMA is able to estimate low/medium-density impervious surfaces with higher accuracy. We propose an improved approach by integrating the two methods, regression analysis aided LSMA, for impervious surface estimation. The proposed method increases the overall accuracy of the impervious surface estimation to 85.24%, which is significantly greater than that of the conventional methods. [ABSTRACT FROM AUTHOR]

Deep-learning-based object detectors have substantially improved state-of-the-art object detection in remote sensing images in terms of precision and degree of automation. Nevertheless, the large variation of the object scales makes it difficult to achieve high-quality detection across multiresolution remote sensing images, where the quality is defined by the Intersection over Union (IoU) threshold used in training. In addition, the imbalance between the positive and negative samples across multiresolution images worsens the detection precision. Recently, it was found that a Cascade region-based convolutional neural network (R-CNN) can potentially achieve a higher quality of detection by introducing a cascaded three-stage structure using progressively improved IoU thresholds. However, the performance of Cascade R-CNN degraded when the fourth stage was added. We investigated the cause and found that the mismatch between the ROI features and the classifier could be responsible for the degradation of performance. Herein, we propose a Cascade R-CNN++ structure to address this issue and extend the three-stage architecture to multiple stages for general use. Specifically, for cascaded classification, we propose a new ensemble strategy for the classifier and region of interest (RoI) features to improve classification accuracy at inference. In localization, we modified the loss function of the bounding box regressor to obtain higher sensitivity around zero. Experiments on the DOTA dataset demonstrated that Cascade R-CNN++ outperforms Cascade R-CNN in terms of precision and detection quality. We conducted further analysis on multiresolution remote sensing images to verify model transferability across different object scales. [ABSTRACT FROM AUTHOR]

Hyperspectral remote sensing image classification has been widely employed for numerous applications, such as environmental monitoring, agriculture, and mineralogy. During such classification, the number of training samples in each class often varies significantly. This imbalance in the dataset is often not identified because most classifiers are designed under a balanced dataset assumption, which can distort the minority classes or even treat them as noise. This may lead to biased and inaccurate classification results. This issue can be alleviated by applying preprocessing techniques that enable a uniform distribution of the imbalanced data for further classification. However, it is difficult to add new natural features to a training model by artificial combination of samples by using existing preprocessing techniques. For minority classes with sparse samples, the addition of sufficient natural features can effectively alleviate bias and improve the generalization. For such an imbalanced problem, semi-supervised learning is a creative solution that utilizes the rich natural features of unlabeled data, which can be collected at a low cost in the remote sensing classification. In this paper, we propose a novel semi-supervised learning-based preprocessing solution called NearPseudo. In NearPseudo, pseudo-labels are created by the initialization classifier and added to minority classes with the corresponding unlabeled samples. Simultaneously, to increase reliability and reduce the misclassification cost of pseudo-labels, we created a feedback mechanism based on a consistency check to effectively select the unlabeled data and its pseudo-labels. Experiments were conducted on a state-of-the-art representative hyperspectral dataset to verify the proposed method. The experimental results demonstrate that NearPseudo can achieve better classification accuracy than other common processing methods. Furthermore, it can be flexibly applied to most typical classifiers to improve their classification accuracy. With the intervention of NearPseudo, the accuracy of random forest, k-nearest neighbors, logistic regression, and classification and regression tree increased by 1.8%, 4.0%, 6.4%, and 3.7%, respectively. This study addresses research a gap to solve the imbalanced data-based limitations in hyperspectral image classification. [ABSTRACT FROM AUTHOR]

Quantifying the gap between actual and exploitable yields (Ya and Ye, respectively) at the regional scale is critical for understanding the scope and hotspots for future yield improvements. Remote sensing can provide geospatially continuous observations of crop growth, which is perceived as an effective means to overcome the scaling issue commonly suffered by regional yield gap (YG) studies. However, the consistency and uncertainties of different modeling paradigms (i.e., model-driven or data-driven paradigm) remain to be investigated when integrating remote sensing data into YG analysis. This study provides a comprehensive assessment of process-based and machine learning approaches for estimating rice YG over Northeast China. Three machine learning models are examined for predicting Ya, the results of which reveal a prevalent deviation in the estimation of extreme yields. A rotation scheme is thereby proposed to correct this error, and the outperforming random forest model is selected to be compared with a biophysical process-based model (BEPS). County-scale validation reveals good accuracy of both the process-based and machine learning models in estimating Ya, with RMSEs of 1355.3 kg ha−1 (17.3%) and 1205.0 kg−1 (15.4%), respectively. The process-based model reproduces well the probability distribution of Ya in peaks and tails, which guarantees the reasonable estimation of Ye, given that Ye is derived from the high percentile of Ya. In contrast, the machine learning model trained over county-scale yield statistics underestimates the heterogeneity of Ya, leading to significant shrinkage of Ye and thus YG. Degraded model consistency is found from the estimation of Ya, Ye to YG, yet the discrepancy presents limited impacts on identifying the regional priorities for closing YG. The remote sensing-based estimations under the "top-down" framework are further evaluated against the results of the Global Yield Gap Atlas (GYGA), which follows a "bottom-up" protocol. Multiscale comparisons (i.e., local, subregional and regional) with the GYGA and other reference results suggest that the pixelwise modeling of crop-growing processes constitutes a more reliable approach for estimating rice YG over Northeast China, although uncertainty exists in both remote sensing approaches in the local YG estimation. The need to train machine learning models using representative datasets at the field scale is strongly emphasized. This study contributes to the first comprehensive understanding of different modeling paradigms for YG analysis and has important implications for promoting model development and ensuring food security. [Display omitted] • Use model- and data-driven approaches to assess regional yield gap (YG) with remote sensing (RS). • Degraded model consistency is found when estimating actual yield, exploitable yield, and YG. • Provide multiscale comparisons between the RS-based (top-down) and GYGA (bottom-up) protocols. • Pixelwise crop growth modelling produces more reliable YG estimates than the ML model. • Training with county-scale yield statistics diminishes the ability of ML models to estimate YG. [ABSTRACT FROM AUTHOR]

This letter presents preliminary results of sugarcane mapping in its tillering period using quad-polarization data. Three TerraSAR-X images were collected in early, middle, and late tillering periods in the Leizhou Peninsula, South China. Polarization features based on Cloude–Pottier decomposition, such as scattering angle and polarization entropy, were used to analyze sugarcane scattering mechanism temporal behavior and the differences between sugarcane and other typical land-cover types. Results show good separability of sugarcane from other land-cover types. At last, based on common features of sugarcane in tillering period, a sugarcane mapping method was proposed with certain high accuracy and good stability. The results suggest that the tillering period, from early April to late May, is a suitable growth period for sugarcane mapping, and polarization features appear promising for crop mapping and monitoring. [ABSTRACT FROM PUBLISHER]

Vegetation is an important element and has such a complexity and uncertainty in remote sensing image that the extraction of vegetation by using remote sensing is a very difficult work. Vegetation is extracted using rich ETM+data in mineralized alteration zone of west Tianshan, Xinjiang to subsequently extract the minerals alteration. In order to ensure better minerals alteration, the feature-oriented principal component analysis + optimal density segmentation was proposed based on the analysis of spectral feature of vegetation and minerals alteration. The results showed that using ETM+ band 3 and band 4 combinations can better extract vegetation and this method is of great value in use. [ABSTRACT FROM AUTHOR]

[Display omitted] • The landuse of wetland areas remained stable despite anthropogenic disturbance. • Wetlands located in the insular area are severely impacted by sea-level rise. • The continental coastal wetlands show the polarization phenomenon. • Buffer zones should be established to alleviate the polarization effects. • RSEI was applied to monitor the wetlands dynamics in NE Italy from 1984 to 2016. The Adriatic coast of Northeast Italy is one of the representative areas holding high ecological value in the Mediterranean. In this research, we have monitored the environmental change of coastal wetland in ten studied sites that covered all the Ramsar sites and most of the natural reserves in Northeast Italy from 1984 to 2016. The results show that the coastal wetland environment has been changing associated with a multitude of influential factors, including economic development, reserve initiatives, and sea-level rise, which have been quantitively analyzed for each of the studied wetland sites. The results of the analysis confirm that (1) with robust reservation policy, the landuse of the coastal wetland sites remained generally stable despite the continuous anthropogenic disturbance; (2) wetlands located in the insular area with low elevation are subject to more severe impact from sea-level rise inundation; (3) the continental coastal wetlands in reserve area show the polarization phenomenon which is strengthened by strict reservation policy and contiguous aggressive anthropogenic activities. To protect the insular coastal wetlands and reduce the polarization effects on continental coastal wetlands, more efficient measures should be taken to strengthen and implement the existing reservation directives and regulations actively. [ABSTRACT FROM AUTHOR]

Oblique sharp and repetitive stripes in remote sensing imagery can distractingly and obstructively affect the interpretation and application of remote sensing data. A new methodology for removing oblique stripes, based on discrete wavelet transform (DWT), is proposed. DWT has been used to transform a multispectral image into a frequency domain to find the frequency components caused by stripes. The new technique has been tested taking CMODIS imagery as an example and evaluated by a number of indices. The results show that the proposed method can remove oblique stripes effectively and keep most image information. [ABSTRACT FROM AUTHOR]

The deep-learning-network performance depends on the accuracy of the training samples. The training samples are commonly labeled by human visual investigation or inherited from historical land-cover or land-use maps, which usually contain label noise, depending on subjective knowledge and the time of the historical map. Helping the network to distinguish noisy labels during the training process is a prerequisite for applying the model for training across time and locations. This study proposes an antinoise framework, the Weight Loss Network (WLN), to achieve this goal. The WLN contains three main parts: (1) the segmentation subnetwork, which any state-of-the-art segmentation network can replace; (2) the attention subnetwork (λ); and (3) the class-balance coefficient (α). Four types of label noise (an insufficient label, redundant label, missing label and incorrect label) were simulated by dilate and erode processing to test the network's antinoise ability. The segmentation task was set to extract buildings from the Inria Aerial Image Labeling Dataset, which includes Austin, Chicago, Kitsap County, Western Tyrol and Vienna. The network's performance was evaluated by comparing it with the original U-Net model by adding noisy training samples with different noise rates and noise levels. The result shows that the proposed antinoise framework (WLN) can maintain high accuracy, while the accuracy of the U-Net model dropped. Specifically, after adding 50% of dilated-label samples at noise level 3, the U-Net model's accuracy dropped by 12.7% for OA, 20.7% for the Mean Intersection over Union (MIOU) and 13.8% for Kappa scores. By contrast, the accuracy of the WLN dropped by 0.2% for OA, 0.3% for the MIOU and 0.8% for Kappa scores. For eroded-label samples at the same level, the accuracy of the U-Net model dropped by 8.4% for OA, 24.2% for the MIOU and 43.3% for Kappa scores, while the accuracy of the WLN dropped by 4.5% for OA, 4.7% for the MIOU and 0.5% for Kappa scores. This result shows that the antinoise framework proposed in this paper can help current segmentation models to avoid the impact of noisy training labels and has the potential to be trained by a larger remote sensing image set regardless of the inner label error. [ABSTRACT FROM AUTHOR]

Accurate and efficient extraction of cultivated land data is of great significance for agricultural resource monitoring and national food security. Deep-learning-based classification of remote-sensing images overcomes the two difficulties of traditional learning methods (e.g., support vector machine (SVM), K-nearest neighbors (KNN), and random forest (RF)) when extracting the cultivated land: (1) the limited performance when extracting the same land-cover type with the high intra-class spectral variation, such as cultivated land with both vegetation and non-vegetation cover, and (2) the limited generalization ability for handling a large dataset to apply the model to different locations. However, the "pooling" process in most deep convolutional networks, which attempts to enlarge the sensing field of the kernel by involving the upscale process, leads to significant detail loss in the output, including the edges, gradients, and image texture details. To solve this problem, in this study we proposed a new end-to-end extraction algorithm, a high-resolution U-Net (HRU-Net), to preserve the image details by improving the skip connection structure and the loss function of the original U-Net. The proposed HRU-Net was tested in Xinjiang Province, China to extract the cultivated land from Landsat Thematic Mapper (TM) images. The result showed that the HRU-Net achieved better performance (Acc: 92.81%; kappa: 0.81; F1-score: 0.90) than the U-Net++ (Acc: 91.74%; kappa: 0.79; F1-score: 0.89), the original U-Net (Acc: 89.83%; kappa: 0.74; F1-score: 0.86), and the Random Forest model (Acc: 76.13%; kappa: 0.48; F1-score: 0.69). The robustness of the proposed model for the intra-class spectral variation and the accuracy of the edge details were also compared, and this showed that the HRU-Net obtained more accurate edge details and had less influence from the intra-class spectral variation. The model proposed in this study can be further applied to other land cover types that have more spectral diversity and require more details of extraction. [ABSTRACT FROM AUTHOR]

Detailed and accurate information on the spatial variation of land cover and land use is a critical component of local ecology and environmental research. For these tasks, high spatial resolution images are required. Considering the trade-off between high spatial and high temporal resolution in remote sensing images, many learning-based models (e.g., Convolutional neural network, sparse coding, Bayesian network) have been established to improve the spatial resolution of coarse images in both the computer vision and remote sensing fields. However, data for training and testing in these learning-based methods are usually limited to a certain location and specific sensor, resulting in the limited ability to generalize the model across locations and sensors. Recently, generative adversarial nets (GANs), a new learning model from the deep learning field, show many advantages for capturing high-dimensional nonlinear features over large samples. In this study, we test whether the GAN method can improve the generalization ability across locations and sensors with some modification to accomplish the idea "training once, apply to everywhere and different sensors" for remote sensing images. This work is based on super-resolution generative adversarial nets (SRGANs), where we modify the loss function and the structure of the network of SRGANs and propose the improved SRGAN (ISRGAN), which makes model training more stable and enhances the generalization ability across locations and sensors. In the experiment, the training and testing data were collected from two sensors (Landsat 8 OLI and Chinese GF 1) from different locations (Guangdong and Xinjiang in China). For the cross-location test, the model was trained in Guangdong with the Chinese GF 1 (8 m) data to be tested with the GF 1 data in Xinjiang. For the cross-sensor test, the same model training in Guangdong with GF 1 was tested in Landsat 8 OLI images in Xinjiang. The proposed method was compared with the neighbor-embedding (NE) method, the sparse representation method (SCSR), and the SRGAN. The peak signal-to-noise ratio (PSNR) and structural similarity (SSIM) were chosen for the quantitive assessment. The results showed that the ISRGAN is superior to the NE (PSNR: 30.999, SSIM: 0.944) and SCSR (PSNR: 29.423, SSIM: 0.876) methods, and the SRGAN (PSNR: 31.378, SSIM: 0.952), with the PSNR = 35.816 and SSIM = 0.988 in the cross-location test. A similar result was seen in the cross-sensor test. The ISRGAN had the best result (PSNR: 38.092, SSIM: 0.988) compared to the NE (PSNR: 35.000, SSIM: 0.982) and SCSR (PSNR: 33.639, SSIM: 0.965) methods, and the SRGAN (PSNR: 32.820, SSIM: 0.949). Meanwhile, we also tested the accuracy improvement for land cover classification before and after super-resolution by the ISRGAN. The results show that the accuracy of land cover classification after super-resolution was significantly improved, in particular, the impervious surface class (the road and buildings with high-resolution texture) improved by 15%. [ABSTRACT FROM AUTHOR]

[Display omitted] • The novel CRSEI minimizes error in continuous monitoring of environmental quality. • Multitemporal bias of standard RSEI is reduced with a change detection algorithm. • CRSEI reveals actual environmental-quality changes in East/Southeast Asian coasts. • North China coasts have low environmental quality but see the most improvement. • Southeast Asian coasts see major environmental deterioration caused by human activity. With urbanization and anthropogenic pressures, the eco-environmental quality of East and Southeast Asian coastal regions has undergone drastic changes in the 21st century. To assess these environmental variations and realize regional sustainable development goals, a quantitative monitoring of the eco-environment is necessary, but challenging. The remote sensing ecological index (RSEI) was recently proposed and applied to quantitatively monitor the eco-environmental quality, but bias during multitemporal observations is a major limitation. Thus, a novel method, Continuous Remote Sensing Ecological Index (CRSEI), was developed in this study to enable constant and objective monitoring of large-scale and long-term environmental changes in the coastal regions of East and Southeast Asia. To reduce the multitemporal bias of pixels caused by imaging conditions and sensor status, a change detection algorithm requiring limited auxiliary data was developed. High-resolution images show that the CRSEI enables adequate multitemporal monitoring of eco-environmental changes with fewer false fluctuations on the local scale. The proposed CRSEI should deepen the current understanding of environmental changes and provide a solid scientific reference for environmental protection on multiple scales. The results of this research indicate that the eco-environmental quality of the east and southeast Asian coast remained generally stable during the studied period (2001–2019), with the mean CRSEI changed slightly from 5918 to 5856. However, on a regional level, each area revealed different conditions. First, the Korean Peninsula had relatively stable and well-maintained eco-environmental quality with a CRSEI of around 5500 during 2001–2019. Second, The North China coast had lower eco-environmental quality compared to other regions, but it showed the most improvement, with the CRSEI increasing from 2719 to 2807. Third, the eco-environmental quality of South China and Myanmar's coastal region remained relatively stable, with a CRSEI of around 5730 and 6720, respectively. Fourth, the continental coast of some Southeast Asian countries, particularly Vietnam and Cambodia, experienced a significant decline in eco-environmental quality, as indicated by a dramatic decrease in the CRSEI from 6751 to 6527. Notably, the areas with declining eco-environmental quality were concentrated in the western border regions of Cambodia, where unrestricted cultivation has led to the destruction of tropical rainforests. [ABSTRACT FROM AUTHOR]