Doctoral candidate trying to automate human work
Written by Jinelle Shengulette
Abdul Haleem Syed is trying to take some work off humans and put it on computers, so the machines can better interpret various images, like those taken by satellites or planes.
Typically an image analyst would try to sort out all the data, things such as the number of trees, location roads and buildings or whatever, to produce maps, predict the path of a forest fire or respond to a natural disaster.
“We want a computer to be able to look at an image and say, ‘That’s a road, that’s a building and that’s a military compound,’–” said Syed, a doctoral candidate at Rochester Institute of Technology Center for Imaging Science. “Of course, the long-term goal of our research is complete automation, but our more immediate goal is to …develop tools that will help the image analysts deal with the massive amount of satellite data that is being collected.”
He presented his research during the International Conference on Geographical Object-based Image Analysis in May in Rio de Janeiro, where Syed won best student paper for his work.
We recently caught up with him to discuss the trip to Brazil and his work at RIT.
The usual day for me: Is a mix of activities involving research and teaching: writing, researching, teaching, grading, literature review, designing algorithms and programming.
Tools I use include: Programming and analysis software and research and documentation systems and software.
At the graduate/PhD level: School is almost like a job. Instead of a boss, you have an adviser, and you have certain goals that you are working towards.
Recently my biggest challenge has been: Programming and coding. Putting my ideas into action requires developing algorithms and being able to code them up. Because I am developing a framework from scratch, having a background like that of a computer science major would have definitely helped and perhaps made things go faster, but I am learning a lot along the way.
Winning best student paper: ... Means a lot. People at the conference recognize me now, and I am not just another conference attendee. It also means I need to set my standards even higher and do better at my next conference.
In five years: I see myself conducting research in this field at a corporate organization or a university.
Shengulette is a Rochester-area freelance writer. Submit story ideas and comments to RocNext@DemocratandChronicle.com.
Last Modified: 5:43pm 19 Jul 12