Image analysis – Applications – Target tracking or detecting
Reexamination Certificate
1999-11-03
2004-08-03
Chang, Jon (Department: 2623)
Image analysis
Applications
Target tracking or detecting
C382S191000
Reexamination Certificate
active
06771798
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates generally to a system for analysis and visualization of hyperspectral data and in particular, the present invention relates to a set of tools for visualization, data acquisition and data utilities for hyperspectral data.
DESCRIPTION OF THE RELATED ART
A hypersensor is a sensor which produces as its output a high dimensional vector or matrix consisting of many separate elements, each of which measures a different attribute of a system or scene under observation. A hyperspectral imaging system, in which the sensed attributes are different wavelength bands, is one example of a hypersensor.
One conventional technique for visualizing the output from a hypersensor is through the use of traditional, image thresholding techniques which implement binary thresholds, where a pixel is either black or white depending on whether it is larger or smaller than the threshold value. An example of a specific currently available system for analysis and visualization of hyperspectral images, is the Environment for Visualizing Images (ENVI), from Research Systems, Inc. The ENVI is a commercial remote sensing program which includes several tools for analysis and visualization of hyperspectral and multispectral data, as well as tools for geolocation and radar image processing. One limitation of ENVI is that it does not provide automatic contrast stretching and scrolling views, which are necessary for continuous data acquisition and display of very long data strips.
In addition to ENVI, other conventional analysis and visualization systems exist. However, these existing systems are designed for post-processing only, i.e., no pre-processing or data acquisition tools are provided, and thus the prior systems are very slow when dealing with data sets, particularly if the sets are multigigabyte in size.
There exists a need for an analysis and visualization system which includes additional features lacking in conventional analysis and visualization systems to provide enhanced visualization and analysis for continuous data acquisition and display of large data, to provide enhanced contrast and scrolling capabilities and to provide pre-processing data acquisition tools.
SUMMARY OF THE INVENTION
It is an object of this invention to provide an improved hyperspectral image analysis and visualization system.
Another object of the present invention is to provide a system for importing collected hyperspectral image data of a scene, identifying targets within that scene, and fitting a shape to the identified targets by shrinkwrapping hyperplanes about the target. Shrinkwrapping is used to select an optional simplex enclosing a set of exemplars determined from an initial screening process. The vertices of this simplex are referred to as endmembers. The role of shrinkwrapping is to select endmembers to represent the scene. Shrinkwrapping is an iterative process in which vertices of the simplex are moved until all data are enclosed by the simplex. In each iteration, faces, i.e., hyperplanes, are moved, i.e., translated and pivoted, until a specified fit parameter is satisfied. Thus, in accordance with the invention, targets are defined in the hyperspectral data of a scene by identifying the target within the scene, fitting a plurality of conjoined hyperplanes or faces about the identified target in the scene; and fitting a shape to the identified target by shrinkwrapping the hyperplanes about the target by translating and rotating the hyperplanes until a specified fit parameter is satisfied.
An additional object of the present invention is to provide a system for enhancing a hyperspectral image, by automatically stretching the image along one or more axes in one or more regions where a target is identified.
A further object of the present invention is to provide a system for enhancing a hyperspectral image by displaying a target on a screen at a higher brightness than the background, such as by displaying the target at full brightness, relative the background, and the background at a selected reduced brightness.
This and additional objects of the invention are accomplished by the structures and processes hereinafter described.
The present invention is adaptable to provide a graphical integration tool for the Intelligent Hypersensor Processing System (“IHPS”) and its output. When so adapted, the present invention provides a framework for rapid contextual visualization of both hyperspectral raw and analyzed data. The framework further serves as a science product integration platform because of its abilities to accept plug-in modules, as well as entire models.
The present invention is further adaptable to provide a unique graphical front end, i.e., interface, configurable to accommodate multiple hyperspectral data analysis algorithms. For example, the present graphical interface can be adapted easily for numerous applications including the commercial satellite industry and other commercial remote sensing endeavors.
The analysis and visualization tools are adaptable to provide additional unique features specific to the exploitation of hyperspectral data. For example, in one specific embodiment, the present invention is adaptable to provide visualization of the output of the Optical Real-time Adaptive Spectral Identification System IHPS/(“ORASIS”) algorithm.
The present invention can also be adapted to accept data in real time from external sources such as from instruments, e.g., sensors, and from online archives.
One advantage of the present system over conventional systems is provided through its adaptation to function in real-time operation. The present system overcomes limitations in existing commercial programs, such as ENVI, which perform some of the operations of the present system. Unlike these conventional systems which require more elaborate procedures and much longer time, which precludes them from near real-time operation, the present system provides features to accommodate near real-time operation.
An additional advantage of the present system is provided through enhanced visualization of a target which is displayed on a darkened background.
Further features and advantages of the present invention will be set forth in, or apparent from, the detailed description of preferred embodiments thereof which follows.
REFERENCES:
patent: 5341439 (1994-08-01), Hsu
patent: 5546472 (1996-08-01), Levin
patent: 5782770 (1998-07-01), Mooradian et al.
patent: 6008492 (1999-12-01), Slater et al.
patent: 6038344 (2000-03-01), Palmadesso et al.
patent: 6075891 (2000-06-01), Burman
patent: 6208752 (2001-03-01), Palmadesso et al.
patent: 6404455 (2002-06-01), Ito et al.
Ren et al., “A Computer-Aided Detection and Classification Method for Concealed Targets in Hyperspectral Imagery.” IEEE Int. Geoscience and Remote Sensing Symposium Proceedings, vol. 2, Jul. 6, 1998, pp. 1016-1018.*
Adams et al. “Simple Models for Complex Natural Surfaces: A Strategy for the Hyperspectral Era of Remote Sensing.” 12thCanadian Symposium on Remote Sensing, IGARSS'89, vol. 1, Jul. 1989, pp. 16-21.*
Yu et al. Automatic Target Detection and Recognition in Multiband Imagery: A Unified ML Detection and Estimation Approach IEEE Trans. on Image Processing, vol. 6, No. 1, Jan. 1997, pp. 143-156.*
Robinson et al. “Comparison of Hyperspectral Sub-Pixel Detection With and Without A Priori Knowledge of Target Features.” IEEE Aerospace Conference, vol. 5, Mar. 21, 1998, pp.183-189.*
Chang et al. “Least Squares Subspace Projection Approach to Mixed Pixel Classification for Hyperspectral Images.” IEEE Trans. on Geoscience and Remote Sensing, vol. 36, No. 3, May 1998, pp. 898-912.
Antoniades John A.
Baumback Mark M.
Bowles Jeffrey H.
Fisher John
Grossman John M.
Chang Jon
Karasek John J.
Legg L. George
The United States of America as represented by the Secretary of
LandOfFree
Hyperspectral visualization extensible workbench does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Hyperspectral visualization extensible workbench, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Hyperspectral visualization extensible workbench will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3340974