Ground penetrating radar (GPR) has been widely used as a non-destructive technique to detect subsurface objects. Manual interpretation of GPR data is tedious and time-consuming. To address this challenge, an automatic method based on a deep instance segmentation framework is developed to detect and segment object signatures from GPR scans. The proposed method develops the Mask Scoring R-CNN (MS R-CNN) architecture by introducing a novel anchoring scheme. By analyzing the characteristics of the hyperbolic signatures of subsurface objects in GPR scans, a set of anchor shape ratios are optimized and selected to substitute the predefined and fixed aspect ratios in the MS R-CNN framework to improve the signature detection performance. In addition, transfer learning technique is adopted to obtain a pre-trained model to address the challenge of insufficient GPR dataset for model training. The detected and segmented signatures can then be further processed for target localization and characterization. GPR data of tree roots were collected in the field to validate the proposed methods. Despite the noisy background and varying signatures in the GPR scans, the proposed method demonstrated promising results in object detection and segmentation. Computational results show that the improved MS R-CNN outperforms the other state-of-the-art methods.

Many vision-based indoor localization methods require tedious and comprehensive pre-mapping of built environments. This research proposes a mapping-free approach to estimating indoor camera poses based on a 3D style-transferred building information model (BIM) and photogrammetry technique. To address the cross-domain gap between virtual 3D models and real-life photographs, a CycleGAN model was developed to transform BIM renderings into photorealistic images. A photogrammetry-based algorithm was developed to estimate camera pose using the visual and spatial information extracted from the style-transferred BIM. The experiments demonstrated the efficacy of CycleGAN in bridging the cross-domain gap, which significantly improved performance in terms of image retrieval and feature correspondence detection. With the 3D coordinates retrieved from BIM, the proposed method can achieve near real-time camera pose estimation with an accuracy of 1.38 m and 10.1° in indoor environments.

Rapid reconnaissance of building damage is critical for disaster response and recovery. Drones have been utilized to collect aerial images of affected areas in order to assess building damage. However, there are two challenges. First, processing many aerial images to detect and classify building damage based on a consistent standard remains laborious and complex, necessitating a new automated solution to achieve accurate building damage detection and classification. Second, drone operations during disaster response rely primarily on human operators’ experience and seldom use the obtained building damage information to optimize drone mission planning. Therefore, this study proposes a new method, which automates building damage reconnaissance with drone mission planning for disaster response operations. Specifically, a deep learning method is developed to detect and classify building damages using a newly labeled dataset consisting of 24,496 distinct instances of building damage. This deep learning method is validated, achieving 71.9% mean average precision. In addition, building damage information is modeled and integrated into mission planning, in order to optimize drones’ task assignments and route calculations. A tornado disaster in Tennessee is used as a case study to quantitatively evaluate this methodology. The present study concludes that optimal drone mission planning during disaster response can be augmented using accurate building damage information acquired from deep learning methods.