ReadLS – An online intervention for improved reading

In this research, we are designing and implementing a study to determine if and how, an on-line intervention
(ReadLS) works to help students with developmental dyslexia learn to read. The project involves reviews of the
academic literature on developmental dyslexia and of the more mainstream literature describing commercially
available reading interventions. Using this knowledge, the interns will design and implement a controlled study
to compare the learning associated with ReadLS to another commercial reading intervention.

Sparse Representations for Embodied AI

Understanding scenes representing real world environments is a challenging problem at the intersection of Computer Vision research and Deep Learning and a necessary pre-requisite for Embodied AI. Embodied AI is an emerging field within Machine Learning that focuses on the challenges that need to be addressed for successful deployment of edge devices such as drones and robots. In this setting estimating the semantics of an environment plays an essential role in addition to how it can be efficiently navigated to solve a variety of tasks that can involve other agents as well.

Anticipation de mouvement des usagers dans l’espace d’une intersection avec feu de circulation à partir de données de sources multiples

Le projet consiste à concevoir, déployer et démontrer le fonctionnement d’un système ouvert de gestion des divers usagers à proximité de feux de circulation. L’implantation proposée démontrera la fusion de capteurs combinée aux mécanismes de localisation et d’authentification de la 5G ainsi qu’à des sources de données multiples en temps réel via l’informatique de périphérie afin d’améliorer la performance environnementale, sécuritaire et opérationnelle des intersections visées.

Investigating the Effect of Cognitive Training Simultaneously with Application of either Active or Sham Transcranial Alternating Current Stimulation on the Executive Brain Functions in Dementia Population

Built upon our successful pilot projects, the goal of this project is to investigate the effect of transcranial alternative current stimulation (tACS) paired with cognitive exercises in a placebo-controlled study on individuals with dementia, and develop novel technologies to monitor its effects and also predict a patient’s response to a treatment at baseline. This project can lead to an efficient optimized personalized treatment strategy for dementia.

Multi-sensor navigation solution with quality assessment and monitoring for real-time tracking of unmanned aerial vehicles beyond visual line of sight

This research project aims at developing an integrated navigation system that combines many technologies to enhance the positioning and tracking capabilities of the system. The navigation system to be developed is for use with UAV systems that require high precision and reliable navigation systems to provide continuous location information to the operator when the UAV is operated remotely or autonomous.

Testing, Validation and QA of Computer Vision Models & Data Sets in a Novel SAAS Environment (2)

Test and expand the capabilities of the Zetane software for application in complex artificial intelligence industrial (AI) projects with the objective of augmenting the users’ ability to gain new insights in model performance and gain more trust on how the data influences the AI models to arrive at the AI’s recommendations.

Detection and Classification of Pavement Defects Using Computer Vision

The objective of the project is to automate the detection of pavement defects. Defects can be of different types: cracks, deformations, potholes and others. These defects are cataloged and detailed in standards established by the Quebec Ministry of Transport in Quebec and various authorities in other regions of the world. At present, the inspection of pavement defects (e.g. potholes, cracks, ruts) is mainly done manually. Inspectors crisscross the roads or scrutinize images taken from inspection vehicles.

Investigating Algorithms for Automotive Keyless Entry Intrusion Detection

Automotive keyless entry systems use wireless communication to communicate information between the key and the car using Bluetooth technology. Such communication is susceptible to security risks including intrusion, where a malicious user injects signals into the system to cause a malfunction potentially resulting in unauthorized access. Thus, it is important to investigate methods that can be used to detect such an intrusion, in order to avoid any risks that may arise due to a potential malfunction.
The proposed research aims to investigate methods that can be used for intrusion detection.

Textile Embedded Vital Health Signs Monitor – Upgrade Testing Equipment

The goal in medicine is prolong life and prevent disease before it spreads and becomes irreversible. There must be a way to easily self-monitor or to allow medical professionals to continuously remotely monitor high risk patients. The main vital signs that ideally should be constantly evaluated are pulse rate, respiration rate, body temperature, oxygen saturation levels and blood pressure. Other secondary figures that should be watched over are glucose levels, cholesterol levels among others.

A Machine Vision- and AI-Based Solution for Optimal Comminution in Mineral Processing Circuits

Comminution, the process of reducing particle size so that valuable minerals can be liberated from the ore, consumes most of the energy used in mining operations. This process consumes an estimated four percent of the world’s electrical power and accounts for 50% of a mine site’s overall power consumption. Although mine-to-mill optimization strategies have been discussed for the past three decades, they have had little overall impact on the industry.