Spring 2005 EOS 758
Quantitative Methods in Remote Sensing
Instructor: Dr. Wenli Yang
Laboratory for Advanced Information Technologies and Standards (LAITS)
George Mason University
6301 Ivy Lane, Suite 620
Greenbelt, MD 20770
Tel: 301-230-0370, E-mail:yang@rattler.gsfc.nasa.gov
Course Discription:
This is an intermediate to advanced remote sensing course emphasizing on digital processing of remote sensing imagery. Students will be introduced to various digital image processing techniques and their applications to the earth observing remote sensing data. Topics of the course will include radiometric and geometric corrections, image enhancement, transformation, segmentation, and classification. Image acquisition sensors and platforms and commonly used data formats for remote sensing data will also be introduced.
The course is designed for 14-week semesters and consists of 13 lecture sessions and a student presentation session.
Prerequisites:
EOS 753, GEOG 579, or permission of instructor.
Course Materials:
1. Text book: Remote sensing digital image analysis: an introduction, 3rd edition, By John A. Richards and Xiuping Jia, Springer, 1999
2. Reference book: Digital image processing, 2nd edition, By Rafael C. Gonzalez and Richard E. Woods, Addison-Wesley, 1993
Assignments:
1. Homework: Homework will be assigned for each of the following topics: geometric correction, enhancement, transformation, feature selection, and classification. Homework is due two weeks after it is handed out.
2. Course project: Each student will design and conduct a project on applying digital image process technics in a remote sensing application area, write a 15- to 20-page (double space) project report and give a 20-minute presentation
at the end of the semester.
Grading:
Homework: 40 (8 for each of five homework assignments)
Course project: 60
A+ : 95-100
A : 90-94
A-: 85-89
B : 80-84
B-: 75-79
C : 70-74
D : 60-69
F : <60
Class Schedule:
| |
Jan. 26 Remote sensing data acquisition systems and data formats |
| |
|
This lecture will briefly review the history of the earth observing remote sensing systems, including platforms and sensors, with an emphasis on the recent sensor systems such as Lansatd-7 ETM+ and EOS Terra/Aqua MODIS. The lecture will also introduce commonly used remote sensing data formats, such as binary, HDF, and geoTiff. |
|
| |
Feb. 2 Radiometric and geometric correction of remote sensing data |
| |
|
This lecture will talk about sources of distortions introduced into remote sensing data, including atmospheric effects, platform and instrumentation errors, geometrical distortions, and methods to identify and correct these distortions. It will also cover image to image and image to map registrations.
|
|
| |
Feb. 9 Image enhancement - part I
|
| |
|
This lecture will discuss image radiometric enhancement techniques. Topics will include contrast enhancement, image histogram modifications (e.g., normalization, equalization), and image to image histogram matching.
|
|
| |
Feb. 16 Image enhancement - Part II
|
| |
|
This lecture will focus on image geometric enhancement techniques. The materials covered in this lecture will include image low and high pass filtering and general convolution filtering, edge detection and enhancement, and line detection and extraction.
|
|
| |
Feb. 23 Image transformation - Part I
|
| |
|
This lecture will discuss image spectral transformations such as band ratio, vegetation indices, the principal component transformation, and the Kauth-Thomas tasseled cap transformation.
|
|
| |
Mar. 2 Image transformation - Part II
|
| |
|
This lecture will introduce image spatial transformation, with an emphasis on the Fourier transformation, including convolution, sampling theory, properties of Fourier transformation, and the image filtering in frequency domain.
|
|
| |
Mar. 9 Image restoration
|
| |
|
This lecture will discuss image degradation and restoration, including degradation model and different restoration techniques such as algebraic approach, inverse filtering, least mean square filter, and interactive restoration.
|
|
|
| |
Mar. 23 Image segmentation
|
| |
|
This lecture will introduce autonomous segmentation of digital imagery, including detection of discontinuities, edge linking, boundary detection, region splitting and merging.
|
|
| |
Mar. 30 Feature selection
|
| |
|
This lecture will discuss techniques of selecting feature space for multispectral imagery, including those based on Jeffries-Matusita distance, principal component transformation, and canonical analysis.
|
|
| |
Apr. 6 Image classification - Part I
|
| |
|
This lecture will focus on image statistical classification. Topics will include maximum likelihood classification, minimum distance classification, parallelepiped classification, spectral clustering, etc.
|
|
| |
Apr. 13 Image classification - Part II
|
| |
|
This lecture will discuss neural network classification algorithms such as Kohonen self-organizing maps, learning vector quantization, adaptive resonance theory, and the backpropagation algorithm.
|
|
| |
Apr. 20 Image classification - Part III
|
| |
|
This lecture will discuss classification labeling techniques such as pixel unmixing, fusion of multisource classification results, and knowledge-based image analysis.
|
|
| |
Apr. 27 Hyperspectral image analysis
|
| |
|
The lecture will focus on the process of hyperspectral remote sensing imagery, including the characteristics, calibration, interpolation, classification, feature reduction and data compression of hyperspectral image data.
|
|
| |
May 4 Project presentation
|
|
| |
May 9 Grades reported to the registrar office
|
|
| |
|
|
Copyright Laboratory for Advanced Information
Technology and Standards, 2002-2005
George Mason University, 6301 Ivy Lane,
Suite 620, Greenbelt, MD 20770 USA
|
|
