Deep learning in computer vision has set new standards in mobile and web-based applications. The power of learning-based computer vision has also tremendous potential in machine vision. Traditionally, machine vision in manufacturing employs analytic solutions often resulting in excellent accuracy but poor robustness. The goal of this project is to increase robustness of a vision-based measurement process in sheet metal manufacturing using deep learning.
The main objective of this project is to design and test millimeter-wave RF front end components for applications in 5G new radio communication systems utilizing the thin-film multilayer LTCC fabrication process developed at ACAMP. With these fabricated and characterized prototypes, ACAMP will be able to showcase its specialized LTCC fabrication process and provide support to potential Canadian technology clients and companies looking to develop technology for the upcoming 5G mobile communications market.
Over the recent years, cryptocurrencies have attracted tremendous amount of attention from both general public and professional investors as a new asset class. However, trading activities of cryptocurrencies are extremely fragmented and unregulated in most of countries around the world. The proposed research project aims to empirically study the microstructure of cryptocurrency exchanges in order to gain insight on what elements are needed to improve the market. In particular, the proposed research focuses on the potential role of smart order router (SOR).
Machine learning in the charitable sector is only beginning to be used successfully. Fundmetric Inc has applied various machine learning algorithms to predict donor behaviour, such as who will become a major ($10,000+) donor and which lapsed donors will return if stewarded correctly. There is ample opportunity for research in this domain, including building a super model across various charities, investigating feature synthesis and importance, applying deep learning to make use of the rich temporal data concerning donations, and learning to sequence email appeals.
Object detection and classification for surveillance applications via deep neural networks have attracted a lot of interests in computer vision (CV) communities. Accurate and fast CV algorithms can alleviate intensive manual labour and reduce human errors due to fatigue and distraction. In detection problem, the aim is to determine bounding boxes which contain interested objects and classify the category of the detected object. Thus, the detection problem can be formulated as a regression problem to localize multiple objects within a frame.
The voltage source converter VSC mimicking the behavior of a synchronous machine provides many advantages for grid operation. This “virtual synchronous generator (VSG)” will be implemented as a real-time simulator model on the RTDS simulator and used to investigate several operating scenarios.
The VS G behaves like a synchronous machine, which is one of the most widely used components of the legacy power system, and so it is well understood. The VSG can provide inertia and damping to the network.
Near-eye displays (NEDs) are small displays that are positioned closed to the eye, which conveniently places visual information in the line of sight of a user. NEDs need to be compact and lightweight as they are typically worn on the head, taking the form of glasses or goggles. In this research, we design and build a thin and transparent NED. The proposed NED uses a high fill-factor embedded concave micromirror array (ECMMA), and light field principles for virtual image formation.
Querying databases without a layer of privacy protection might lead to serious privacy issues. Such issues include access patterns and communication volume patterns. By combining the state-of-the-art privacy standard (differential privacy) and encryption in provides resilience to a host of attacks on remote databases, including data reconstruction attacks. However, there is still research work needed in building a private access system on top of an encrypted database.
D-Wave’s quantum computer is good at solving a specific type of problems known as Ising spin problems. However, in order to solve one of these spin problems, you must first solve another hard problem—embedding the spin problem on D-Wave’s quantum processor.
From the land of discrete mathematics, this embedding problem falls into a well studied branch of graph theory known as graph minors. Being that this problem is difficult in and of itself, D-Wave has developed a heuristic solution. This project’s main aim is to help improve this embedding process.