ECE-S 433
INSTRUCTORS
Prof. Fernand Cohen, Rm. 7-126, Tel. x1420, e-mail: fscohen@cbis.ece.drexel.edu
Prof. Athina Petropulu, Rm. 7-221, Tel. x2358, e-mail: athina@cbis.ece.drexel.edu
TEACHING ASSISTANT
Georgia Georgiou, Rm. 7-104, Tel. x2174, e-mail: ggeo@cbis.ece.drexel.edu
CLASS SCHEDULE
Lectures: MON&WED 12:00-12:50 RM 7-308
Recitations: FRI 12:00-12:50 RM 7-403
TEXT/REFERENCES
Class Notes
John G. Proakis and Dimitris G. Manolakis, Digital Signal Processing
[Second Edition], Upper
Saddle River, NJ: Prentice Hall Co, 1996.
J.R. Deller, Jr., J.G. Proakis, J.H.L. Hansen, Discrete-time processing
of speech signals, Macmillan, 1993.
A.K. Jain, Fundamentals of Digital Image Processing, Prentice Hall
1989.
GRADING POLICY
Projects: 50%
Homework: 25%
Quizzes: 25%
Late assignments will not be accepted.
Projects will be assigned according to the schedule given in Course
Description. They will be based on material covered in class and lab/recitation
sessions, and could be implemented in the Multimedia Signal Processing
Laboratory (Rm 7-403).
Homework will be assigned periodically.
Solutions will be provided on the homework due date.
There will be a weekly 7 mins quiz to be held every Friday during recitation time. The lowest two quiz grades will be omitted. No make-ups will be given.
Please note:
Copying homework, project, or quiz from someone else, or allowing someone
else to copy for your own, is considered to be cheating. Any students cheating
will receive a zero. In addition, the ECE Department Committee on Cheating
will be notified and the student will be referred to the committee chair
for further action.
TENTATIVE COURSE OUTLINE
PART1: Speech Processing (A.P. Petropulu)
Speech Production and Modeling (1 week)
anatomy of the speech production system; phonemics/phonetics; acoustic
theory of speech production; discrete time modeling
Speech Analysis (2 weeks)
short-term processing; linear prediction and applications on
pitch estimation,
formant estimation; cepstral analysis and applications to recognition.
Project I: Model your vocal tract and use your model to synthesize speech.
Coding (1 week)
time-domain coding; frequency-domain coding
Recognition (1 week)
speaker independent vs speaker dependent; speaker recognition;
speaker verification
Project II: Design a voice operated door unlocking system
PART2: Image Modeling and Recognition (F.S. Cohen)
Edge detection (1 week ) -- Computer Assignment I;
Boundary representation and shape recognition (1 week);
Moments, invariants, and shape recognition (1 week) -- Computer Assignment
II;
Texture representation and shape from texture (1 week) -- Computer
Assignment III;
Shape Estimation from Stereo, Motion and Object Tracking (1 week).


