Drone-based Optical and Thermal Videos of Rotor Blades Taken in Normal Wind Turbine Operation

0 ratings - Please login to submit your rating.


Blade damage inspection without stopping the normal operation of wind turbines has significant economic value. This study proposes an AI-based method AQUADA-Seg to segment the images of blades from complex backgrounds by fusing optical and thermal videos taken from normal operating wind turbines. The method follows an encoder-decoder architecture and uses both optical and thermal videos to overcome the challenges associated with field application. A memory is designed between the encoder and decoder to improve the method’s performance by utilizing time history information in the videos to achieve temporal complementarity. The designed memory shares information between optical and thermal modalities to achieve multimodal complementarity. We collected a large-scale dataset, i.e., 100 video pairs and over 55,000 images, of optical-thermal videos of blades in operational wind turbines to train and test the method. Experimental results show that AQUADA-Seg: i) achieves near real-time thermal-optical blade video segmentation and can analyze videos with complex backgrounds in real-world field applications; ii) achieves 0.996 and 0.981 MIoU on optical and thermal videos, respectively, outperforming state-of-the-art methods, particularly in the videos with complex backgrounds. This study provides an essential step towards automated blade damage detection using computer vision without stopping the normal operation of wind turbines.


This is a large-scale drone-based optical-thermal wind turbine blade video dataset. This dataset contains 36 optical-thermal video pairs and over 20,778 images. 


Subject: Request for Access to Wind Turbine Blade Video Dataset for Research

Dear [Recipient's Name],

I am writing to request access to your drone-based optical-thermal wind turbine blade video dataset for my research. My study focuses on advancing blade damage inspection methods for wind turbines, aiming to enhance efficiency and reduce operational costs by enabling inspections without halting turbine operations.

Your dataset, comprising 36 optical-thermal video pairs and over 20,778 images, is crucial for my work. It offers a unique opportunity to test and refine AQUADA-Seg, an AI-based segmentation method I am developing, which promises to significantly improve the accuracy and speed of blade damage detection under real-world conditions.

Access to your dataset will not only support my research but also contribute to the broader field of sustainable energy by improving wind turbine maintenance and reliability.

Thank you for considering my request. I am looking forward to the possibility of utilizing your dataset to make meaningful advancements in wind turbine maintenance technology.

Submitted by Majid Memari on Wed, 02/21/2024 - 13:31