I used spamassasin which then forwards all my mail to gmail account, dont get but 1-5 pieces a day now, but over 12k spam (within month) in spam box, and 11k mail in trash (my own spam settings)
True Color in Real Time: The Challenge of Mobile Imaging
[Extracted from J. Luo, A. Singhal, G. Braun, R. T. Gray, O. Seignol, and N. Touchard; "Displaying images on mobile devices: Capabilities, issues, and solutions"; Special Session on Wireless Imaging, Proceedings of 2002 International Conference on Image Processing, September 22-25, Rochester, NY. http://icip2002.com]
It seems so simple and so obvious: a camera in your device let's you take a picture. Then you send it - just like a text message or e-mail. Wireless imaging, using your cell phone or other hand-held device to send, receive, and view images, is everywhere in advertising. It's the latest manifestation of the convergence of imaging and information technologies. And it's fraught with challenges.
The major technical challenges are obvious:
Connection bandwidth
Coverage area
Error correction
Power supply
Multiple standards
Rendering/display
Image management
Ease of use
No so obvious is the combination of imaging and color technologies that must be embedded in these devices, affecting the immediate perception of image quality. Many of these devices simply cannot display all the colors in an input image, because the input image must be stored in a memory buffer with a reduced bit-depth. In addition, representing an image at a reduced bit-depth can reduce the amount of bandwidth needed for transmission or the amount of memory needed to store an image - both useful in mobile communications.
Just over a decade ago, many computers used an 8-bit color palette to store an image that was to be displayed on screen. Such representations allow only 256 unique color values. This is significantly less than the 16,777,216 possible color values associated with a typical 24-bit color image. This problem has attracted renewed interest with the recent boom of cell phones and \personal digital assistants. The problem is even more acute for these hand-held devices, owing to their severely limited display size. Tables 1 & 2 present a summary of some of the popular handheld devices (PDAs and cell-phones) along with their OS, display resolution, color bit-depth, and other characteristics.
Dithering & Palettization
Color palettization refers to the conversion of an image containing a larger set of possible colors to an image with a different (perhaps reduced) set of palette colors. For example, a typical 24-bit input image could have millions of colors, whereas a typical 8-bit color palette has only 256 colors. It is most desirable to determine the set of palette colors based on the distribution of colors in the input image. Furthermore, preserving important colors, such as human skin tones, in the palettized image is critical.
One common, simple color palette ("websafe" palette) sets six levels of quantization in each of the three color channels (red, green, and blue). Other fixed color palettes have also been developed. Using any fixed color palette will result in an image that has quantization errors that can produce visible contours in the image. In addition, a specific image may not contain colors in all parts of the color space. As a result, there may be some of the 256 color values that never get used, and the effects of quantization errors are more visible.
One way to limit the visibility of the quantization errors is to use a multi-level halftoning algorithm to preserve the local mean of the color value [1]. Since halftoning essentially trades spatial resolution for bit-depth, it is typically not suitable for very small displays, although it is a viable option for PDAs with larger screens (e.g., 8-bit 320 x 240).
An alternative approach is to select the palette of colors on the basis of the distribution of color values in the specific image. This approach avoids wasting color values that will never be used to represent that particular image. One such method is vector quantization (VQ), which involves the selection of an initial color palette, followed by an iterative refinement scheme [2]. Another VQ method [3] starts with all of the colors of an image and groups colors into clusters by merging one nearest neighbor pair of clusters at a time until the number of clusters equals the desired number of palette colors. A third class of VQ methods uses splitting techniques to divide the color space into smaller sub-regions and selects a representative palette color from each sub-region [4]. In general, splitting techniques are computationally more efficient than either the iterative or merging techniques and can provide a structure to the color space that enables efficient pixel mapping at the output stage.
While VQ mechanisms can yield more visually appealing images, they are computationally intensive. A sequential scalar quantization (SSQ) method, proposed by Allebach et al. [4], partitions a histogram of the distribution of the input colors into a number of sub-regions or color space cells, such that each partitioned color cell is associated with a color in the output color palette. This approach is more efficient computationally than vector quantization schemes. We chose to use SSQ as the underlying color palettization method in our image-rendering scheme.
Image-dependent palettization methods offer significant advantage in that they assign palette colors based on the distribution of input colors, choosing colors that represent the most commonly occurring hues in a particular image. This approach reduces the average quantization errors throughout the image, although there may still be large quantization errors in important image regions. For example, consider an image containing a human face that only occupies a small image area. The number of pixels that represent skin-tone colors may be relatively small, and therefore, the likelihood that adequate palette colors are assigned to skin-tone colors is low. As a result, when the image is represented by the set of chosen palette colors, there may be objectionable colors or contours in the face. Because this image region may be very important to an observer, these artifacts may be much more objectionable than they would have been if they had occurred elsewhere in the image. Existing techniques do not provide any mechanism for minimizing the quantization artifacts in these important regions unless they are large enough to comprise a significant portion of the color distribution.
Recently, Kodak image scientists developed a way to meet these goals by supplementing the distribution of the input colors with a distribution of selected "important" colors [5]. In particular, they found they could supplement skin tones by appending image skin-tone patches generated from a statistical sampling of the skin color probability density function. A major advantage of this approach is that explicit skin detection, which can be error prone, is avoided. In addition, this approach is useful with any color palettization scheme. Subjective evaluation has shown the efficacy of this scheme. Once the set of palette colors is determined, a palettized color image can be generated. Generally, the palette color for each pixel of the image will be identified by an index value indicating which palette color should be used for that pixel. For example, if there are 256 palette colors used for a particular image, each pixel of the output image can be represented by an 8-bit number, i.e., palette index, in the range 0-255. When the image is displayed, the palette index can be used to determine the corresponding color value (red, green, and blue) for each of the palette colors.
There are a few key observations based on the above study. First, sequential scalar quantization is effective (and extremely fast); the custom 240-color palette by SSQ (16 Windows system colors are preserved) handsomely beats a fixed "Web Safe" palette (6 quantization levels in each channel). Second, the introduction of supplementary skin colors leads to significantly better rendition of human skin areas without, in general, adversely affecting other areas. The latter is mostly because each supplemented skin color patch is small enough to not claim a color in the palette unless it also occurs in the image. The size of individual patches scales with the size of the input image while the total number of patches remains constant.
References
[1] R. S. Gentile, E. Walowit, and J. P. Allebach, "Quantization and multilevel halftoning of color images for near original image quality," J. Opt. Soc. Am. A 7, 1019-
1026, 1990.
[2] R. S. Gentile, J. P. Allebach, and E. Walowit, "Quantization of color images based on uniform color spaces," J. Imaging Technol. 16, 11-21, 1990.
[3] R. Balasubramanian et al., "A new approach to palette selection for color images," J. Imaging Technol. 17, 284-290, 1991.
[4] J. P. Allebach et al., "Sequential product code quantization of digital color image," U.S. Patent 5,544,284.
[5] J. Luo, K. Spaulding, and Q. Yu, "A novel color palettization scheme for preserving important colors," submitted to SPIE Conference on Electronic Imaging, 2003.
[6] A. D. Cropper, R. S. Cok, and R. D. Feldman, "Organic LED System and Applications", SPIE 4105, 19-29, 2000.
Slashdot Mirror
ok i will just buy red hat then kkthxbye
EOM
thats right!
fp
I used spamassasin which then forwards all my mail to gmail account, dont get but 1-5 pieces a day now, but over 12k spam (within month) in spam box, and 11k mail in trash (my own spam settings)
oh yeah
Pay royalties?? HAHAHAH NEVER!!! The net is phree!! Keep it like tha' warez??? hahaha goatse for president
So cool.. not So new... not Lame poster.. yes
The /. is never far
hey ASSHOLE your second fucking post.. MORON
I will never get it. FUCK SHIT CUNT
cause i poop on this interview
True Color in Real Time: The Challenge of Mobile Imaging [Extracted from J. Luo, A. Singhal, G. Braun, R. T. Gray, O. Seignol, and N. Touchard; "Displaying images on mobile devices: Capabilities, issues, and solutions"; Special Session on Wireless Imaging, Proceedings of 2002 International Conference on Image Processing, September 22-25, Rochester, NY. http://icip2002.com] It seems so simple and so obvious: a camera in your device let's you take a picture. Then you send it - just like a text message or e-mail. Wireless imaging, using your cell phone or other hand-held device to send, receive, and view images, is everywhere in advertising. It's the latest manifestation of the convergence of imaging and information technologies. And it's fraught with challenges. The major technical challenges are obvious: Connection bandwidth Coverage area Error correction Power supply Multiple standards Rendering/display Image management Ease of use No so obvious is the combination of imaging and color technologies that must be embedded in these devices, affecting the immediate perception of image quality. Many of these devices simply cannot display all the colors in an input image, because the input image must be stored in a memory buffer with a reduced bit-depth. In addition, representing an image at a reduced bit-depth can reduce the amount of bandwidth needed for transmission or the amount of memory needed to store an image - both useful in mobile communications. Just over a decade ago, many computers used an 8-bit color palette to store an image that was to be displayed on screen. Such representations allow only 256 unique color values. This is significantly less than the 16,777,216 possible color values associated with a typical 24-bit color image. This problem has attracted renewed interest with the recent boom of cell phones and \personal digital assistants. The problem is even more acute for these hand-held devices, owing to their severely limited display size. Tables 1 & 2 present a summary of some of the popular handheld devices (PDAs and cell-phones) along with their OS, display resolution, color bit-depth, and other characteristics. Dithering & Palettization Color palettization refers to the conversion of an image containing a larger set of possible colors to an image with a different (perhaps reduced) set of palette colors. For example, a typical 24-bit input image could have millions of colors, whereas a typical 8-bit color palette has only 256 colors. It is most desirable to determine the set of palette colors based on the distribution of colors in the input image. Furthermore, preserving important colors, such as human skin tones, in the palettized image is critical. One common, simple color palette ("websafe" palette) sets six levels of quantization in each of the three color channels (red, green, and blue). Other fixed color palettes have also been developed. Using any fixed color palette will result in an image that has quantization errors that can produce visible contours in the image. In addition, a specific image may not contain colors in all parts of the color space. As a result, there may be some of the 256 color values that never get used, and the effects of quantization errors are more visible. One way to limit the visibility of the quantization errors is to use a multi-level halftoning algorithm to preserve the local mean of the color value [1]. Since halftoning essentially trades spatial resolution for bit-depth, it is typically not suitable for very small displays, although it is a viable option for PDAs with larger screens (e.g., 8-bit 320 x 240). An alternative approach is to select the palette of colors on the basis of the distribution of color values in the specific image. This approach avoids wasting color values that will never be used to represent that particular image. One such method is vector quantization (VQ), which involves the selection of an initial color palette, followed by an iterative refinement scheme [2]. Another VQ method [3] starts with all of the colors of an image and groups colors into clusters by merging one nearest neighbor pair of clusters at a time until the number of clusters equals the desired number of palette colors. A third class of VQ methods uses splitting techniques to divide the color space into smaller sub-regions and selects a representative palette color from each sub-region [4]. In general, splitting techniques are computationally more efficient than either the iterative or merging techniques and can provide a structure to the color space that enables efficient pixel mapping at the output stage. While VQ mechanisms can yield more visually appealing images, they are computationally intensive. A sequential scalar quantization (SSQ) method, proposed by Allebach et al. [4], partitions a histogram of the distribution of the input colors into a number of sub-regions or color space cells, such that each partitioned color cell is associated with a color in the output color palette. This approach is more efficient computationally than vector quantization schemes. We chose to use SSQ as the underlying color palettization method in our image-rendering scheme. Image-dependent palettization methods offer significant advantage in that they assign palette colors based on the distribution of input colors, choosing colors that represent the most commonly occurring hues in a particular image. This approach reduces the average quantization errors throughout the image, although there may still be large quantization errors in important image regions. For example, consider an image containing a human face that only occupies a small image area. The number of pixels that represent skin-tone colors may be relatively small, and therefore, the likelihood that adequate palette colors are assigned to skin-tone colors is low. As a result, when the image is represented by the set of chosen palette colors, there may be objectionable colors or contours in the face. Because this image region may be very important to an observer, these artifacts may be much more objectionable than they would have been if they had occurred elsewhere in the image. Existing techniques do not provide any mechanism for minimizing the quantization artifacts in these important regions unless they are large enough to comprise a significant portion of the color distribution. Recently, Kodak image scientists developed a way to meet these goals by supplementing the distribution of the input colors with a distribution of selected "important" colors [5]. In particular, they found they could supplement skin tones by appending image skin-tone patches generated from a statistical sampling of the skin color probability density function. A major advantage of this approach is that explicit skin detection, which can be error prone, is avoided. In addition, this approach is useful with any color palettization scheme. Subjective evaluation has shown the efficacy of this scheme. Once the set of palette colors is determined, a palettized color image can be generated. Generally, the palette color for each pixel of the image will be identified by an index value indicating which palette color should be used for that pixel. For example, if there are 256 palette colors used for a particular image, each pixel of the output image can be represented by an 8-bit number, i.e., palette index, in the range 0-255. When the image is displayed, the palette index can be used to determine the corresponding color value (red, green, and blue) for each of the palette colors. There are a few key observations based on the above study. First, sequential scalar quantization is effective (and extremely fast); the custom 240-color palette by SSQ (16 Windows system colors are preserved) handsomely beats a fixed "Web Safe" palette (6 quantization levels in each channel). Second, the introduction of supplementary skin colors leads to significantly better rendition of human skin areas without, in general, adversely affecting other areas. The latter is mostly because each supplemented skin color patch is small enough to not claim a color in the palette unless it also occurs in the image. The size of individual patches scales with the size of the input image while the total number of patches remains constant. References [1] R. S. Gentile, E. Walowit, and J. P. Allebach, "Quantization and multilevel halftoning of color images for near original image quality," J. Opt. Soc. Am. A 7, 1019- 1026, 1990. [2] R. S. Gentile, J. P. Allebach, and E. Walowit, "Quantization of color images based on uniform color spaces," J. Imaging Technol. 16, 11-21, 1990. [3] R. Balasubramanian et al., "A new approach to palette selection for color images," J. Imaging Technol. 17, 284-290, 1991. [4] J. P. Allebach et al., "Sequential product code quantization of digital color image," U.S. Patent 5,544,284. [5] J. Luo, K. Spaulding, and Q. Yu, "A novel color palettization scheme for preserving important colors," submitted to SPIE Conference on Electronic Imaging, 2003. [6] A. D. Cropper, R. S. Cok, and R. D. Feldman, "Organic LED System and Applications", SPIE 4105, 19-29, 2000.
Enough of this Red Hat, linux, gnome sheeeeeeeeeet. WinXP for corporate users is what its all about!!!
Come on idiots........ use the friggin newsgroups, verizon even gives you a free one with DSL
I didnt even have a chance to read the article!!!!
Why thank you!!!! Mods.. Hook this man up with some points!!!
is good to poop on
good for me to poop on!
Bravo fox network for a great find!!Keep those quality shows ROLLIN!
for me to poop on!!!
ewwwwwwwwwwwwww yeahhhhh
but i had to kill her
I used to lover her, ohhh hyeah, but I had to killl her!
Hell yeah!! Can i get a witness!
Who let the dogs out????? what was this article about again?????
you still smell
I didnt waste my time reading that article