학술논문

AI-Supported Tree Detection Using Sensor Data Fusion for a Land Surveying Robot
Document Type
Conference
Source
2023 Smart Systems Integration Conference and Exhibition (SSI) Smart Systems Integration Conference and Exhibition (SSI), 2023. :1-5 Mar, 2023
Subject
Bioengineering
Components, Circuits, Devices and Systems
Computing and Processing
Photonics and Electrooptics
Power, Energy and Industry Applications
Robotics and Control Systems
Signal Processing and Analysis
Transportation
Robot vision systems
Data integration
System integration
Object detection
Cameras
Surface structures
Reliability
multispectral camera sensor
AI-based algorithms
tree detection
artificial object
land surveying robot
Language
Abstract
Within the project, Automatic Land Surveying 5.0 a robot was developed that captures the surface structures three-dimensionally. Special focus was placed on the capturing of three-dimensional objects, in particular trees in the landscape. The approach chosen was to develop a 360° panoramic camera and detect the trees with a specially trained AI algorithm. Since only real (living) trees are relevant for landscape surveying, a method had to be developed to reliably distinguish real trees from artificial trees. The object detection of an AI is not sufficient for this objective, because artificial trees can fool AI. Therefore, a multispectral camera sensor was developed that provides additional data to distinguish living trees from artificial trees.