Abstract:Land-area classification (LAC) research offers a promising avenue to address the intricacies of urban planning, agricultural zoning, and environmental monitoring, with a specific focus on urban areas and their complex land usage patterns. The potential of LAC research is significantly propelled by advancements in high-resolution satellite imagery and machine learning strategies, particularly the use of convolutional neural networks (CNNs). Accurate LAC is paramount for informed urban development and effective land management. Traditional remote-sensing methods encounter limitations in precisely classifying dynamic and complex urban land areas. Therefore, in this study, we investigated the application of transfer learning with Inception-v3 and DenseNet121 architectures to establish a reliable LAC system for identifying urban land use classes. Leveraging transfer learning with these models provided distinct advantages, as it allows the LAC system to benefit from pre-trained features on large datasets, enhancing model generalization and performance compared to starting from scratch. Transfer learning also facilitates the effective utilization of limited labeled data for fine-tuning, making it a valuable strategy for optimizing model accuracy in complex urban land classification tasks. Moreover, we strategically employ fine-tuned versions of Inception-v3 and DenseNet121 networks, emphasizing the transformative impact of these architectures. The fine-tuning process enables the model to leverage pre-existing knowledge from extensive datasets, enhancing its adaptability to the intricacies of LC classification. By aligning with these advanced techniques, our research not only contributes to the evolution of remote-sensing methodologies but also underscores the paramount importance of incorporating cutting-edge methodologies, such as fine-tuning and the use of specific network architectures, in the continual enhancement of LC classification systems. Through experiments conducted on the UC-Merced_LandUse dataset, we demonstrate the effectiveness of our approach, achieving remarkable results, including 92% accuracy, 93% recall, 92% precision, and a 92% F1-score. Moreover, employing heatmap analysis further elucidates the decision-making process of the models, providing insights into the classification mechanism. The successful application of CNNs in LAC, coupled with heatmap analysis, opens promising avenues for enhanced urban planning, agricultural zoning, and environmental monitoring through more accurate and automated land-area classification.

Analyzing Land Shape Typologies in South Korean Apartment Complexes Using Machine Learning and Deep Learning Techniques

Classification of land lot shapes in real estate sector using a convolutional neural network

Identifying urban form typologies in Seoul using a new Gaussian mixture model-based clustering framework

The detection of residential developments in urban areas: Exploring the potentials of deep-learning algorithms

Recognition and Classification of Typical Building Shapes Based on YOLO Object Detection Models

Urban Land Use Classification Using LiDAR Geometric, Spatial Autocorrelation and Lacunarity Features Combined with Postclassification Processing Method

An automatic approach for tree species detection and profile estimation of urban street trees using deep learning and Google street view images

Knowledge-guided land pattern depiction for urban land use mapping: A case study of Chinese cities

Detection of infill development and contributing factors using deep learning and multilevel modeling

Land-Cover Classification Using Deep Learning with High-Resolution Remote-Sensing Imagery

Developing a two-level machine-learning approach for classifying urban form for an East Asian mega-city

Classification of residential and non-residential buildings based on satellite data using deep learning

Automatic Building Roof Plane Extraction in Urban Environments for 3D City Modelling Using Remote Sensing Data

Developing High Quality Training Samples for Deep Learning Based Local Climate Zone Classification in Korea

Lightweight and Robust Representation of Economic Scales from Satellite Imagery

Boundary Regularized Building Footprint Extraction From Satellite Images Using Deep Neural Network

A Novel Active-Learning Based Residential Area Segmentation Algorithm

A Structured Approach to the Analysis of Remote Sensing Images

Quantitative land price analysis via computer vision from street view images

Road Mapping in Low Data Environments with OpenStreetMap

UrbanClassifier: A Deep Learning-Based Model for Automated Typology and Temporal Analysis of Urban Fabric Across Multiple Spatial Scales and Viewpoints