Ashley Napier

About

Ashley Napier has made significant contributions to the field of computer vision and deep learning, with a particular focus on automated image analysis for infrastructure inspection. Her most-cited work, "Image segmentation of underfloor scenes using a mask regions convolutional neural network with two-stage transfer learning" (2020, 35 citations), exemplifies her expertise in applying advanced neural network architectures to real-world engineering challenges. Napier’s research centers on developing robust segmentation models capable of accurately identifying and delineating objects in complex, low-visibility environments—such as underfloor spaces—where traditional methods often fail. By pioneering a two-stage transfer learning approach, she demonstrated how pre-trained models can be effectively adapted to specialized, data-scarce domains, significantly improving detection accuracy and efficiency. This work has direct implications for automated building inspection, maintenance robotics, and structural health monitoring. While her citation count reflects the growing interest in her methodology, Napier’s true impact lies in bridging the gap between cutting-edge AI techniques and practical industrial applications, offering a scalable solution for non-destructive evaluation. Her research continues to inspire further exploration into domain-adapted deep learning for challenging visual tasks.

Research Focus

Key Achievements

Top Papers

1. 1
   Image segmentation of underfloor scenes using a mask regions convolutional neural network with two-stage transfer learning

   35 citations · 2020

Key Collaborators

Contact & Links