You have been engaged as a consultant by Wivenhoe Innovative Software Enterprises (WISE) because of your extensive knowledge of computer vision gained through CE866. WISE is developing a vision-based system for use in cars to alert the driver if he or she strays outside their lane. Their prototype hardware consists of a forward-looking movie camera mounted in the centre of the car. Your job is to write the vision software to process video captured from the camera.
WISE would like you to use the white lines painted at the boundaries of the lanes to identify them. Real-time execution is not important, the program just needs to be able to identify the lane boundaries reliably. You should annotate image frames in some way, perhaps by painting the lanes on the image or by writing some text on them, before they are displayed by your software. You are free to use any algorithm you like; one approach that suggests itself immediately is to search in specific parts of the image for white regions, then ascertain whether the direction of travel is within them. (If you're having trouble working out an algorithm, do speak with a demonstrator.)
To help you develop the software, you have been provided with a single movie on which to work. This is a forty-second clip of about 6 Mbytes in size.
You are required to use OpenCV, though you have a choice of programming in C, C++, Java or Python. WISE uses Linux exclusively, with an installation identical to that in Lab 1, so your software must run there.
|Submission deadline:||Wednesday 17th November at 11:59:59|
|What to submit:|
Do not submit the movie!
|Marks returned:||three weeks|
|Overall marks breakdown:||according to the usual assessment criteria|
Remember to identify your work with your registration number only. The coursework system allows you to upload your work as often as you like, so do keep uploading your files as you develop them.
|Last updated on 2016-10-24 10:58:35||Web pages maintained by Adrian F. Clark [contact]|