Enhancing Line Tracking through Image Modeling and Restoration: A UCSD ECE 191 Project, Study Guides, Projects, Research of Electrical and Electronics Engineering

Download Enhancing Line Tracking through Image Modeling and Restoration: A UCSD ECE 191 Project and more Study Guides, Projects, Research Electrical and Electronics Engineering in PDF only on Docsity! UCSD Corporate Affiliates Engineering Design Process ECE 191 Projects 4 Jan 2003 Enhanced Line Tracking Through Image Modeling and Restoration Submitted by: ORINCON Corporation Point of Contact: Dr. Don Pace, (858) 795-1232 ORINCON Corporation has developed advanced line-like feature tracking algorithms most recently deployed as signal processing automation upgrades in the Navy’s Advanced Rapid COTS Insertion (ARCI) program. ORINCON line trackers identify and track sources of energy that appear as line objects on bearing vs. time (BTR) and on bearing vs. frequency vs. time (LOFARgram) displayed data. Tracking is always performed in a recursive manner, with best current estimates of broadband or narrowband line-like feature parameters (location in bearing and frequency, bandwidth, etc.) always available to an observer or to down-stream processing functions. In essence, the line tracking problem is one of identifying line-like components in gram “images” as the images are updated in time. Such an image-based interpretation suggests that advanced image processing techniques may be used to enhance the line-like features in the gram data and therefore improve the overall performance of the line tracking algorithm. Objective: Identify, implement, and demonstrate an image processing algorithm that will enhance line-like objects, and therefore improve the overall performance of line tracking software. The algorithm must be computationally efficient, which may mean that it must operate on the gram images in a recursive manner. Suggested Tasks:  Identify suitable image modeling, restoration, or enhancement algorithms.  Develop Matlab prototype of image modeling and restoration algorithm  Develop a simple simulation scenario generator (in Matlab)  Process simulated data sets and define algorithm behavior (in Matlab)  Define metrics and assess performance of algorithm on real data samples (to be provided) References: 1. R. L. Kashyap and K. Eom, “Robust Image Modeling Techniques with an Image Restoration Application,” IEEE ASSP, Vol 36, No 8, pps. 1313-1325, August 1988.