Land Cover Mapping from Colorized CORONA Archived Greyscale Satellite Data and Feature Extraction Classification

Abstract

Land cover mapping is often performed via satellite or aerial multispectral/hyperspectral datasets. This paper explores new potentials for the characterisation of land cover from archive greyscale satellite sources by using classification analysis of colourised images. In particular, a CORONA satellite image over Larnaca city in Cyprus was used for this study. The DeOldify Deep learning method embedded in the MyHeritage platform was initially applied to colourise the CORONA image. The new image was then compared against the original greyscale image across various quality metric methods. Then, the geometric correction of the CORONA coloured image was performed using common ground control points taken for aerial images. Later a segmentation process of the image was completed, while segments were selected and characterised for training purposes during the classification process. The latest was performed using the support vector machine (SVM) classifier. Five main land cover classes were selected: land, water, salt lake, vegetation, and urban areas. The overall results of the classification process were then evaluated. The results were very promising (>85 classification accuracy, 0.91 kappa coefficient). The outcomes show that this method can be implemented in any archive greyscale satellite or aerial image to characterise preview landscapes. These results are improved compared to other methods, such as using texture filters.

Practicing Historical Ecology : Methods for the Collection, Analysis, and Integration of Interdisciplinary Historical Data

“…Over 860,000 images were acquired around the world from 1960 to 1972, with a spatial resolution ranging from as low as 2 meters to 100 meters. There are several examples of the use of Corona imagery for archaeological prospection and mapping (Casana 2020) land use mapping (Agapiou 2021), and other applications. These are the oldest available satellite remote sensing systems, and the data have now all been declassified, and are available for download without cost through the USGS portal (https://catalog.data.gov/dataset/corona-satellite-photography).…”

Revolutionizing urban mapping: deep learning and data fusion strategies for accurate building footprint segmentation

“…These computational methods can be based on features like colour, texture, or shape 26 . Various classifiers, such as a support vector machine (SVM), have been used, for instance, for texture-based aerial image segmentation 7 . A random forest (RF) classifier for spectral-based structure segmentation, instead, has been explored when operating on satellite images 8 , 9 .…”

“…Rule-based methods 6 rely on predefined criteria and heuristics, while machine learning methods, such as support vector machines 7 and random forests 8 , 9 , use labelled training data for automatic learning and identification of building structures. Nevertheless, limitations can be encountered when using the above-mentioned methods.…”

Land Cover Classification from Hyperspectral Images via Weighted Spatial–Spectral Joint Kernel Collaborative Representation Classifier

“…The changes in LULC produce a significant impact on environmental factors such as climate, water balance, biodiversity, and terrestrial ecosystems [2]. Highly accurately monitoring and updating land cover information is not only crucial for environmental protection and scientific studies, but also plays an important role in land resource planning and management, landscape pattern analysis, sustainable development, and many others [3][4][5][6]. Remote Sensing (RS) technology, as one of the most important means for earth observation, has been widely used in land cover classification and mapping through spaceborne or airborne visible, multispectral, hyperspectral, and other imaging sensors due to its promising accuracy and high efficiency [7][8][9].…”