Vision-based Drone Identification and Tracking System - BC-638

Project type: Research
Desired discipline(s): Engineering - computer / electrical, Engineering, Engineering - mechanical, Computer science, Mathematical Sciences, Mathematics, Physics / Astronomy, Natural Sciences
Company: Bluvec Technologies Inc.
Project Length: Longer than 1 year
Preferred start date: As soon as possible.
Language requirement: English
Location(s): Burnaby, BC, Canada; Canada
No. of positions: 2
Desired education level: Master'sPhD
Search across Mitacs’ international networks - check this box if you’d also like to receive profiles of researchers based outside of Canada: 

About the company: 

Founded in Vancouver, Canada, Bluvec Technologies Inc. is the pioneer developer of Deep Signal Inspection (DSI) technology and a leading supplier of counter-drone technology in the industry.

The company's R&D team is gathered from the world's leading institutions, including multiple senior information security, network security, and wireless experts, with 30% of R&D team members owning Ph.D. degrees and master's degrees accounting for 60%.

Our company uses self-researched and developed DSI (Deep Signal Inspection) technology to provide world-class Counter Drone solutions.  The DSI technology is designed to detect, identify, localize, defend and manage civilian drones, covering 98% of currently available drone types, and is widely used in government agencies, defense military, aviation airports, prisons, stadiums, industrial facilities, etc.

Describe the project.: 

This system can detect, identify, locate and track drones in the surrounding low-altitude range 24 hours a day. The system integrates active detection, type identification, information display and control, target positioning functions in one. Compared with other similar equipment at this stage, it has the unique functional advantages of independent detection, active discovery, accurate positioning, and is suitable for low-altitude and sensitive areas, such as airports, petrochemicals industries and large venues. The equipment can integrate with wireless detection equipment to effectively making up the shortcomings and weaknesses of wireless equipment that cannot detect "silent flight" drones, and greatly improving the comprehensive use of wireless detection equipment.
This is a research project. The main used technologies contain image signal processing, fast focusing technology, computer vision, deep learning and etc.
Currently, the detection distance is 500 meters. In the next stage, our aim is to extend the detection distance to 3 kilometres. Meanwhile, we need to increase our detection precision.
We provide 2 positions for interns. One is Camera Researcher which requires the intern to research and select the camera hardware for our project. The other is Machine Learning Researcher which requires the intern to design and implement object detection and real time tracking algorithms.

Required expertise/skills: 

Camera Researcher
1. Research and selection of the camera hardware, including lens, image sensors, ISP (Image Signal Processing) chips and module interfaces.
2. Design and implementation of hardware assembly, and writing hardware schematic and related documents.
3. Image acquisition and image quality analysis of camera modules, and writing image quality evaluation reports.
4. Failure analysis and debugging of the camera hardware at both module and product level.
1. Major in Optics, Electronics, Image Signal Processing, Physics or related fields.
2. Familiar with the photosensitive principle of image sensors, and familiar with the architecture and principle of colours.
3. Familiar with the principle and algorithm of camera hardware assembly.
4. Familiar with fast-focusing technology.

Machine Learning Researcher
1. Research on real-time object detection and tracking algorithms.
2. Leverage open source libraries and resources to design and develop deep learning models.
3. Analyse and improve model performance and speed.
4. Write production-ready Python code.
1. Experience with computer vision and deep learning.
2. Familiar with either object detection or real-time tracking algorithms.
3. Familiar with deep learning models such as YOLO and SSD, and able to adapt it to new applications
4. Familiar with PyTorch or TensorFlow, and computer vision libraries such as OpenCV.
5. Highly self-motivated and quick learner
6. Smart, practical, passionate to create great products!