UAV coal-rock recognition and 3D coal seam model dynamic correction technology in open-pit mine

Area of Science:

• Geological Engineering
• Mining Engineering
• Geospatial Technology

Background:

• High-precision 3D coal seam models are vital for open-pit mining but often lack spatial resolution due to data limitations.
• Existing static models struggle to incorporate real-time geological data, hindering accurate mining design and production.
• Challenges in obtaining realistic geological data result in insufficient spatial resolution for current 3D coal seam models.

Purpose of the Study:

• To develop and validate a novel UAV-based coal-rock recognition and dynamic 3D coal seam model correction technology.
• To enhance the accuracy and spatial resolution of 3D coal seam models in open-pit mining environments.
• To enable continuous improvement of coal seam models by integrating real-time geological data during mining operations.

Main Methods:

• Generation of UAV color point clouds for coal-rock recognition.
• Application of improved region-growing and Alpha-shape algorithms for accurate boundary point extraction.
• Dynamic correction of 3D models using spatial interpolation and TIN update algorithms with real-time data integration.

Main Results:

• High accuracy achieved in coal-rock recognition (97.44%) and boundary point extraction (91.85%).
• Significant reduction in the standard deviation of the 3D coal seam model to 0.32 m after correction.
• Average elevation error reduced by 78.76% to 0.34 m compared to the initial model.

Conclusions:

• The proposed UAV-based dynamic correction technology effectively enhances the precision of 3D coal seam models.
• Continuous integration of real-time data and advanced algorithms leads to substantial improvements in model accuracy.
• This technology offers a viable solution for optimizing mining design and production in open-pit coal mines.