Facsimile and static presentation processing – Static presentation processing – Attribute control
Reexamination Certificate
1999-03-11
2001-04-03
Rogers, Scott (Department: 2724)
Facsimile and static presentation processing
Static presentation processing
Attribute control
C382S156000, C382S157000, C382S162000, C382S167000, C382S274000, C345S156000
Reexamination Certificate
active
06211971
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates generally to image processing techniques. In particular, the invention relates to techniques for gathering image data at N different frequencies or frequency bands and mapping the non-redundant image data into a format suitable for display on a color display screen or printer using a finite number of colors, such as combinations of red, green and blue or on a black and white display or printer as gray-scales.
BACKGROUND OF THE INVENTION
Image processing is commonly used to acquire, manipulate, store and display images. It has many useful applications, which illustratively include the areas of photographic processing, photographic restoration and enhancement, medical imaging using x-ray and other frequencies of interest, and satellite imaging for capturing features on the earth's surface, to name a few. The images acquired are typically digitized into imaging data prior to processing, either directly through the use of a digital camera or indirectly by scanning a pre-existing image.
Imaging data generally consists of an array of pixels, where each pixel represents a position on the image and a spectral profile of the position. The spectral profile is represented by an N-dimensional vector having N spectral intensities. For many applications, N is 3 and each of the three spectral intensities represents a respective intensity in the visible light portion of the spectrum, such as red, green or blue color. Combinations of red, green and blue intensities may be perceived as a large range of colors in the visible spectrum. For other applications, N may be greater or less than 3 and the respective spectral intensities may be measured within and outside of the visible portion of the electromagnetic spectrum. For example, in the case of the TMS and MAMS imaging sensors, N is 11 and includes frequencies in the infra-red portion of the electromagnetic spectrum. In the case of AVIRIS imaging data, N is 224 and includes a wide range of frequencies. AVIRIS and similar sensor data may include sufficient spectral information on the materials represented in the imaging data to allow classification of the materials.
After acquisition, imaging data may be modified to a particular display format, such a monitor or printer. The display format is generally pixel based, where each pixel represents a position on the image and includes an M-dimensional vector representing M color intensities. For color renderings, M is typically 3, with each respective color intensity being red, green or blue. For black and white renderings, M is typically one.
After an image is acquired in N-dimensional imaging data, it may be desirable to improve or enhance contrast within the image for rendering in an M-dimensional display format. For example, a photograph or other image may have poor contrast through over-exposure, under-exposure, uneven exposure, age or fading. It would be desirable to acquire the image, for example by scanning it, or its negative, in 3 dimensional red, green and blue space (N=3) and to enhance the contrast for rendering on a display or printer also having three color dimensions (M=3).
It would also be desirable to reveal and enhance contrasts present in N-dimensional imaging data in an M-dimensional display space, where N>M. This would be used to reveal contrasts between objects that are not apparent in visible light to the naked eye. For example, from an aerial view, different trees within a group of trees may appear indistinguishable in the visible spectrum because they are a similar shade of green. However, the different trees may reflect and/or emit radiation from non-visible portions of the electromagnetic spectrum differently. By choosing N spectral bands for gathering the imaging data from the trees, where at least one of the N bands is outside of the visible spectrum and is reflected differently by the trees, contrast within the gathered imaging data will exist, which can be enhanced. Choosing N bands, including bands outside of the visible spectrum, is also useful, for example, for revealing contrasts between crops that are healthy (sufficiently watered and fertilized) and those that are not, and for revealing contrasts between the boundaries of buildings, both of which may be difficult if not impossible to discern in the visible spectrum. Once contrasts are present in the acquired N-dimensional imaging data, it may be desirable to reveal or enhance those contrasts for display in an M-dimensional display format.
Conventional techniques for revealing contrasts on an M-dimensional display based on N-dimensional imaging data, where N>M include: (1) choosing M bands out of the N available bands and displaying these chosen bands in the M-dimensional space; (2) using principal component analysis of the N bands and displaying only the M most important components in the M-dimensional display format; and (3) using an ad hoc linear combination of the N bands to create imaging data in the M-dimensional display format. Each of these techniques, however, results in loss of data. In the case of technique (1), N minus M spectral bands are discarded. Therefore, a final image rendered using technique (1) may lack significant details that are present in the underlying N-dimensional imaging data. Techniques (2) and (3) may salvage more of the underlying N-dimensional imaging data than technique (1), however, as conventionally used, do not maximize use of the N-dimensional imaging data.
It would be desirable to enhance contrasts within images by making use of all non-redundant imaging data. It would further be desirable to maximize the contrast between portions of an image that appear the same color in visible light but different in at least a portion of the electromagnetic spectrum, where N is greater than or equal to M. It would further be desirable to depict objects using both visible light information and information from other portions of the electromagnetic spectrum in a way that maximizes the contrast between different objects but which nonetheless preserves as much as possible the way the objects appear in the visible portion of the electromagnetic spectrum.
SUMMARY OF THE INVENTION
According to the present invention, a method and apparatus enhance visible contrast within an acquired image for display. The method and apparatus utilize all N bands of the original N-dimensional image to produce an M-dimensional enhanced image for display.
In one embodiment of the invention, a method creates an enhanced image from an original image to improve visible contrast in the original image. The original image includes pixels, each pixel having N spectral intensities. In an assigning step, the method assigns N-dimensional prototypes based on the plurality of pixels. Then each N-dimensional prototype is related with an M-dimensional prototype. The M-dimensional prototypes define a palette of colors which are used to represent the enhanced image for display. The M-dimensional prototypes are then spread through a dynamic range in M dimensions. This step enhances contrast by spreading the “color palette” over the range of permissible colors for display. Then, in a creating step, an enhanced image is created based on the N spectral intensities of each pixel, the N-dimensional prototypes and the related M-dimensional prototypes. The enhanced image includes M color dimensions where N may be either greater than or equal to M.
After creating the enhanced image, the enhanced image may be displayed on a monitor, printed, stored in a database or transmitted over a network. In a preferred embodiment of the invention, the assigning step is performed using Vector Quantization. Also in a preferred embodiment, the spreading step spreads the M-dimensional prototypes to achieve substantial histogram equalization in M dimensions. The spreading step is amenable to several implementations, including input training using a general regression neural network (IT-GRNN), input training using a back propagation neural network (IT-BPNN), and an auto associative n
Lockheed Martin Missiles & Space Co.
Rogers Scott
Swidler Berlin Shereff & Friedman, LLP
LandOfFree
Method and apparatus to compress multi-spectral images to a... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus to compress multi-spectral images to a..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus to compress multi-spectral images to a... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2488723