Two color quantization algorithms for efficient image compression
Date of Award
Master of Science in Computer Science
Information Systems & Computer Science
Fernandez, Proceso L., Jr., Ph.D.
Digital Images are an ubiquitous aspect of modern-day communication. A vast of information constantly transmitted over the internet consists of image data. Therefore, compressing these images so as to reduce the amount of bandwidth and space required to transmit and process these images is an essential concern. This study presents a two new lossy color quantization image compression techniques. The new techniques both make use of partitioning the color values of the image into a specified number of bins, then replacing each bin element with the mean value of the bin. The techniques are called Uniform Partition Mean-based Color Quantization (UPMCQ) and Optimal Partition Mean-based Color Quantization (OPMCQ). The results of this study show that both algorithms are capable of producing compressed images which have a higher Peak Signal-to-Noise Ratio (PSNR) in comparison to similar algorithms, without compromising compression ratio. The algorithms are efficient and relatively simple to implement, making them viable for use as alternatives to existing compression algorithms.
(2017). Two color quantization algorithms for efficient image compression. Ateneo de Manila University.