Towards reducing chemical usage for weed control in agriculture using UAS imagery analysis and computer vision...

Area of Science:

• Agricultural Engineering
• Precision Agriculture
• Computer Vision

Background:

• Current weed management in agriculture often involves uniform herbicide application, leading to overuse.
• Site-specific weed control (SSWC) offers a more targeted approach to reduce chemical usage.
• Unmanned Aerial Systems (UAS) provide valuable spatial data for precision agriculture applications.

Purpose of the Study:

• To implement site-specific weed control (SSWC) in a corn field using UAS imagery.
• To develop and apply a computer vision algorithm for identifying crop rows and weeds.
• To evaluate the effectiveness of SSWC in reducing herbicide application compared to conventional methods.

Main Methods:

• Utilized UAS to capture high-resolution imagery of the corn field.
• Developed a 'Crop Row Identification' algorithm to distinguish corn rows from weeds using computer vision.
• Generated a grid-based weed prescription map and executed targeted herbicide spraying with a commercial sprayer.

Main Results:

• The SSWC approach successfully identified and mapped weed distribution within the corn field.
• Herbicide application was significantly reduced, with 26.2% of the acreage saved from spraying.
• The method demonstrated potential for chemical usage reduction even in high weed infestation scenarios.

Conclusions:

• Site-specific weed control using UAS and computer vision is a viable strategy for optimizing herbicide application in corn production.
• The developed workflow provides a practical method for implementing SSWC from data acquisition to field application.
• This approach offers a promising opportunity to reduce environmental impact and input costs in modern agriculture.