Low-cost UAV as a Source of Image Data for Detection of Land Cover Changes

Sedlák, Pavel; Komárková, Jitka; Jech, Jakub; Mašín, Oldřich · 2019 · OpenAlex-citations

DOI: 10.29333/jisem/5894

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This study investigates the efficacy of low-cost unmanned aerial vehicles (UAVs) as a data source for monitoring land cover changes, specifically focusing on the coastal zones of small water bodies. The research is motivated by the increasing importance of small water bodies in the context of water scarcity and the need for frequent monitoring of their shorelines, which are often difficult to access. While remote sensing via satellites or manned aircraft is common for large areas, UAVs offer higher spatial resolution and faster data acquisition for smaller sites. The authors aim to demonstrate that a mid-range consumer drone, specifically the DJI Phantom 3 with a built-in camera, can effectively document landscape changes through conventional digital image processing methods. The experimental design involved a case study of the "Skříň" pond near Pardubice, Czech Republic. Data collection occurred over eight time horizons between February and August 2018, spaced approximately one month apart to capture seasonal variations. The UAV flights were automated using DJI GO software, with altitudes adjusted between 39.6 meters and 61 meters depending on conditions, maintaining 60% front and side overlaps. Due to the use of a standard RGB camera, multispectral indices were not available; instead, the study relied on visual interpretation. Images were processed using Image Composite Editor 2.0 to create mosaics and ArcGIS 10.5.1 for on-screen digitization. The researchers manually delineated four categories within a specific rectangular area of interest: water surface, ice, drawdown zone (periodically flooded areas), and coastal vegetation. The results revealed distinct seasonal patterns in the coastal zone. In February and March, significant ice cover was observed, with March showing the largest ice area and consequently the smallest visible water surface. By early April, ice had melted, leading to the largest recorded water surface area due to increased water levels. Vegetation coverage was minimal in early April but expanded significantly through the summer, reaching its peak in August. Conversely, the drawdown zone fluctuated, appearing minimal in March and expanding in late summer as water levels decreased due to evaporation and low precipitation. The high spatial resolution of the UAV imagery allowed for clear differentiation and accurate measurement of these categories across all time horizons. The study concludes that low-cost UAVs are a viable and flexible tool for monitoring small water bodies, providing high-resolution data that facilitates detailed visual interpretation of landscape changes. Although limited to visible spectral bands, the data successfully captured dynamic changes in water levels, ice cover, and vegetation growth. This approach offers a cost-effective alternative to traditional surveying or satellite imagery for localized environmental monitoring, particularly in areas with accessibility constraints. The findings support the broader adoption of consumer-grade drones in geoinformation technologies for precise, time-sensitive landscape analysis.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• telematics ubi feedback