Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Color Quantization Based on Multiple Look-Up Table Reference

IP.com Disclosure Number: IPCOM000036194D
Original Publication Date: 1989-Sep-01
Included in the Prior Art Database: 2005-Jan-28
Document File: 3 page(s) / 27K

Publishing Venue

IBM

Related People

Doi, A: AUTHOR [+3]

Abstract

Disclosed is a device for rapidly quantizing colored images for the raster type of graphics devices as illustrated in the figure. The device creates "multiple look-up tables" so that the errors caused by the color quantization may cancel out each other and refers these multiple look-up tables by turns or randomly in loading the images into the display buffer.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 52% of the total text.

Page 1 of 3

Color Quantization Based on Multiple Look-Up Table Reference

Disclosed is a device for rapidly quantizing colored images for the raster type of graphics devices as illustrated in the figure. The device creates "multiple look- up tables" so that the errors caused by the color quantization may cancel out each other and refers these multiple look-up tables by turns or randomly in loading the images into the display buffer.

Any graphics workstations and personal computers require representing the color of each pixel of the images by some discrete value in the display buffer. This discrete value is called the index. The index of the display buffer is converted through a Look-Up Table into the digital signal of red, green, and blue. Each digital signal is converted into an analog signal for CRT display by the Digital-to-Analog Converter. Popular types of graphics devices allow a small size for the index space, such as 256 colors (8 bits) and so on. In the case of such a small index space, the conventional treatment largely decreases the quality of colored images. As an example, it creates undesirable stripes at the places where the color changes gradually. The method described here keeps the visual quality of the colored images as follows.

(1) The user creates his color image in a large virtual index space which is the product color space of red, green and blue. For an example, 16*16*16 colors where 16 is the intensity level for red, green and blue. The error diffusion method is applied in this stage if it is necessary.

(2) The color-quantization system counts the color frequency (pixel numbers belonging to particular colors) of the user's image and selects the representative colors according to the size of the real index space. Here, the system selects the representative colors so that they may reproduce exactly the real color in most of pixels but may distribute roughly equally in the virtual index space. To do this, a part of the representative colors...