Intelligent System for Detecting Environmental Problems Based on Multispectral Satellite Images

As a key parameter of land surface energy balance models, near surface air temperature (NSAT) is an important indicator of the surface atmospheric environment and the urban thermal environment. At present, NSAT data are mainly captured by meteorological ground stations. In areas with a sparse distribution of meteorological stations, however, it is not possible to describe the heterogeneity of NSAT in continuous space. With the rapid development of satellite remote sensing technologies, there is now a significant method to retrieve NSAT from multispectral satellite images based on machine learning methods. In the literatures published so far, there is little reported research concerning the comprehensive evaluation and/or the systematic comparison of NSAT retrieval performances based on different machine learning models. Hence, the three most commonly-used machine learning models, Support Vector Regression (SVR), Multilayer Perceptron Neural Network (MLBPN), and Random Forest (RF), have been employed for the NSAT retrieval from various multispectral satellite images of MODIS daytime and nighttime data, Landsat 8 data, and Sentinel-2 data. Comparison of the NSAT retrieval results generated by the different machine learning models from the different types of satellite images reveals that (a) the RF-based model has a better NSAT retrieval performance than the SVR- or MLBPN-based models with respect to both the accuracy and stability, and (b) the NSAT results retrieved from the MODIS data were generally better than those from the Landsat 8 and Sentinel-2 data. To sum up, the conducted research in this article does not only provide a reference for practical applications relevant to NSAT retrievals, but also proposes an efficient RF-based model for NSAT retrieval from multispectral satellite images in continuous space.

Remote sensing techniques are a powerful tool for monitoring littoral zones. Optical sensors can be used to quantify water quality parameters such as suspended sediments. It is possible to estimate the Total Suspended Matter (TSM) concentration using multi-spectral satellite images. In order to extract meaningful information, the satellite data needs to be validated with in situ measurements. The main objective of this work was to quantify the TSM in sea breaking zone, using multi-spectral satellite images. A part of the northwest coast of Portugal, centered around Aveiro, was chosen as a test area. Several methodologies have been used to establish a relationship between the above sea water reflectance and the TSM concentration. Various field trips were done in order to simultaneously obtain water samples and reflectance measurements. A relationship between TSM concentration and reflectance was established for the range 400 - 900 nm. Data from Landsat TM, SPOT HRVIR and ASTER were calibrated and geometric corrected. The reflectance values were used to estimate the TSM concentration using the relationships established using the field measurements. The model coefficients and correlation factors, for identical bands on different sensors, presented a high similarity. The results have been incorporated in a Geographical Information System (GIS).

The main aim of this paper is to develop the accurately map of soil chemical parameters were used for development of the agriculture, forestry, ecological planning, and crop yield production. Soil chemical properties analysis and forecast model was developed and validated with the Wavelet transform methods and multispectral satellite images. At the study area sites, satellite images and soil samples were collected during a similar time. Three most important soil chemical properties such as organic carbon, pH and EC were chosen to development of predication modeling based on the soil chemical information. This valuable information of parameters was analysed according to conventional methods. The observed data of soil was used for the predication modeling of soil chemical properties by MATLAB software. The identification of soil chemical properties was the subject of multi-spectral satellite images through algorithm of soil predication modeling. The real chemical characteristics of the soil are associated to wavelet transformation methods. Forecasting of soil chemical properties and this model can be given more accurate information related to soil nutrient parameters. Now a day’s machine learning programming is an easy to applied on the natural resources and agriculture studies. The chemical characteristics of the soil are compared with the different spectrum wavelengths of the MATLAB program. Therefore, four wavelets models like Daubechies, Symlet, Biorthogonal and Coiflet were selected to development of predication modelling, which wavelet model can be given more accurate information with best model of the soil chemical properties. Also, the coefficient of five key components and soil-chemical values were associated in the MATLAB software. In the semi-arid regions in, India, which components have been highly correlated with soil parameters in the predicated modeling. More detailed information of soil chemical characteristics was provided by four selected wavelet models developed based on the observed data and satellite data. Prediction of soil chemical values has been identified through low and high-frequency satellite images and artificial neural network model. In this study, the neural network wavelet model was used to predicted values related to soil chemical properties in the semi-arid region. The developed two models, like polynomial and ANN, have been validated and compared to the soil chemical properties data, which models can be fitted with the study of soil chemical properties. The results of the study area can be more beneficial for development of agriculture activities, climate change approaches, crop and soil suitability planning. From the results of models have been given a fast and quickly information of soil nutrient parameters without laboratory analysis. The results of predicated values can be more helpful to precision farming related activates and soil fertility mapping to provide the farmers and agriculture scientist.

Ship detection in multispectral remote-sensing images is critical in marine surveillance applications. The previously proposed ship-detection methods for multispectral satellite imagery usually work well under ideal conditions. When meeting complex environments such as shadows, mists, or clouds, they fail to detect ships. To solve this problem, we propose a novel spectral-reflectance-based ship-detection method. Research has shown that different materials have unique reflectance curves in the same spectral wavelength range. Based on this observation, we present a new feature using the reflectance gradient across multispectral bands. Moreover, we propose a neural network called lightweight fusion networks (LFNet). This network combines the aforementioned reflectance and the color information of multispectral images to jointly verify the regions with ships. The method utilizes a coarse-to-fine detection framework because of the large-sense-sparse-targets situation in remote-sensing images. In the coarse stage, the proposed reflectance feature vector is used to input the classifier to rule out the regions without ships. In fine detection, the LFNet is used to verify true ships. Compared with some traditional methods that merely depend on appearance features in images, the proposed method takes advantage of employing the reflectance variance in objects between each band as additional information. Extensive experiments have been conducted on multispectral images from four satellites under different weather and environmental conditions to demonstrate the effectiveness and efficiency of the proposed method. The results show that our method can still achieve good performance even under harsh weather conditions.

Uncertainty of soil reflectance retrieval from SPOT and RapidEye multispectral satellite images using a per-pixel bootstrapped empirical line atmospheric correction over an agricultural region

In this paper a quantitative and qualitative comparison between performance of Multilayer Perceptron and Fuzzy-Artmap neural networks for classification of land cover from multispectral satellite images is reported. It describes parameters and conditions that help to produce accurate classified images. Additionally, it makes a comparison between the two types of neural networks models describing both their advantages and disadvantages.

Classification of ships and icebergs in the Arctic in satellite images is an important problem. We study how to train deep neural networks for improving the discrimination of ships and icebergs in multispectral satellite images. We also analyze synthetic-aperture radar (SAR) images for comparison. The annotated datasets of ships and icebergs are collected from multispectral Sentinel-2 data and taken from the C-CORE dataset of Sentinel-1 SAR images. Convolutional Neural Networks with a range of hyperparameters are tested and optimized. Classification accuracies are considerably better for deep neural networks than for support vector machines. Deeper neural nets improve the accuracy per epoch but at the cost of longer processing time. Extending the datasets with semi-supervised data from Greenland improves the accuracy considerably whereas data augmentation by rotating and flipping the images has little effect. The resulting classification accuracies for ships and icebergs are 86% for the SAR data and 96% for the MSI data due to the better resolution and more multispectral bands. The size and quality of the datasets are essential for training the deep neural networks, and methods to improve them are discussed. The reduced false alarm rates and exploitation of multisensory data are important for Arctic search and rescue services.

To date, countless satellite image fusions have been made, mainly with panchromatic spatial resolution to a multispectral image ratio of 1/4, fewer fusions with lower ratios, and relatively recently fusions with much higher spatial resolution ratios have been published. Apart from this, there is a small number of publications studying the fusion of aerial photographs with satellite images, with the year of image acquisition varying and the dates of acquisition not mentioned. In addition, in these publications, either no quantitative controls are performed on the composite images produced, or the aerial photographs are recent and colorful and only the RGB bands of the satellite images are used for data fusion purposes. The objective of this paper is the study of the addition of multispectral information from satellite images to black and white aerial photographs of the 80s decade (1980–1990) with small difference (just a few days) in their image acquisition date, the same year and season. Quantitative tests are performed in two case studies and the results are encouraging, as the accuracy of the classification of the features and objects of the Earth’s surface is improved and the automatic digital extraction of their form and shape from the archived aerial photographs is now allowed. This opens up a new field of use for the black and white aerial photographs and archived multispectral satellite images of the same period in a variety of applications, such as the temporal changes of cities, forests and archaeological sites.

The accurate estimation of aboveground biomass (AGB) in rubber plantations is essential for predicting rubber production and assessing carbon storage. Multispectral sensors mounted on unmanned aerial vehicles (UAVs) can obtain high spatiotemporal resolution imagery of rubber plantations, offering significant advantages in capturing fine structural details and heterogeneity. However, most previous studies primarily focused on developing biomass estimation models for rubber using machine learning (ML) algorithms in conjunction with feature selection methods based on UAV-acquired multispectral imagery. The reliance on feature selection methods limits the model’s generalizability, robustness, and predictive accuracy. In contrast, deep learning (DL) exhibits considerable promise in extracting features from high-resolution UAV-based multispectral imagery without the need for manual selection. Nonetheless, it remains unclear whether DL can surpass traditional ML methods in improving the AGB estimation accuracy in rubber plantations. To address this, our study evaluated the performance of three ML algorithms (random forest regression, RFR; XGBoost regression, XGBR; categorical boosting regression, CatBoost) combined with feature selection techniques and a deep convolutional neural network (DCNN) using multispectral imagery obtained from UAV for the AGB estimation of rubber plantations. The results indicate that the RFR combined with a principal component analysis (PCA) for feature selection yielded the best performance (R2 = 0.81, RMSE = 11.63 t/ha, MAE = 9.27 t/ha) between the three ML algorithms. Meanwhile, the DCNN model derived from the G, R, and NIR spectral bands achieved the highest estimation accuracy (R2 = 0.89, RMSE = 6.44 t/ha, MAE = 5.72 t/ha), where it outperformed the other ML methods. Our study highlights the great potential of combining UAV-based multispectral imagery with DL techniques to improve AGB estimation in rubber plantations, offering a new perspective for estimating the physiological and biochemical growth parameters of forests.

In this paper, we propose a fuzzy-based multi-spectral (MS) and panchromatic (PAN) image fusion approach which provides a tradeoff solution between spectral and spatial fidelity and is able to preserve more detail in terms of spectral and spatial information. First, we perform principal component analysis on the multi-spectral images and utilise the first principal component to extract matched low and high frequency coefficients. We then apply fuzzy-based image fusion rules to fuse the first principal component with the PAN image, followed by fusing the approximation coefficients. The proposed approach is tested on several satellite images and shown to provide a feasible and effective approach.

The accuracy of mathematical modeling of the hydrological regime for a specific territory is determined by the quality of the digital elevation model (DEM) of this area. A particular challenge is the modeling of floodplain areas with a large number of reservoirs with changing water levels during flood periods. A distinctive feature of such a territory is the complex structure of small channels, vegetation and wetlands. Small changes in topographic elevations can significantly change flood patterns, requiring the construction of a high-precision DEM. Objective – the goal is the analysis of the quality of a digital terrain model based on constructing a system of changing coastlines, processing multispectral satellite images of the area at different time. The software for processing multi-channel satellite images with identification of water bodies in batch mode with subsequent vectorization of coastlines has been created. The possibility of updating altitude data using a set of boundaries of reservoirs using infrared data from the Landsat 7 ETM+ and Landsat 8-9 OLI satellites is shown. The critical values of intensities in the infrared channels have been determined, providing the best identification of the boundaries of reservoirs. The construction of geographically referenced boundaries of water bodies can be used to solve various environmental, hydrological and socio-economic problems. In particular, it is proposed to update elevation data for periodically flooded floodplain areas.

Augmenting a deep-learning algorithm with canal inspection knowledge for reliable water leak detection from multispectral satellite images

Recently, there has been a remarkable growth in Artificial Intelligence (AI) with the development of efficient AI models and high-power computational resources for processing complex datasets. There has been a growing number of applications of machine learning in satellite remote sensing image data processing. In this work, machine learning methods were applied for crop classification of temporal multi- spectral satellite image to achieve better prediction of crop-wise area statistics. In India, agriculture has a huge impact on the national economy and most of the critical decisions are dependent on agricultural statistics. Sentinel-2 satellite image data for the Guntur district region of Andhra Pradesh, India has been used as the study area. The main reason for selecting this region is the diversity of agricultural crops and the availability of ground truth. As a baseline, the performance of machine learning models like Support Vector Machine (SVM), Random Forest (RF), Decision Tree (DT), and K-Nearest Neighbor (KNN) on crop classification was evaluated on 2016 single time multi-spectral satellite image. SVM performed well with an F1 score of 0.94. Further, the performance of SVM, RF, 1D and 2D Convolutional Neural Network (CNN), Recurrent Neural Network (RNN) with Long Short-Term Memory (LSTM) and RNN with Gated Recurrent Unit (GRU) on 2018 Kharif season temporal multi-spectral satellite image has been evaluated for estimation of crop areas. The results show that SVM has the best F1 score of 0.99 and achieved a 95.9% agreement with the ground surveyed crop areas.

The advances in Machine Learning (ML) and computer technologies enabled to process satellite images using programming. Environmental applications that handle Remote Sensing (RS) data for spatial analysis use such an approach, for example, Python’s library scikit-learn using algorithms on pattern identification, predictions or image classification. This paper presents an ML method of satellite image processing using Geographic Resources Analysis Support System (GRASS) Geographic Information System (GIS). The aim is to classify multispectral Landsat images using ML for identification of changes in salt pans of West Mauritania, Africa over the period 2014–2023. We define 10 classes of land cover categories and perform analysis of geological, lithological and landscape setting, and then introduce the principles, algorithms and processing of the ML methods of GRASS GIS. The following classification models were employed to implement image classification with training: Random Forest (RF), Decision Tree, Gradient Boosting and Support Vector Machine (SVM). The results were compared with clustering performed by k-means and maximum likelihood discriminant analysis. The cartographic visualisation and validation was implemented through accuracy analysis. Results for the best performing SVM model with seven-band input produced an overall accuracy of 76%, for the RF model – 73%, compared to 69% for Decision Tree Classifier – 69% and for Gradient Boosting Classifier – 67%. The SVM model embedded in GRASS GIS generates robust land cover maps with good accuracy from multispectral satellite images. The paper demonstrated an ML-based automated approach to satellite image processing, which links Artificial Intelligence (AI) with cartographic tasks.

Radiometric normalization of multi-temporal satellite image is very important for change detection or image mosaic. In this paper, a simple and effective radiometric normalization approach is proposed for processing high resolution multispectral satellite images such as IKONOS or QuickBird. This approach involves three main steps: (1) extracting pixel values from image pairs, (2) sorting original pixel values of image pairs in ascending order, and (3) assigning each pixel a new value based on its reflectance value relative to all other pixels. The effectiveness of the proposed approach is confirmed with experimental results compared to other normalization methods.