Artificial Intelligence
BIC has invested heavily in its artificial intelligence (AI) capabilities since its inception. From acquiring supercomputers customised to facilitate AI technologies and computations, to recruiting a team of experts specialised in AI and computer vision. This investment has paid off in the form of making the Centre a leader in artificial intelligence-based solutions and a go-to name for partners, both in industry and academia for state-of-the-art AI-based solutions. BIC successfully delivers projects demanding artificial intelligence with deep learning and neural networks as the core technology. Enabling new opportunities across various industries and tackling challenges and obstacles previously proved to be unfeasible with classical approaches.
APPLICATION
- Signal processing is “where it all started” and continues to be a core part of the research activities. Leveraging deep learning for such tasks has allowed for never-before-seen results in various fields including; Automated Defects Recognition (ADR), Healthcare, predictive models, Digital Twin, condition monitoring etc.
- Computer vision algorithms, such as convolutional neural networks (CNNs), are employed in multiple research tasks, from spotting brain cancer, all the way to detecting production defects in photovoltaic panels.
- Our expertise is also within the use of Generative Adversarial Networks (GANs) for single image super resolution (SISR)
KEY PROJECTS
ASSAI
OLEDSOLAR
Opto-electronic devices are opening exciting new applications every day. With new display options using pliable substrates such as plastic and flexible glass, OLEDs manufacturers are bringing a wide range of new applications in lighting (e.g. energy efficient lighting) and different type of displays. Similarly, with the emergence of thin-film technologies in the solar cells market, new applications ranging from installations on curved surfaces to building integrated PV has become possible. However, to meet the industry requirements for mass production, including low cost, manufacturing volumes and efficiency, many challenges still need to be addressed. OLEDSOLAR aims to tackle these challenges by developing innovative manufacturing processes for critical steps in the production of opto-electronic devices including OLEDs, OPVs and CIGs solar cells. BIC’s responsibilities in this project are:
- Defect detection and localisation using image processing and machine learning
- Defect identification using machine learning
- Identification of affecting factors in a production line based on input sensors’ data
- Production line optimisation based on environmental and affective factors
- Sample quality control based on image processing