[ MECHANICS ] [ CONTENT ] [ COURSEWORK ] [ SCHEDULE ]

DEPARTMENT OF ELECTRICAL AND COMPUTER ENGINEERING

NORTHEASTERN UNIVERSITY

ECEU692     INTRO TO SUBSURFACE SENSING AND IMAGING      SPRING 2004

MECHANICS

Instructors: This class will be co-taught by Prof. Charles DiMarzio and Prof. Dana Brooks

Prof. DiMarzio	Prof. Brooks
334 Egan Center	423 DA
617 373 2304	617 373 3352
dimarzio@ece.neu.edu	brooks@ece.neu.edu
http://www.ece.neu.edu/faculty/dimarzio.html	http://www.cdsp.neu.edu/info/faculty/brooks/brooks.html

Class and Office Hours:

Class Hours are Mon Wed 2:50pm-4:30pm
Office Hours: Prof. Brooks, Mon. 10AM-Noon in 423 DA, and electronically on a schedule to be announced weekly, Prof. DiMarzio Thu 10AM-Noon in 302 Stearns, and Wed 9-10PM electronically.

Prerequisites:

Officially, a course in linear systems and a course in electromagnetics, or by permission of instructor. However the only true required prerequisites are a course in circuits, a course in differential equations, and a course in linear systems. In particular, Computer Engineer students who are interested in the course but do not have the prerequisites are urged to contact one of the instructors to discuss their interest.

Textbook:

There will be no textbook for this class. It will be taught from notes which will be posted on the web. In addition occasional readings may be assigned as necessary. Prof.'s Brooks and DiMarzio will maintain a library of references books and articles for the course that they will lend to students during the quarter per student request; the list of available books will be on the course web site. In addition, students are encouraged to look in the NU library for relevant texts.

Back to top

COURSE CONTENT

Course Description:

This course is an introductory unified look at the emerging field of subsurface sensing and imaging (SSI). Major themes include the interrelatedness of the three technological levels of

1. sensing,
2. modeling and signal processing, and
3. computational technology,

the similarity of SSI across diverse problem domains and size scales, and the variety of information extraction strategies such as localized imaging and the use of multiple views in space, wavelength, and other accessible dimensions of the sensing modalities. The course will be organized around hands-on experience with a particular SSI modality (optical imaging) and will include experimental measurement and subsequent visualization and processing of the students' measured data.

Course Objectives:

The student completing this course should

• Gain a broad understanding of the interrelated sensing, modeling, processing, and computational aspects of Subsurface Sensing and Imaging (SSI).
• Gain specific understanding of these ideas through study of a particular SSI modality, optical imaging.
• Learn aspects of basic optical theory as related to SSI.
• Learn aspects of basic signal and image processing as related to SSI.
• Learn aspects of modeling and linear algebra as related to SSI.
• Become familiar with the basic principles and some specific examples of SSI imaging with a variety of sensing/measurement modalities including ultrasonic, electrical, Xray, and electromagnetic,
• Become familiar with a variety of application areas of SSI for both medical/biological and environmental purposes, including microscopy, anatomical and functional medical imaging, and imaging of buried objects in the ground and in the ocean,
• Be able to draw on their detailed understanding of optical imaging to strengthen their insight into how other SSI modalities work,
• Gain specific familiarity with some optical measurement and measurement verification techniques for hyperspectral SSI.
• Gain familiarity with extracting subsurface information from hyperspectral measurements.

What You Should Expect From This Course:

• This is the third time this course is being taught, and it is a course in a brand-new area. As such it is by necessity a ``work-in-progress''. The two of us are really excited about teaching it again, and we are hoping for your input in making it work.
• This course is being co-taught. This is a relatively new teaching model for us and for the Northeastern ECE department. Prof. DiMarzio's area of expertise is optics and optical imaging, while Prof. Brooks' area is signal and image processing, especially as applied to biomedical problems. (We are research collaborators, and are both participants in the National Science Foundation funded Engineering Research Center for Subsurface Sensing and Imaging Systems (CenSSIS), headquartered at Northeastern (for more information on CenSSIS check www.censsis.neu.edu).) We feel that co-teaching is a particularly appropriate model for this course because SSI is an inherently interdisciplinary field, and CenSSIS in particular is devoted to emphasizing connections across SSI sub-disciplines and applications. We are working together to refine the course, are jointly responsible for the entire course, and expect to both be present and participate in most of the classes during the quarter.
• We have chosen to focus on a specific SSI modality in some detail, rather than to simply present a broad overview of the field. Specifically we will look at hyperspectral SSI, a form of sensing and imaging where optical measurements are made through a large number of narrow-band sensors and then models of the propagation of light and of the environment being imaged are used to extract information of interest from the measurements.
• We will emphasize similarities and connections across three important aspects of SSI:

  1. interrelatedness between sensing, modeling, processing, and computation,
  2. similarities across broad range of problem domains and size scales, and
  3. connections between model problems, application problems, and experimental / processing problems

• We will attempt to make the course meet your needs. This means that we will try to modify the material being presented and the tasks being assigned as we learn more about the background, interests, and capabilities of the class.
• Our attitude about this course is that it is somewhat like a graduate school course. In particular the assumption is that everyone taking the course is doing so because they are interested, motivated, and want to learn.
• We expect everyone to get a good grade if they make the expected effort. Details on grading are below. In general, grades in college courss are typically based on a weighted combination of accomplishment (what you learned) and effort (how hard you worked). In different courses this weighting can vary widely. In this course the weights will be more-or-less evenly balanced; however strong effort will be sufficient to earn a good grade, leading to our expectation of a good outcome. In particular there will be no attempt to make the class average correspond to any particular grade.

What We Will Expect From You:

• We expect you to be interested in the subject and in making the course work well,
• We expect you to be willing to work hard to learn, and not to shy away from challenges.
• At the same time, we expect you to ask questions of yourselves, of each other, and of us, when you do not understand something.
• Several aspects of this course--the interdisciplinary subject matter, the co-teaching format, the integrated experimental/processing approach--are ``experimental'' in themselves (and in some sense a paradigm for the interdisciplinary, collaborative, integrated topic of SSI). We expect you to want to be a part of this ``experiment''.
• We expect you to be active learners, sharing responsibility for making the course work. We want the class to be stimulating and interactive, and we expect you to work with us to make this happen. If you have difficulties, complaints, suggestions, even on the rare occasion complements, we want you to share them with us.
• We expect everyone to work hard, and as as a result we expect everyone to get a good grade, which they will deserve.

Capstone Possibilities:

If you are a senior taking this course and are interested in doing a capstone project related to SSI, please speak to us about it.

Back to top

COURSEWORK

Groups:

You will form groups of students, probably with 3 or 4 students per group. These groups will work together when taking data in Prof. DiMarzio's lab and also work as a team on their term project.

Lab Work:

We presently anticipate that there will be two assignments during the quarter that will require you to take data using facilities in Prof. DiMarzio's lab. This will be considered as part of your homework assignments. For each such assignment, each group will have to schedule time (a 2 hour slot) to use the lab and take data. All lab work will be done under the supervision of graduate students working in the lab. Homework sets will be the individual responsibility of each student.

Homework Sets:

We plan to have 5 homework sets assigned during the quarter. They will involve a combination of analytical work and computer work, including presentation and processing of data you take in the lab. Each assignment will generally consist of a relatively small number of relatively difficult problems --- the idea is to get you to stretch yourselves to understand, develop, and extend ideas presented in class. Each assignment will need to be turned in, in electronic form, as a well-organized document.

In-Class Problem Presentations:

One common complaint of both students and instructors is that students never really get a chance to close the loop on their homework since once they turn it in it tends to be forgotten, and whatever was supposed to be learned in doing the homework gets lost. Another complaint is that students do not get enough opportunity to practice oral presentation skills. To address both of these problems, we plan to try the following scheme: on or just after the day each homework assignment is due, we will have an in-class problem session for about half the 100-minute class period, in which students will take turns presenting their solutions to the homework problems, with each presentation planned for to last approximately 10 minutes. Students performance will count towards their homework grade. Although of course accuracy of solutions is important, much more important are evidence of effort and thought in both the problem solutions and the presentations. Students will be chosen randomly so that each student will present at least one problem during the semester. After the problem set, students may resubmit one (or possibly more, depending on the assignment) of their homework problems at the next class for a revised grade. Further details of this plan will be discussed in class.

Project:

A term project will be assigned early in the quarter. You will work on this project as a group; however each individual will have clearly defined responsibilities within the team. The groups will have considerable choice of project topic. Each group will write a project proposal (tentatively due Jan. 30), propose and then be responsible to meet a sequence of milestones during the quarter, including a formal progress report approximately mid-way through the project time period, do a joint presentation at the end of the quarter on their project, and write a project report, summarizing the work of the group.

Software:

You will be expected to make extensive use of MATLAB for data acquisition and processing as well as other homework assignments. You are free to use any word processing or typesetting software you choose. In the (perhaps unlikely) event you choose to use to use L^ATEX we will be happy to provide extensive help. We will provide help with Matlab in response to student request---the stronger and more specific the request, the more help will be provided. This help may take the form of in-class live demonstrations, extra scheduled Matlab sessions, work in regular office hours, provision of example Matlab programs, etc.

Electronic Communication:

We will make regular use of the course web page
( www.ece.neu.edu/courses/eceu692/2004sp ) and email (to Prof. DiMarzio, Prof. Brooks, or both). Both instructors make regular use of email and check it frequently. We will also use Blackboard, but for both of us it is the first time using it, so please bear with us, and check the course website for updates. Any student who does not have a Blackboard account should apply for one immediately at http://www.edtech.neu.edu/blackboard/accounts/students.php . We will assume you check your email regularly, and it is your responsibility to make sure that we have a good email address (or addresses) for you. Our goal is to post the lecture notes to the web page (generally powerpoint and/or pdf files) for each class at least 24 hours before the class; if we meet that goal it will be your responsibility to download the slides and bring them to class. If we don't we'll bring copies to class with us.

Grading Scheme:

Grades will be based 60% on the homework sets (12% per set) and 40% on the projects.

Back to top

TENTATIVE COURSE SCHEDULE

Tentative Schedule:

Note that this schedule should be considered very much as a working plan, to be modified as the course develops:

Dates	Week	Topic
Jan 7	1	Intro to SSI. What we will cover, and why.
Jan 12	2	Imaging with Light: The optical spectrum, geometric optics, light sources, detectors, cameras. Brightfield microscopy, point-spread functions, and optical transfer functions. Homework Set 1 and Project Assignment out.
Jan 14	2	Imaging with Light: continued. PSF's and deconvolution.
Jan 19	3	Holiday
Jan 21	3	PSF's and deconvolution or Imaging with light: continued. Homework set 1 due. Problems session 1. Homework Set 2 out (includes lab problem).
Jan 26	4	PSF's and deconvolution or Imaging with light: Weak-scattering and Beer's Law. Simple models, spectral mixing, instrumentation, Plans for experiments. Optional Project Pre-proposals due
Jan 28	4	Linear systems introduction, with spectral unmixing example
Feb 2	5	Linear systems, continued. Sampling issues, diffraction and apodization. Project Proposal Due
Feb 4	5	Imaging with light: Strong scattering.
Feb 9	6	Imaging with light: Confocal and multi-photon microscopy.
Feb 11	6	Imaging with light: Microscopy: continued. Some radiometry. Homework set 2 due including lab problem. Problems session. Problem set 3 out.
Feb 16	7	Holiday
Feb 18	7	Radiometry, continued. Sampling issues, diffraction and apodization. Linear systems in strong scattering.
Feb 23	8	Tomographic Imaging with Light: Forward models Homework set 3 due. Problems session 3. Problem set 4 out.
Feb 25	8	Tomographic Imaging with Light: Forward models, continued. Project Status Report Due.
Mar 1-5	Break	Spring Break Week
Mar 8	9	Tomographic Imaging with Light: Inverse solutions
Mar 10	9	Tomographic Imaging with Light: Inverse solutions continued
Mar 15	10	Tomographic Imaging with Light: Inverse solutions continued
Mar 17	10	OTHER SSI MODALITIES: Chosen from Acoustics and Ultrasound, Xray, Ground Penetrating Radar, Electrical Impedance Tomography, Magnetic Resonance Imaging, Mixing of sound and light, and possibly others. Order TBD; some topics to be presented by guest lecturers. Homework set 4 due, including lab problem. Problems session 4. Problem set 5 out.
Mar 22	11	OTHER SSI MODALITIES
Mar 24	11	OTHER SSI MODALITIES
Mar 29	12	OTHER SSI MODALITIES
Mar 31	12	OTHER SSI MODALITIES Homework set 5 due. Problems session 5 Wednesday.
Apr 5	13	OTHER SSI MODALITIES
Apr 7	13	OTHER SSI MODALITIES
Apr 12	14	OTHER SSI MODALITIES Draft of project final report due.
Apr 14	14	OTHER SSI MODALITIES
Apr 20-23	Finals	Project Presentations

Back to top --------------030304090701000000090605--