Title: CSci 6971: Image Registration Lecture 1: Introduction January 13, 2004
1CSci 6971 Image Registration Lecture 1
IntroductionJanuary 13, 2004
Prof. Chuck Stewart, RPI Dr. Luis Ibanez, Kitware
2Outline
- Syllabus
- Registration problem
- Applications of registration
- Components of a solution
- Thematic questions underlying registration
- Software toolkits
3Syllabus - Topic
- Image registration
- Determining the mapping between two images of the
same object, similar objects, the same region or
similar regions - All aspects of the problem will be covered
- Underlying mathematics
- Images
- Algorithms
- Implementations
- Applications
- Special emphasizis on software toolkits
4Syllabus - Instructors
Prof. Chuck Stewart 107 Amos Eaton Rensselaer
Polytechnic Institute 518-276-6731 stewart_at_cs.rpi.
edu
Dr. Luis Ibanez Kitware Corporation
(518)-371-3971 x 112 luis.ibanez_at_kitware.com
5Syllabus - Distributed Course
- NSF Center for Subsurface Sensing and Imaging
Systems (CenSSIS) Course - Four universities
- Boston University,
- Northeastern University,
- Rensselaer Polytechnic Institute,
- University of Puerto Rico Mayaguez
- Lectures live at Rensselear
- Lectures recorded in voice-annotated powerpoint
for remote students - Lectures missed due to weather or travel will
also be available via voice-annotated powerpoint
6Syllabus - Office Hours
- At Rensselaer
- Tuesday and Friday immediately following lecture
- Distributed office hours
- Web-meeting
- Anyone can participate
7Syllabus - Prerequisites
- Data structures
- Calculus
- Linear algebra
- Vectors and matrices
- Experience working with images
- C programming experience
- Templates!
8Syllabus - Requirements
- Weekly homework assignments and programming
projects - Extended programming project (due Tuesday, April
6) - 10-page research paper (due Tuesday, May 4)
- Each is worth 30 of the semester grade
- Late assignments will not be accepted without
prior arrangement or a verified personal
emergency - Last 10 is for acting as a scribe during
lecture. This is a pass/fail requirement that
off-campus students automatically pass.
9Syllabus - Course Materials
- Voice-annotated powerpoint lectures will be
placed on CenSSIS website - Software toolkits will include tutorials
- Reading materials will also be placed on the
CenSSIS website
10Syllabus - Topics
- Introduction
- Mathematical background
- First examples
- Intensity-based registration and ITK
- Feature-based registration and the CenSSIS/RPI
toolkit - Initialization techniques
- Multiresolution techniques
- Mutual information
- Deformable registration
- Video registration and image mosaics
- Trends and research questions
11Syllabus - Academic Integrity
- Students may discuss homework and programming
assignments - Solutions must be written in students own words
- Extended programming project and research paper
must be individual work with appropriate
citations - A serious incident will result in failing the
course
12Registration Problem Definition
13Example Mapping Function
q (912,632)
p (825,856)
Pixel scaling and translation
14Registration Problem Definition
q (912,632)
q T(pq)
- Problems
- Form of mapping function T
- Unknown mapping parameters q
- Unknown correspondences, p,q
Chicken-and-egg problem
15Applications Multimodal Integration
- Two or more different sensors view same region or
volume - Different viewpoints
- (Some specialized sensors have two or more
coincident modalities, so registration is not
needed.) - Different information is prominent in each image
- The images may even have different dimensions!
- Range images vs. intensity images
- CT volumes vs. fluoro images
16Example MR-CT Brain Registration
MR
CT
- MR (magnetic resonance) measures water content
- CT measures x-ray absorption
- Bone is brightest in CT and darkest in MR
- Both images are 3d volumes
Source http//www-ipg.umds.ac.uk/d.hill/hhh/10/10
.pdf
17MR-CT Registration Results
Superimposed images, with bone structures from CT
in green
Aligned images
18Retinal Angiogram and Color Image
19Applications Image Mosaics
- Many, partially overlapping images
- No one gives a complete view
- Goal stitch images together
- Requires
- Limited camera viewpoint such as rotation about
optical center - Simple surface geometry such as plane or
quadratic
20Retinal Image Mosaics
21Sea-Floor Mosaics
Courtesy Woods Hole Oceanographic Institution
22Spherical Mosaics
Images from Sarnoff Corporation
23Applications Building 3d Models
- Range scanners store an (x,y,z) measurement at
each pixel location - Each range image gives a partial view
- Must register range images and texture map them
- Applications
- Reverse engineering
- Digital architecture and archaeology
24Examples
http//www1.cs.columbia.edu/allen/NEW/workshop.ht
ml
25Applications Change Detection
- Images taken at different times
- Following registration, the differences between
the images may be indicative of change - Deciding if the change is really there may be
quite difficult
26Retinal Change Example
27Regions Showing Change
28Applications Video Super-Imposed on 3d Model
Taken from Sarnoff Corporation research
29Other Applications
- Multi-subject registration to develop organ
variation atlases. - Used as the basis for detecting abnormal
variations - Object recognition - alignment of object model
instance and image of unknown object - Industrial inspection
- Compare CAD model to instance of part to
determine errors in manufacturing process
30Steps Toward a Solution
- Analyze the images
- Determine the appropriate image primitives
- Determine the transformation model
- Geometric and intensity
- Design an initialization technique
- Develop constraints and an error metric on the
transformation estimate - Design a minimization algorithm
- Develop a convergence criteria
31Software Toolkits
- ITK
- Medical image processing, segmentation, and
registration toolkit - C, heavily templated, data flow architecture
- Registration stresses intensity-based approaches
- VXL
- Computer vision applications
- C, moderate templating
- Registration stresses feature-based approaches
- CenSSIS tool suite is a hybrid of these
- They can be used together
- Programming styles are different
32Summary Pervasive Questions
- Three questions to consider in approaching any
registration problem - What intensity information or image structures
is/are consistent between the images to be
registered? - What is the geometric relationship between the
image coordinate systems? - What prior information can be used to constrain
the domain of possible transformations?
33Looking Ahead Lecture 2 - Friday, January 16
- Mathematical background, part 1
- Vectors and matrices
- No meeting on Rensselaer campus
- Voice-annotated lecture will be posted Thursday
night, January 15
34Homework Problem
- Due Tuesday, January 20 at 12 noon (via email to
Professor Stewart) - Problem
- Find an application of registration, preferably
in a research area of interest to you. In a
short write-up (less than a full page), describe
the problem and attempt to sketch answers to the
three Pervasive Questions posed.