Abstract
Deep learning-based marker detection for autonomous drone landing is widely studied, due to its superior detection performance. However, no study was reported to address non-uniform motion-blurred input images, and most of the previous handcrafted and deep learning-based methods failed to operate with these challenging inputs. To solve this problem, we propose a deep learning-based marker detection method for autonomous drone landing, by (1) introducing a two-phase framework of deblurring and object detection, by adopting a slimmed version of deblur generative adversarial network (DeblurGAN) model and a You only look once version 2 (YOLOv2) detector, respectively, and (2) considering the balance between the processing time and accuracy of the system. To this end, we propose a channel-pruning framework for slimming the DeblurGAN model called SlimDeblurGAN, without significant accuracy degradation. The experimental results on the two datasets showed that our proposed method exhibited higher performance and greater robustness than the previous methods, in both deburring and marker detection.
Highlights
Unmanned aerial vehicles (UAVs) or drones are successfully used in several industries
In the 1st fold validation, half of the images in the SMBD-DB1 were used for training, and the other half was for testing
In an attempt to facilitate the training process of the marker detection convolutional neural networks (CNN), we considered the distribution of object bounding boxes in the training dataset, in order to generate a set of anchor boxes used by You only look once version 2 (YOLOv2)
Summary
Unmanned aerial vehicles (UAVs) or drones are successfully used in several industries They have a wide range of applications such as surveillance, aerial photography, infrastructural inspection, and rescue operations. These applications require that the onboard system can sense the environment, parse, and react according to the parsing results. Scene parsing is a function that enables the system to understand the visual environment, such as recognizing the type of objects, place of objects, and regions of object instances in a scene. These problems are the main topics in computer vision—classification, object detection, and object segmentation. Deploying deep learning models to a UAV onboard system raises new challenges—(1) difficulties of scene parsing in cases of low-resolution or motion-blurred input, (2) difficulties of deploying the model to an embedded system with limited memory and computation power, and (3) balancing between model accuracy and execution time
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
