AI driven Multi-modal sensing and fusion for autonomous industrial inspection and smart decision-making

This AI driven Multi-modal sensing and fusion for autonomous industrial inspection and smart decision-making project at University of Dundee sets out to fundamentally transform the use of image processing and deep learning technologies in pivotal industrial sectors, including manufacturing, oil & gas, and renewable energy.

This AI driven Multi-modal sensing and fusion for autonomous industrial inspection and smart decision-making AI driven Multi-modal sensing and fusion for autonomous industrial inspection and smart decision-making project at University of Dundee University of Dundee will first delve into the development of advanced algorithms for optimized image data acquisition. This involves not just the capture of images but also their intelligent processing to ensure maximum relevance and quality achieved. The use of deep learning techniques, particularly in the context of image quality assessment, will play a crucial role in the smart interpretation and processing of industrial informatics.

Context

In addition, the project aims to push the boundaries of current image enhancement and analytic techniques. This is particularly relevant for object detection in visually degraded scenes, a common challenge in industrial settings. Advanced deep learning architectures such as convolutional neural network (CNN) and generative adversarial network (GAN) will be used as baseline. Few shot learning and transfer learning strategies will also be introduced to tackle with challenging scenarios characterized by limited data samples and less-than-ideal annotations.

At its core, the research will investigate the synergistic application of multi-modal sensing and fusion technologies, such as 

• depth imaging

• optical imaging

• multibeam sonar

• stereo vision

• multispectral and hyperspectral imaging (HSI)