Finger Counter is a computer-vision system that counts the numbers of fingers held up in front of a video camera in real time. The system is designed as a simple and universal human-computer interface: potential applications include educational tools for young children and supplemental input devices, particularly for persons with disabilities. The interface is language independent and requires minimal education and computer literacy.
Finger Counter uses background differencing and edge detection to locate the outline of the hand. The system then processes the polar-coordinate representation of the pixels on the outline to identify and count fingers: fingers are recognized as protrusions that meet particular threshold requirements. The system also logs the frequency of different inputs over a given time interval. We implemented the Finger Counter interface under Linux using Video4Linux. The system was tested extensively under various lighting and background conditions. During testing, the system successfully counted the fingers of numerous subjects with disparate hand shapes and sizes and skin color. Finally, we incorporated the Finger Counter interface into a children's game for learning and entertainment
S. Crampton and M. Betke, "Counting Fingers in Real Time: A Webcam-Based Human-Computer Interface with Game Applications," Accepted at the Conference on Universal Access in Human-Computer Interaction, affiliated with HCI International 2003, Crete, Greece, June 2003.
S. Crampton and M. Betke, "Finger Counter: A Human-Computer Interface." 7th ERCIM Workshop "User Interfaces for All," UI4ALL 2002, pp. 195-196, Paris, France, October 2002.
Published material is based upon work supported by the National Science Foundation under Grant No. 0093367 and Office of Naval Research Grant N000140110444.
"Counting Fingers in Real Time: A Webcam-Based Human-Computer Interface with Game Applications" won the Chancellor's Award at Boston University's Science and Technology Day 2003.
Here is a video clip of a Finger Counter user: fingercounter-kevin.mpg
(6.4 MB)
EcoTracker takes infrared images and automatically identifies and tracks bat-like objects that appear in them. (The infrared images were collected outside bat caves in south Texas by Dr. Thomas Kunz of the Department of Biology, Boston University.)
EcoTracker determines each bat's position and velocity at each frame, and displays the bat objects with colored trails to illustrate their flight patterns. Proposed work on the system initially will concentrate on evaluating and refining the tracker's parameters and determining its error rate. To do this, we will compare hand-marked images with the tracker's annotations. Also, we would like to incorporate the periodicity of the wing flapping into the tracking system. Other proposed algorithmic work will include tracking in 3D, based upon heuristical information regarding the size of an average bat.
We also may collect radar data at the same locations to track the entire flock of bats as they search for and devour insects in the atmosphere. To track the flock, we anticipate developing a 3D optical-flow algorithm. Estimates of the number of bats per cave run from 1-3 million.
S. Crampton, M. Betke, T. Kunz, and J. Horn, "A Prototype Computer-Vision System for Tracking Bats," Presented at the Bat Project at NCEAS, Santa Barbara, CA, January 2003.
Bat Tracker is part of a pending NSF grant.
Here is a video clip showing the EcoTracker in action: ecotracker.mpg (746 KB)