Convolutional Neural Network-Based Land Cover Classification Using 2-D Spectral Reflectance Curve Graphs With Multitemporal Satellite Imagery

Recent advances in land use and land cover (LULC) classification with remote sensing imagery are driven by state-of-the-art models such as Convolutional Neural Networks (CNNs). Advanced CNN architecture like ResNet can enhance overall classification performance by incorporating residual skip connections. The integration of 3D feature extraction and ResNet architecture suggests a potential improvement in classification tasks. This paper explores the potential of the 3D ResNet model for LULC classification, comparing it with baseline approaches (Support Vector Machine, Random Forest, XGBoost, 1D CNN, 3D CNN) and state-of-the-art 3D models (3D VGG, 3D DenseNet) using WorldView-2 satellite imagery. The 3D ResNet-18 model, fine-tuned via transfer learning on multispectral images, demonstrates significant improvements in classification performance over machine learning models. It achieves the highest Overall Accuracy (OA) of 99.66% and Kappa Accuracy (KA) of 99.39% on the primary dataset. Despite having slightly lower performance on the external validation dataset (OA:82.89%, KA:80.05%) than 3D DenseNet, it is highly efficient with processing times of 490.2 minutes and 3.6 minutes for both datasets respectively. McNemar’s test results show 3D ResNet and 3D DenseNet have significant differences in classification performance (p<0.05) against other models consistently for both datasets.

Over large areas, land cover classification has conventionally been undertaken using satellite time series. Typically temporal metric percentiles derived from single pixel location time series have been used to take advantage of spectral differences among land cover classes over time and to minimize the impact of missing observations. Deep convolutional neural networks (CNNs) have demonstrated potential for land cover classification of single date images. However, over large areas and using time series their application is complicated because they are sensitive to missing observations and they may misclassify small and spatially fragmented surface features due to their spatial patch-based implementation. This study demonstrates, for the first time, a one-dimensional (1D) CNN single pixel time series land classification approach that uses temporal percentile metrics and that does not have these issues. This is demonstrated for all the Conterminous United States (CONUS) considering two different 1D CNN structures with 5 and 8 layers, respectively. CONUS 30 m land cover classifications were derived using all the available Landsat-5 and -7 imagery over a seven-month growing season in 2011 with 3.3 million 30 m land cover class labelled samples extracted from the contemporaneous CONUS National Land Cover Database (NLCD) 16 class land cover product. The 1D CNNs and, a conventional random forest model, were trained using 10%, 50% and 90% samples, and the classification accuracies were evaluated with an independent 10% proportion. Temporal metrics were classified using 5, 7 and 9 percentiles for each of five Landsat reflective wavelength bands and their eight band ratios. The CONUS and detailed 150 × 150 km classification results demonstrate that the approach is effective at scale and locally. The 1D CNN classification land cover class boundaries were preserved for small axis dimension features, such as roads and rivers, with no stripes or anomalous spatial patterns. The 8-layer 1D CNN provided the highest overall classification accuracies and both the 5-layer and 8-layer 1D CNN architectures provided higher accuracies than the random forest by 1.9% - 2.8% which as all the accuracies were > 83% is a meaningful increase. The CONUS overall classification accuracies increased marginally with the number of percentiles (86.21%, 86.40%, and 86.43% for 5, 7 and 9 percentiles, respectively) using the 8-layer 1D-CNN. Class specific producer and user accuracies were quantified, with lower accuracies for the developed land, crop and pasture/hay classes, but no systematic pattern among classes with respect to the number of temporal percentiles used. Application of the trained model to a different year of CONUS Landsat ARD showed moderately decreased accuracy (80.79% for 7 percentiles) that we illustrate is likely due to different intra-annual surface variations between years. These encouraging results are discussed with recommended research for deep learning using temporal metric percentiles.

Land cover is all types of features that exist on the earth's surface on certain land, either artificial or natural. Information related to monitoring and processing satellite image data to obtain land cover classification can be done in various ways, one of which is machine learning methods. This study aims to apply machine learning methods in monitoring land cover using Landsat-8 imagery, to obtain a technique that has high accuracy and is suitable for monitoring land cover. This study uses machine learning methods, namely Support Vector Machine (SVM) and Random Forest (RF). The classification of land cover in this study consists of five classes, namely, built-up areas, water bodies, vacant land, agriculture, and vegetation, where the determination of this land cover class is based on the type of land cover that exists on the RTRW Map of Mataram City in 2011-2031. match the image used. This study shows that the method with the best accuracy is the Support Vector Machine (SVM) method with overall accuracy and kappa accuracy values of 0.9101 and 0.8748. However, there is a misclassification caused by several factors such as the reflectance value of each pixel which is almost the same, the cropping period, and other factors. These factors need to be considered because they affect the land cover classification results.

Land Use and Land Cover (LULC) classification plays a pivotal role in understanding and managing environmental resources. This study presents a novel methodology utilizing sentinel satellite data in conjunction with two robust machine learning algorithms: Random Forest (RF) and Support Vector machine (SVM) on Google Earth Engine platform. Sentinel data, renowned for its high- resolution multispectral imagery, provides rich information for classification. Google Earth Engine provides access to vast geospatial datasets and computational resource, enabling effective analysis. RF and SVM, distinguished for their ability to handle complex dataset, are employed to optimize classification accuracy. A systematic work flow for preprocessing of Sentinel imagery is outlined, followed by implementation of RF and SVM algorithm with a focus on accurately classifying vegetation, built-up areas, barren land and water bodies. Evaluation metrics including overall accuracy and kappa coefficient demonstrate the efficacy of the proposed methodology. A compelling study demonstrates the utility of RF and SVM within GEE for precise LULC mapping, highlighting their pivotal role in supporting informed decision-making for environmental planning and conservation initiatives. Keywords GEE, SVM, RF, Sentinel Data, LULC

AN ASSESSMENT OF SPATIAL VARIATION OF LAND SURFACE CHARACTERISTICS OF MINNA, NIGER STATE NIGERIA FOR SUSTAINABLE URBANIZATION USING GEOSPATIAL TECHNIQUES

Landsat-like moderate resolution remote sensing images are widely used in land use and land cover (LULC) classification. Limited by coarser resolutions, most of the traditional LULC classifications that are based on moderate resolution remote sensing images focus on the spectral features of a single pixel. Inspired by the spatial evaluation methods in landscape ecology, this study proposed a new method to extract neighborhood characteristics around a pixel for moderate resolution images. 3 landscape-metric-like indexes, i.e., mean index, standard deviation index, and distance weighted value index, were defined as adjacent region features to include the surrounding environmental characteristics. The effects of the adjacent region features and the different feature set configurations on improving the LULC classification were evaluated by a series of well-controlled LULC classification experiments using K nearest neighbor (KNN) and support vector machine (SVM) classifiers on a Landsat 8 Operational Land Imager (OLI) image. When the adjacent region features were added, the overall accuracies of both the classifiers were higher than when only spectral features were used. For the KNN and SVM classifiers that used only spectral features, the overall accuracies of the LULC classification were 85.45% and 88.87%, respectively, and the accuracies were improved to 94.52% and 96.97%. The classification accuracies of all the LULC types improved. Highly heterogeneous LULC types that are easily misclassified achieved greater improvements. As comparisons, the grey-level co-occurrence matrix (GLCM) and convolutional neural network (CNN) approaches were also implemented on the same dataset. The results revealed that the new method outperformed GLCM and CNN approaches and can significantly improve the classification performance that is based on moderate resolution data.

Nowadays, Satellite images have become a major source of data for many aspects of development. Land and crops classification using satellite images is a recent important subject. From the other side, Deep Convolutional Neural Networks (DCNNs) is a powerful technique for understanding images. This paper describes a pixel based crops and land cover classification originating from one source satellite imagery represented by Sentinel satellite and based on several dates for the same agricultural season. We propose a DCNN architecture based on multi-temporal data that was fed to a one-dimension (1-D DCNN). The proposed architecture is compared with other methods of satellite image classification algorithms; such as Support Vector Machines (SVMs), Random Forests (RFs) and k-Nearest Neighbors (k-NNs). Experiments are conducted for the mutual experiment of major crops and land cover classification for Al-Fayoum governorate in Egypt. The 1-D DCNN achieves about 89% accuracy using 10 spectral bands from Sentinel-2 satellite imagery database for the area of interest. The proposed architecture although it outperforms other methods, needs further research to optimize the memory usage.

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.

In recent years, the classification of land change has revolutionized the ability to monitor and understand dynamic changes occurring on the Earth’s surface. Artificial intelligence (AI) techniques must improve the performance and accuracy of land change detection by extracting spectral features from several Convolutional Neural Networks (CNNs) and integrating them. In this study, AI techniques were applied to classify the land use and land cover (LULC) of the Najran city map in Saudi Arabia based on 2020 Landsat 8 satellite imagery. This was achieved using several hybrid models combining CNN and random forest (RF) models, namely AlexNet-RF and GoogLeNet-RF, as well as the combined spectral features of AlexNet-GoogLeNet with RF. The results showed that LULC classification using a hybrid system was superior to CNN and proved that the proposed hybrid system of combined spectral features extracted from AlexNet-GoogLeNet with RF provided better results than using the hybrid system proposed by AlexNet with RF and GoogLeNet with RF. The proposed hybrid system of combined spectral features extracted from AlexNet-GoogLeNet with RF achieved an accuracy of 96.95%, a Kappa coefficient of 0.9638, sensitivity of 96.95%, AUC of 98.4%, and specificity of 99.83%. The proposed hybrid methods aim to enhance the classification accuracy and increase the robustness of the system, ensuring consistent performance across diverse earth-change scenarios. It substantially impacts various domains, including environmental monitoring, disaster management, and sustainable urban planning.

Accurate Land Use and Land Cover (LU/LC) classification is essential for sustainable resource management, urban development, and environmental conservation. The integration of remote sensing data with supervised machine learning algorithms has significantly enhanced classification accuracy and efficiency. This study evaluates the performance of five widely used supervised learning algorithms namely 1) Classification and Regression Tree (CART), 2) Gradient Boost Tree (GB), 3) K-Nearest Neighbours (KNN), 4) Support Vector Machine (SVM) and 5) Random Forest (RF) for LU/LC mapping in study area of East Godavari District, Andhra Pradesh, India over a time period of 2 years between 2023 and 2025. High-resolution Landsat-8 imagery is processed and classified using above algorithms, with model performance assessed based on overall accuracy, Kappa coefficient, precision and F1-score. The findings indicated that Gradient Tree Boost demonstrated superior performance compared to the other classifiers, attaining the highest accuracy of 98.26% along with a Kappa coefficient of 0.9761. Random Forest closely followed, achieving an accuracy of 97.39% and a Kappa value of 0.9642. Additionally, both SVM and KNN exhibited strong classification capabilities, with respective accuracies of 96.52% and Kappa values of 0.9522, highlighting their effectiveness in land cover classification applications. The study also examines the computational efficiency and reliability of each classifier, offering insights into their suitability for LU/LC analysis in diverse landscapes. The findings contribute to the optimization of machine learning techniques for remote sensing applications, aiding in data-driven decision-making for land management. Future research can explore deep learning-based classification models and multi-temporal analysis to further enhance LU/LC mapping accuracy.

Emotion recognition plays a pivotal role in advancing Human-Computer Interaction (HCI) by enabling systems to understand and respond to human emotional states. Among various modalities, Speech Emotion Recognition (SER) stands out as a non-invasive, cost-effective, and temporally efficient approach for detecting emotions. This study leverages the RAVDESS dataset, which provides high-quality audio recordings for emotion classification. The proposed methodology involves preprocessing audio signals to remove noise, extracting temporal and spectral features in both time and frequency domains, and implementing machine learning models for multi-class emotion classification. The study evaluates and compares the performance of several classification models, including Random Forest (RF), Multilayer Perceptron (MLP), Support Vector Machine (SVM), Convolutional Neural Networks (CNN), and Decision Trees (DT). Experimental results demonstrate promising accuracy levels, highlighting the potential of these machine learning techniques in SER. This research contributes to the integration of SER within Brain-Computer Interfaces (BCI) and other emotion-aware applications, paving the way for enhanced interactive systems. Keywords – Speech Emotion Recognition -Machine Learning - Random Forest -Multilayer Perceptron - Support Vector Machine - Convolutional Neural Networks - Decision Tree - RAVDESS Dataset

Remote sensing of environmental processes and properties using Landsat imagery has evolved since its inception in 1972. Monitoring of land cover and its changes over space and time using classification algorithms is one of the most important uses of remote sensing. However, the reliability of the land cover products from remotely sensed data is dependent upon the accuracy of different classification parameters. In this study, we have applied and tested two land cover classification algorithms: support vector machine (SVM) and maximum likelihood (ML) for land cover classification of the Kathmandu Valley, Nepal, between 1988 and 2016. The results show that SVM has better classification accuracies compared to ML.

Accurate classification of land cover indices is important for diverse disciplines (e.g., ecology, geography, and climatology) because it serves as a basis for various real world applications. For detection and classification of land cover, remote sensing has long been used as an excellent source of data for finding different types of data attribute present in the land cover. A variety of feature extraction and classification methods in machine learning have been used to classify land cover using satellite images. In recent years, deep learning have recently emerged as a dominant paradigm for machine learning in a variety of domains. The objective of this paper presents the multi-labeled land cover indices classification using Google Earth Satellite images with deep convolutional neural network (DCNN). Since the lack of massive labeled land cover dataset, the own created labeled dataset for Ayeyarwaddy Delta is applied and tested with AlexNet. Then the results of land cover classification are compared with Multiclass-SVM using confusion matrices. According to the tested results, 76.6% of building index, 81.5% road index, 91.8% of vegetation index and 93.2% of water index can be correctly classified by using DCNN. The confusion matrix for Multiclass-SVM, 78.9% of building index, 72.7% road index, 94.2% of vegetation index and 98.1% of water index can be correctly classified.

The ability to rapidly produce accurate land use and land cover maps regularly and consistently has been a growing initiative as they have increasingly become an important tool in the efforts to evaluate, monitor, and conserve Earth’s natural resources. Algorithms for supervised classification of satellite images constitute a necessary tool for the building of these maps and they have made it possible to establish remote sensing as the most reliable means of map generation. In this paper, we compare three machine learning techniques: Random Forest, Support Vector Machines, and Light Gradient Boosted Machine, using a 70/30 training/testing evaluation model. Our research evaluates the accuracy of Light Gradient Boosted Machine models against the more classic and trusted Random Forest and Support Vector Machines when it comes to classifying land use and land cover over large geographic areas. We found that the Light Gradient Booted model is marginally more accurate with a 0.01 and 0.059 increase in the overall accuracy compared to Support Vector and Random Forests, respectively, but also performed around 25% quicker on average.

The coronavirus COVID-19 pandemic has caused a global health crisis. According to According to the World Health Assembly, one of the best preventative measures is to wear a face mask while out outdoors (WHO). This work presents a hybrid model for face mask identification that combines deep and traditional machine learning. I have trained the proposed system, which consists of convolutional neural networks (ConNN), support vector machines (SVM), and random forests (RF), in three stages, the first stage, used ConNN, the second stage, used the same ConNN with the SVM method, and in the third stage, used ConNN and RF. This paper suggests three different kinds of masked face recognition datasets: the Incorrectly Masked Face Dataset (IMFD), the Correctly Masked Face Dataset (CMFD), and the combination for MaskedFace-Net, a worldwide masked face detection system. Two objectives are presented for the realistic masked face datasets: i) to identify individuals whose faces are covered or not covered, ii) to identify faces whose masks are put on properly or improperly (for example, at airport entrances or among crowds). The suggested model is made up of two parts. The first part is designed for feature extraction using a convolutional neural networks. In contrast, the second section is made to classify face masks using SVM and RF methods. The ConNN achieved 99.92%. and achieved for ConNN and SVM 99.94%. ConNN and RF 98.79%. Moreover,The system has been tested in real world scenarios and can recognize and classify any image selected by Google with high accuracy. we a comparison and the results aim to evaluate the proposed model.