Image analysis – Shark detection

This is the only project that can be published because it has already made the news

The objective was to design and develop a software for image recognition. Using Computer Vision and Machine Learning techniques, the aim was to recognize sharks from a video. A drone flies over the water with a camera and the software, either on board or remotely, analyzes the video to alert of the presence of a shark. The project was part of a bigger one which involved the air-frame and hardware departments.

 

Accomplishments
  • Designed and delivered a 1500000$ worth software which was tested with positive results. 
  • Lead a team on a cutting-edge project, implementing Computer Vision, Machine Learning and CPU/GPU techniques.
  • Researched for the latest innovations and applied state of the art methodologies with positive results. 
Overcome impediments
  • In order to build that kind of programs, I needed a computer with specific requirements. Since the company was a startup, I suggested the best specifications and built a computer to match them.
  • I designed the architecture and estimated timing. To work quicker, the department needed more programmers, therefore I wrote the job opening ad and joined the interviews. I was also in charge of the technical interviews.
  • I interviewed biologists on shark’s peculiarities in order to design a software to locate them.  
Skilled acquired or improved:

Computer Vision, Object Oriented, software architecture design, C++, UML, SQL, GPU architecture, Python, Cuda, OpenCV, Eclipse, Ubuntu, Agile methodologies, team leader, team player.