Use of UAV-Borne Spectrometer for Land Cover Classification

Abstract

Unmanned aerial vehicles (UAV) are being used for low altitude remote sensing for thematic land classification using visible light and multi-spectral sensors. The objective of this work was to investigate the use of UAV equipped with a compact spectrometer for land cover classification. The UAV platform used was a DJI Flamewheel F550 hexacopter equipped with GPS and Inertial Measurement Unit (IMU) navigation sensors, and a Raspberry Pi processor and camera module. The spectrometer used was the FLAME-NIR, a near-infrared spectrometer for hyperspectral measurements. RGB images and spectrometer data were captured simultaneously. As spectrometer data do not provide continuous terrain coverage, the locations of their ground elliptical footprints were determined from the bundle adjustment solution of the captured images. For each of the spectrometer ground ellipses, the land cover signature at the footprint location was determined to enable the characterization, identification, and classification of land cover elements. To attain a continuous land cover classification map, spatial interpolation was carried out from the irregularly distributed labeled spectrometer points. The accuracy of the classification was assessed using spatial intersection with the object-based image classification performed using the RGB images. Results show that in homogeneous land cover, like water, the accuracy of classification is 78% and in mixed classes, like grass, trees and manmade features, the average accuracy is 50%, thus, indicating the contribution of hyperspectral measurements of low altitude UAV-borne spectrometers to improve land cover classification.

An Incongruence-Based Anomaly Detection Strategy for Analyzing Water Pollution in Images from Remote Sensing

“…Other studies, such as [66,77,78,100], have more significant differences regarding both accuracy and recall in comparison to our study. In these studies, the levels of accuracy and recall were achieved after applying a method based on: (1) seasonal autoregressive integrated moving average (SARIMA) model for autocorrelation analysis in [77].…”

“…The results achieved by this study demonstrate higher accuracy and recall than previous studies' results, such as [25,66,[76][77][78][80][81][82][98][99][100]. Therefore, the results suggest that our strategy provides Outcomes from this study align with those presented by Kittler et al in [13] and Weinshall et al in [15], because they perform research involving the use of incongruences and recognize their potential to be applied to anomaly detection.…”

“…(2) Gaussian process (GP) based non-parametric time-series prediction in [78]. (3) Maximum likelihood classification in [100]. (4) Joint fully convolutional auto-encoders (FC AE) model in [66].…”

“…However, [25] and [82] are very specific case studies. Whereas in this paper and in [66,[76][77][78][80][81][82][98][99][100] the areas of study are heterogeneous, in [25], results were presented only related to an ice shelf in the Antarctic region, which is a homogeneous area of study. According to [100], results tend to be higher for homogeneous areas of study compared to heterogeneous ones.…”

“…Whereas in this paper and in [66,[76][77][78][80][81][82][98][99][100] the areas of study are heterogeneous, in [25], results were presented only related to an ice shelf in the Antarctic region, which is a homogeneous area of study. According to [100], results tend to be higher for homogeneous areas of study compared to heterogeneous ones. Whereas in this paper and in [66,[76][77][78][80][81][82]98,99] the altitude of the remote sensing is high; in [82], the sensor was very close to the investigated outliers when the images were captured.…”