YIQ vector quantization in a new color palette architecture.

Area of Science:

• Computer Graphics
• Image Processing
• Data Compression

Background:

• Color palettes are used in CRT displays to conserve frame buffer memory for real-time graphics.
• Current color quantization algorithms cluster colors as 3-D vectors, failing to exploit image space redundancies.

Purpose of the Study:

• To develop a novel vector quantization (VQ) technique for color palettes that exploits both image and color space redundancies.
• To enhance frame buffer memory efficiency and color palette flexibility in graphics and imaging operations.

Main Methods:

• Proposed a simple product vector quantization (VQ) technique.
• Utilized the NTSC YIQ color space instead of RGB.
• Modified the color palette architecture to accommodate the YIQ-based product VQ algorithm.

Main Results:

• Reduced frame buffer memory usage for color images by half compared to existing algorithms.
• Achieved comparable image quality with the new VQ technique.
• Created a modified color palette that is nearly ten times smaller and more flexible than current RGB palettes.

Conclusions:

• The proposed product VQ technique offers significant improvements in space efficiency for both frame buffers and color palettes.
• The method achieves these gains without complicating frame buffer architecture or increasing computational complexity.
• This approach enables more efficient use of memory resources in real-time graphics and imaging.