Next Article in Journal
Coastal Visibility Distance Estimation Using Dark Channel Prior and Distance Map Under Sea-Fog: Korean Peninsula Case
Previous Article in Journal
A Decade of Ground Deformation in Kunming (China) Revealed by Multi-Temporal Synthetic Aperture Radar Interferometry (InSAR) Technique
Open AccessArticle

Research on Method of Farmland Obstacle Boundary Extraction in UAV Remote Sensing Images

1
College of Biosystems Engineering and Food Science, Zhejiang University, Hangzhou 310058, China
2
Key Laboratory of Agricultural Internet of Things, Ministry of Agriculture Rural Affairs, Shaanxi 712100, China
*
Author to whom correspondence should be addressed.
Sensors 2019, 19(20), 4431; https://doi.org/10.3390/s19204431
Received: 19 September 2019 / Revised: 6 October 2019 / Accepted: 11 October 2019 / Published: 12 October 2019
(This article belongs to the Section Remote Sensors, Control, and Telemetry)
Aimed at the problem of obstacle detection in farmland, the research proposed to adopt the method of farmland information acquisition based on unmanned aerial vehicle landmark image, and improved the method of extracting obstacle boundary based on standard correlation coefficient template matching and assessed the influence of different image resolutions on the precision of obstacle extraction. Analyzing the RGB image of farmland acquired by unmanned aerial vehicle remote sensing technology, this research got the following results. Firstly, we applied a method automatically registering coordinates, and the average deviations on the X and Y direction were 4.6 cm and 12.0 cm respectively, while the average deviations manually by ArcGIS were 4.6 cm and 5.7 cm. Secondly, with an improvement on the step of the traditional correlation coefficient template matching, we reduced the time of template matching from 12.2 s to 4.6 s. The average deviation between edge length of obstacles calculated by corner points extracted by the algorithm and that by actual measurement was 4.0 cm. Lastly, by compressing the original image on a different ratio, when the pixel reached 735 × 2174 (the image resolution reached 6 cm), the obstacle boundary was extracted based on correlation coefficient template matching, the average deviations of boundary points I of six obstacles on the X and Y were respectively 0.87 and 0.95 cm, and the whole process of detection took about 3.1 s. To sum up, it can be concluded that the algorithm of automatically registered coordinates and of automatically extracted obstacle boundary, which were designed in this research, can be applied to the establishment of a basic information collection system for navigation in future study. The best image pixel of obstacle boundary detection proposed after integrating the detection precision and detection time can be the theoretical basis for deciding the unmanned aerial vehicle remote sensing image resolution.
Keywords: UAV remote sensing; coordinate registration; template matching; boundary extraction UAV remote sensing; coordinate registration; template matching; boundary extraction
MDPI and ACS Style

Fang, H.; Chen, H.; Jiang, H.; Wang, Y.; Liu, Y.; Liu, F.; He, Y. Research on Method of Farmland Obstacle Boundary Extraction in UAV Remote Sensing Images. Sensors 2019, 19, 4431.

Show more citation formats Show less citations formats
Note that from the first issue of 2016, MDPI journals use article numbers instead of page numbers. See further details here.

Article Access Map by Country/Region

1
Back to TopTop