Search research articles

Search research articles

Home
Image-based Obstacle Detection Methods For The Safe Navigation Of Industrial Unmanned Aerial Vehicles.

Home
Image-based Obstacle Detection Methods For The Safe Navigation Of Industrial Unmanned Aerial Vehicles.

Related Concept Videos

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Experiment Video

Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation

Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation

Published on: February 9, 2024

Image-based obstacle detection methods for the safe navigation of industrial unmanned aerial vehicles.

Liang Wang¹, Yin Shen¹, Haichao Li¹

¹Hangzhou Aolida Elevator Co., Ltd., Hangzhou, 311600, P. R. China.

Scientific Reports

|October 15, 2025

View abstract on PubMed

Summary

This summary is machine-generated.

This study introduces a new model for unmanned aerial vehicles (UAVs) to improve obstacle detection. The Texture-variant Obstacle Object Classification Model (TOOCM) enhances accuracy by adapting feature extraction to changing object textures.

Keywords:

Object detection Obstacle classification ResNet Texture variant Unmanned aerial vehicles

More Related Videos

Collecting and Processing Drone-based Remotely Sensed Data for Use in Forest Recovery Monitoring

Collecting and Processing Drone-based Remotely Sensed Data for Use in Forest Recovery Monitoring

Published on: October 24, 2025

Related Experiment Videos

Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation

Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation

Published on: February 9, 2024

Collecting and Processing Drone-based Remotely Sensed Data for Use in Forest Recovery Monitoring

Collecting and Processing Drone-based Remotely Sensed Data for Use in Forest Recovery Monitoring

Published on: October 24, 2025

Area of Science:

Robotics and Automation
Computer Vision
Artificial Intelligence

Background:

Industrial unmanned aerial vehicles (UAVs) require real-time object and obstacle recognition for autonomous navigation.
Variable object textures often cause feature disappearance, reducing detection accuracy.

Purpose of the Study:

To propose a novel Texture-variant Obstacle Object Classification Model (TOOCM) for enhanced feature extraction and dynamic texture representation.
To improve the accuracy of obstacle detection and object classification in UAVs by capturing small texture fluctuations.

Main Methods:

Developed the TOOCM model using an N-layer ResNet architecture with dynamic layer augmentation.
Reconstructed convolutional layers adaptively based on texture concentration and object size variations.
Implemented layer-wise learning updates during UAV navigation for continuous performance optimization.

Main Results:

Achieved a 14.09% increase in detection accuracy.
Improved precision by 14.53%.
Reduced loss by 14.17%, especially in high-density obstacle environments.

Conclusions:

The proposed adaptive feature learning approach effectively enhances obstacle detection and classification accuracy for UAVs.
TOOCM demonstrates superior performance in complex environments compared to traditional fixed-layer deep models.