Image analysis – Color image processing
Reexamination Certificate
2000-09-28
2004-08-31
Wu, Jingge (Department: 2623)
Image analysis
Color image processing
C358S520000, C348S362000
Reexamination Certificate
active
06785414
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates generally to control of exposure times of color digital cameras.
2. Discussion of the Related Art
The old adage that “a picture is worth a thousand words” has taken on a heightened degree of meaning in the wake of the development of digital image processing technologies. Digital image processing methods enable pictorial data to be manipulated in ways that enhance the means by which humans can interpret this data. Initially, this capability gave rise to an interest in transforming images, stored via traditional media products, into digital formats. Typically, this has been performed by use of an image scanner. More recently, efforts have focused on methods of directly creating digital images. Digital cameras are an end product of these efforts.
Unlike traditional film cameras, digital cameras use a solid-state device called an image sensor to capture the image. Image sensors are semiconductor chips containing a grid of hundreds of thousands to millions of photosensitive diodes called photosites. Each photosite corresponds to a single pixel of the captured image. Light energy entering the digital camera impinges on the image sensor causing the semiconductor material to release electrons that are captured by the photosites. In this way, each photosite records, as an electric charge, the accumulation of light energy that impinges on it. As a photosite is exposed to light for longer periods, increasing amounts of light energy reach it, and a greater electric charge is stored. The closing of the shutter initiates a process whereby the charge at each photosite is measured and converted to a digital representation. All photosites, taken in aggregate, store a digital mapping of the image that can be used to reproduce it.
While recording the amount of light energy that reaches them, photosites do not distinguish this light energy by its different wavelengths. Therefore, photosites are, by themselves, insensitive to color. Digital cameras perceive color by use of color filters placed between the impinging light and the photosites. A color filter passes light energy at or near the wavelength corresponding to the color of the filter and attenuates light energy outside this band. This has the relative effect of favoring the accumulation, on a photosite, of light energy at wavelengths at or near the wavelength corresponding to the color of the filter. This allows digital cameras to leverage a natural phenomenon of visible light. All of the colors of visible light can be realized through various combinations of the three primary colors of red, green, and blue. Therefore, it is possible, by using filters set to these three primary colors, to capture the various hues of an image and reproduce the actual coloring of the scene by reassembling the various filtered components.
Color filtering typically occurs by one of two methods. One process requires taking a series of exposures in which a different colored filter is used with each iteration. These filters can be individual color filters or a composite unit that can be controlled to filter different colors. Composite filters using liquid crystal display technology are commonly used. Light passing through a given filter reaches each of the photosites on the image sensor. This facilitates a high degree of accuracy when the image is reproduced. However, this high degree of accuracy depends on the ability to hold the objects in the image stationary for the time necessary to record the three images and change the filters between each exposure. A modification of this approach involves using a digital camera with three image sensors. A different colored filter is used for each sensor so that all three primary colors of the image are captured in a single exposure. Digital cameras with three image sensors are relatively expensive.
Another alternative method involves affixing an individual filter to each photosite such that the overall image sensor appears as a mosaic. The Bayer pattern is a typical arrangement of photosite filters. Empirically derived to match the perceptive qualities of the human eye, it uses twice as many green filters as blue or red. Using filters affixed to and covering individual photosites allows all colors of an image to be captured in a single exposure. True colors are divined from the three primary colors by use of a comparison algorithm. The degree of accumulated energy recorded at a given photosite is compared with the levels stored at surrounding photosites to determine the true color of the image at the position corresponding to the given photosite. This is the color that is rendered when the image is reproduced. Owing to the large number of photosites, performance of the algorithm requires a considerable amount of processing time.
While this approach allows all colors of an image to be captured in a single exposure, the fidelity of the reproduction suffers from the fact that each primary color is recorded by only a portion of the total number of photosites. Accuracy is also impeded by reliance on the interpolation algorithm to estimate the true color of a given location on the image. Additionally, a certain amount of the accumulated electric charge at a given photosite results from electrons released from the semiconductor material due to thermal energy. The interpolation algorithm can be limited in its ability to correct for variations, among the photosites, in this thermal noise.
Image sensors traditionally have used charge-coupled devices (CCDs) for their photosites. Development of this technology has enabled the production of high quality image sensors. More recently however, image sensors have been made using complementary metal oxide semiconductor (CMOS) technology. With CCD technology, after an image is captured, stored charges are transferred from the photosites row by row to a readout register for subsequent processing by devices on other chips. As charges stored in the row nearest the readout register are transferred to it, charges stored in the next adjacent row are transferred to the row nearest the readout register. Charges stored in all other rows are likewise transferred to their respective adjacent rows nearest the readout register. The charges on each row are “coupled” to those on the adjacent row so that when one row moves, the next also moves to fill the vacated photosites.
As a category of electron devices, CCDs have limited applications. They are fabricated on wafers using relatively expensive processes at foundries that specialize in their production. In contrast, CMOS technology is widely used for chips supporting a variety of electronic products. Therefore, use of CMOS allows manufacturers to enjoy tremendous economies of scale and to reduce their manufacturing expenses substantially. Additionally, development of CMOS processing technologies has realized the manufacture of wafers with high yields of useable chips, further reducing costs. Also, image sensors made from CMOS can include processing circuits on the same chip. Yet, while CMOS technology offers several advantages over CCDs, fabrication of image sensors from CMOS is still in its early stages of development and CMOS imagers do not yet have the quality of CCD sensors.
Digital cameras offer several advantages over other imaging technologies. With a digital camera, one can immediately view the captured image on a screen, transfer the image in a digital format via a host of communications technologies, display the image via a variety of media, edit images with digital processing techniques, save the costs of buying and developing film, reduce the use of toxic chemicals used in traditional photography, and ergonomically configure the design of the camera for use with other equipment, such as a microscope. With digital image processing, images can be cropped to emphasize specific portions, expanded or reduced in size, altered for brightness, blended with other images, or filtered to sharpen or blur outlines of objects or create a special effect. While it is possible
Marek, Jr. Louis F.
McStravick, III Arnold W.
Ryan Kevin W.
Media Cybernetics Inc.
Sterne Kessler Goldstein & Fox P.L.L.C.
Wu Jingge
LandOfFree
System and method for establishing an aggregate degree of... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System and method for establishing an aggregate degree of..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and method for establishing an aggregate degree of... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3280860