Bias-Compensation Augmentation Learning for Semantic Segmentation in UAV Networks

• Published in: IEEE internet of things journal 2024-06, Vol.11 (12), p.21261-21273
• Main Authors: Yu, Tiankuo, Yang, Hui, Nie, Jiali, Yao, Qiuyan, Liu, Wenxin, Zhang, Jie, Cheriet, Mohamed
• Format: Article
• Language: English
• Subjects: Accuracy; Adaptation models; Autonomous aerial vehicles; Bias; Bias compensation augmentation learning; Compensation; Computation offloading; Computational modeling; computing power resource scheduling; Data collection; Disaster relief; Disasters; Image reconstruction; Image segmentation; Learning; Neural networks; Processor scheduling; Rescue operations; Search and rescue missions; Seismic activity; Semantic segmentation; Semantics; Task analysis; unmanned aerial vehicle (UAV) networks; Unmanned aerial vehicles
• ISSN: 2327-4662; 2327-4662
• DOI: 10.1109/JIOT.2024.3373454

In the realm of emergency disaster relief, it is paramount to attain a thorough comprehension of the semantic information associated with the local disaster scene for strategic rescue path planning and immediate rescue operations for affected individuals. Unmanned aerial vehicle (UAV) networks are widely utilized for rapid data collection in the aftermath of disasters due to their flexibility and maneuverability, assisting in rescue decision-making. However, some disasters, such as seismic events and floods, have disrupted the initially structured ground shape information, leading to a disparate distribution of data collected by various UAV groups. This exposes traditional semantic segmentation models susceptible to shortcut bias, posing challenges in adapting to semantic segmentation tasks in disaster scenarios. Thus, this article proposes a bias-compensation augmentation learning-based semantic segmentation framework, which substantially enhances the extraction capability of semantic information. Initially, we exploit an artificial augmentation neural network for bias-awareness to determine the relative bias values of the collected image data. Subsequently, considering the limited computing power resources in UAV networks, we present a bias compensation computation offloading strategy to achieve a relatively balanced distribution of semantic information across UAV nodes, optimizing the tradeoff between network scheduling efficiency and model accuracy. We conduct reconstruction validation on the FloodNet data set, and a plethora of experimental results demonstrate that, compared to traditional methods, this approach greatly improves the accuracy of pixel-level semantic segmentation by over 86.5%. Moreover, the average combined processing time is also reduced by over 50%, enhancing the utilization efficiency of limited computational resources.