Abstract: Developing a universal and effective UAV low-altitude remote sensing image processing method is crucial for precise pesticide application and real-time agricultural monitoring using drones. In the previous article, we discussed the use of 3D remote sensing technology in plant growth monitoring, pest detection, and the current state of research both domestically and internationally. Today, we will focus on how to utilize a low-cost visible light ultra-low-altitude agricultural remote sensing platform to extract and analyze agricultural data, providing a reference for accurate drone spraying and agricultural monitoring.
Advantages of UAV Ultra-Low Altitude Remote Sensing Technology
Currently, there is serious pesticide pollution in farmlands in China. While UAV low-altitude application technology combined with GPS can plan routes for spraying, it often results in uniform spraying without targeting specific areas. In contrast, UAV remote sensing technology can quickly capture high-resolution images and agricultural data from key areas, enabling precise crop management and reducing chemical inputs. This makes it an essential tool in precision agriculture.
Compared to traditional remote sensing methods, multi-spectral UAV imagery is easier to collect, offers higher spatial resolution, and has lower costs. When integrated with UAV spraying systems, it opens new possibilities for precision agriculture, pest analysis, and field monitoring.
Vegetation Index and UAV Low-Altitude Remote Sensing Image Processing
Vegetation indices are widely used in remote sensing to reflect plant health and other key parameters like chlorophyll content, coverage, leaf area index, and biomass. Commonly used indices include NDVI and RVI, which combine visible and near-infrared bands. However, these methods often require expensive equipment, long acquisition times, and suffer from low spatial resolution, making them less suitable for real-time field applications.
Visible light-based vegetation indices such as NGRDI, GLI, RGRI, and ExG have been developed to overcome these limitations. These indices offer a cost-effective way to monitor crops and detect pests and diseases using low-altitude UAVs.
Research Process and Method
We conducted a series of studies on image correction, visible light vegetation index extraction, and crop region identification for ultra-low altitude remote sensing images. Our goal was to develop a low-cost image processing method that supports precision pesticide application and agricultural information extraction.
Step 1: Acquisition of Visible Light Remote Sensing Images
We used a custom-built electric quadcopter equipped with an APM flight controller, OSD display, barometer, and a wide-angle visible light camera. The system allowed real-time altitude monitoring and stable image transmission.
Step 2: Image Correction
To correct geometric distortions, we applied the Zhang plane calibration method. This involved capturing images of a checkerboard pattern from different angles and depths to calculate the internal distortion matrix and correct the test images.
Step 3: Visible Band Vegetation Index Calculation
We calculated several vegetation indices using RGB pixel values. For example:
$$
\text{NGRDI} = \frac{G - R}{G + R}, \quad \text{GLI} = \frac{2G - R - B}{G + R + B}
$$
Results and Analysis
After analyzing the statistical differences between vegetation and non-vegetation regions, we found that GLI and ExG indices showed the best performance in vegetation extraction, achieving over 97% accuracy. In contrast, NGRDI and RGRI indices had overlapping values between vegetation and non-vegetation areas, leading to lower classification accuracy.
For those interested in further study, two additional knowledge points are provided:
1. Why is the image captured by the drone distorted?
UAV images often suffer from geometric distortions due to terrain variations and changes in flight attitude. In our case, the distortion was mainly caused by barrel distortion from the camera lens, as the camera wasn't strictly calibrated.
2. What is the Zhang plane calibration method?
This offline calibration technique involves capturing images of a planar calibration pattern from multiple angles. Using these images, we calculated the camera's internal parameters and corrected the distortion in the test images.
In conclusion, the use of GLI and ExG indices for vegetation extraction from ultra-low altitude UAV images shows great potential. This cost-effective method provides high-precision results and holds significant value for precision agriculture, targeted spraying, and pest monitoring.
[Description] This article is reprinted from the column "Professional Road to Precision Agriculture" by Professor Lan Yubin, a leading figure in China's agricultural aviation industry. Feifei College is a UAV technology and modern agricultural knowledge-sharing platform established by Extreme Flying Technology, offering both online knowledge services and offline training programs. With over 50,000 registered students and more than 5 million article reads, Feifei College continues to promote the concepts of drone technology and precision agriculture among new farmers.
Karaoke Speakers,Karaoke Bluetooth Speaker,Led Karaoke Speaker,Karaoke Speaker For Family
Comcn Electronics Limited , https://www.comencnspeaker.com