creator Waite, Sheldon advisor Oruklu, Erdal This work describes an implementation of a system that recognizes tra c signs within an image or video frame. In addition it covers a brief history of automotive technology and the safety motivations for the technology presented here. An algo- rithm is presented that takes RGB image data, extracts relevant pixels, lters the image, labels prospective tra c signs and evaluates them against template tra c sign images. Furthermore a system is described that uses a Virtex 5 Xilinx FPGA and the EDK tool kit to create an embedded processor and a hardware IP periph- eral that together implement the algorithm. This implementation is shown to have positive results both in terms of timing performance and accuracy. Submitted by Dana Lamparello (dlampare@iit.edu) on 2013-03-04T19:43:33Z No. of bitstreams: 2 Waite_Real-Time_Traffic_Signal_Recognition_Final_Final.pdf: 20295762 bytes, checksum: 2fcc9e9571bda78aecaac38e41b22e7c (MD5) Waite_Real-Time_Traffic_Signal_Recognition_Final_Final_signe.pdf: 13373 bytes, checksum: f421dc1ec8de76de89cc8cce46d2f0f6 (MD5) Made available in DSpace on 2013-03-04T19:43:33Z (GMT). No. of bitstreams: 2 Waite_Real-Time_Traffic_Signal_Recognition_Final_Final.pdf: 20295762 bytes, checksum: 2fcc9e9571bda78aecaac38e41b22e7c (MD5) Waite_Real-Time_Traffic_Signal_Recognition_Final_Final_signe.pdf: 13373 bytes, checksum: f421dc1ec8de76de89cc8cce46d2f0f6 (MD5) Previous issue date: 2012-07 M.S. in Computer Engineering, July 2012 2012-07-12 2012-07 http://hdl.handle.net/10560/2918 en Thesis born digital application/pdf In Copyright http://rightsstatements.org/page/InC/1.0/ Restricted Access ECE / Electrical and Computer Engineering Illinois Institute of Technology Affiliated department