Lossless compression of color map images by context tree modeling

Area of Science:

• Computer Science
• Image Processing
• Data Compression

Background:

• Lossless compression of color map images is crucial for efficient storage and transmission.
• Previous methods utilized binary layer separation and context tree models, exploiting interlayer dependencies.
• Directly operating on color values was hypothesized to yield comparable or superior compression.

Purpose of the Study:

• To extend context-tree-based compression methods to operate directly on color values, bypassing layer separation.
• To develop an optimized n-ary context tree model for color image compression.
• To evaluate the performance of the proposed method against existing techniques.

Main Methods:

• An n-ary context tree is generated by constructing a complete tree to a predefined depth.
• Nodes that do not contribute to compression improvements are pruned from the tree.
• The method operates directly on color values, exploiting spatial and inter-color dependencies.

Main Results:

• The proposed method achieves significant lossless compression results for color map images.
• Experimental results demonstrate superior performance compared to existing methods across various color map datasets.
• The direct color value approach effectively captures dependencies without explicit layer separation.

Conclusions:

• Operating directly on color values with an optimized context tree model is an effective strategy for lossless image compression.
• The developed method offers improved compression efficiency for color map images.
• This approach represents a advancement in lossless image compression techniques.