Atmospheric correction is one of the key parts of remote sensing preprocessing because it can influence and change the final classification result. This research examines the impact of five different atmospheric correction processing on land cover classification accuracy using Sentinel-2 satellite imagery. Those are surface reflectance (SREF), standardized surface reflectance (STDSREF), Sentinel-2 atmospheric correction (S2AC), image correction for atmospheric effects (iCOR), dark object subtraction (DOS) and top of the atmosphere (TOA) reflectance without any atmospheric correction. Sentinel-2 images corrected with stated atmospheric corrections were classified using four different machine learning classification techniques namely extreme gradient boosting (XGB), random forests (RF), support vector machine (SVM) and catboost (CB). For classification, five different classes were used: bare land, low vegetation, high vegetation, water and built-up area. SVM classification provided the best overall result for twelve dates, for all atmospheric corrections. It was the best method for both cases: when using Sentinel-2 bands and radiometric indices and when using just spectral bands. The best atmospheric correction for classification with SVM using radiometric indices is S2AC with the median value of 96.54% and the best correction without radiometric indices is STDSREF with the median value of 96.83%.

Atmospheric correction is used to remove atmospheric effects on remotely sensed imagery, so it could change its properties and, in the end, all the derived results from it. The objective of this research was to analyse the effect of different atmospheric corrections on radiometric indices using Landsat 8 and Sentinel-2 satellite. Radiometric indices used in this research are Normalized difference vegetation index (NDVI), Normalized difference water index (NDWI), Soil-adjusted vegetation index (SAVI) and Modified soil-adjusted vegetation index (MSAVI). The analysis was done using five pairs of contemporary images from Landsat 8 and Sentinel-2. Based on those images, our results indicate that NDVI values between those two datasets did not have a statistically significant difference (p >0.05) and have the strongest correlation for all used atmospheric corrections (r1 = 0.989, r2 = 0.967, r3 = 0.983). We concluded that it is possible to combine radiometric indices values from Landsat 8 and Sentinel-2 data.

Satellite image fusion techniques have been used for more than two decades. Development of satellite sensors for very high-resolution satellite imagery monitoring contributed to the development of new image fusion techniques. We examine the quality of Ehlers, Brovey transform, modified intensity–hue–saturation (M-IHS), and high-filter resolution merge fusion methods on vegetation indices values combining high-resolution WorldView-4 and low-resolution Sentinel-2 imagery. For image fusion, four bands—blue, green, red, and near-infrared—were used. The effect of fusing Sentinel-2 and WorldView-4 imagery is tested on the various vegetation indices [normalized difference vegetation index (NDVI), blue normalized difference vegetation index (BNDVI), and green normalized difference vegetation index (GNDVI)]. M-IHS fusion showed the best result, with the least difference between indices calculated based on the original Sentinel-2 and fused bands. Difference between vegetation indices calculated using original Sentinel-2 and M-IHS fused bands for 30.08.2015, 30.09.2016, and 30.09.2017 is between −0.041 to 0.037. The second part of this research is an evaluation of three different vegetation classes (grassland, mixed vegetation, and forest) on three different years and three different months within a year. Difference between NDVI, GNDVI, and BNDVI calculated based on the original Sentinel-2, and M-IHS fused bands are ranging from −0.043 to 0.103, −0.041 to 0.050, and −0.052 to 0.030 for grassland, mixed vegetation, and forest, respectively. On the other hand, the difference between vegetation indices for grassland, mixed vegetation, and forest are from −0.110 to 0.050, −0.041 to 0.038, and −0.052 to 0.038, when observing August, September, and October of 2017, respectively.

Image fusion methods aim at fusing low resolution and high-resolution image to obtain a new image that provides new information for the specific application. The main goal of this article is multitemporal Sentinel-2 image fusion using single WorldView-4 satellite image for urban area monitoring. Fusing those images should provide Sentinel-2 image with similar radiometric band value as original Sentinel-2 image, but with a spatial resolution of WorldView-4. Ehlers, Brovey Transform, Modified Intensity-Hue-Saturation, High-Pass Filtering, Hyperspherical Colour Space and Wavelet resolution merge fusion techniques were used for spatial enhancement of Sentinel-2 images. Original and fused images were first compared using standard statistical parameters, mean, median and standard deviation. Image quality analysis was conducted with different objective image quality measures like root mean square error, peak signal to noise ratio, universal image quality index, structural similarity index, relative dimensionless global error, spatial correlation coefficient, relative average spectral error, spectral angle mapper, multi-scale structural similarity index. Using these quality measures helped in determining the spectral and spatial preservation of fused images. Hyperspherical colour space method was selected as the best method for image fusion of Sentinel-2 and WorldView-4 image-based on standard statistical parameters and quality measures.

Accurate delineation of fields is difficult in fragmented landscapes where single-date images provide no seasonal cues and supervised models require labels. We propose a method that explicitly represents phenology to improve zero-shot delineation. Using 22 cloud-free PlanetScope scenes over a 5 × 5 km area, a single harmonic model is fitted to the NDVI per pixel to obtain the phase, amplitude and mean. These values are then mapped into cylindrical colour spaces (Hue–Saturation–Value, Hue–Whiteness–Blackness, Luminance-Chroma-Hue). The resulting recoloured composites are segmented using the Segment Anything Model (SAM), without fine-tuning. The results are evaluated object-wise, object-wise grouped by area size, and pixel-wise. Pixel-wise evaluation achieved up to F1 = 0.898, and a mean Intersection-over-Union (mIoU) of 0.815, while object-wise performance reached F1 = 0.610. HSV achieved the strongest area match, while HWB produced the fewest fragments. The ordinal time-of-day basis provided better parcel separability than the annual radian adjustment. The main errors were over-segmentation and fragmentation. As the parcel size increased, the IoU increased, but the precision decreased. It is concluded that recolouring using harmonic NDVI time series is a simple, scalable, and interpretable basis for field delineation that can be easily improved.