Digital Image Shape Recognition
Senior Project: 2008-2009
Sameed Abdul, Swain Brooks, George Engelbrecht, Fayaz Ghiasy and Lekha Pokharel
TKO Enterprises intends to provide products that incorporate low-cost,
application-oriented analysis of images for consumers. One possible device
might be a programmable embedded device with the capability to distinguish one
object from another. For example, the device might be able to distinguish a
person from a large dog. The applications for such devices are numerous
and range from security to data mining.
The goal of this project was to develop a test bed for the analysis of image
data and the testing of various image processing algorithms. The project can
then be used to develop tested algorithms for specific user applications.
The analysis goals for the project were to first operate on computer-generated
imagery, such as solid objects and geometrical shapes.
The resulting application is a platform for algorithm development and
evaluation. A user can select a set of images, as well as a set of algorithms
(along with algorithm parameters) to be applied to those images.
Results can then be analyzed to determine the quality of the algorithms used.
The system is constructed in a manner that allows algorithms to be "pluggable",
i.e. a user can easily construct new algorithms (in practically any language)
and incorporate those new algorithms into the test platform, as long as the
algorithm conforms to a well-defined "algorithm interface".

Algorithm Selection
Image Selection
Rectangle Detection
Hamster Blur
|