Optimising Fractal Encoding Using Cuttlefish Optimisation Algorithm

Abstract

Increase in demand for better appearance and less storage requirement of an image has led to explore various image compression techniques. Due to technological advancement of photo capturing devices such as single-lens reflex camera (SLR), digital SLR, smart phone cameras, and satellite sensors, more detailed information can be recorded in a single image. A coloured image captured by high wavelength sensors produces large-sized image as it contains highly correlated data. Many image compression and analysis techniques have been developed to aid the interpretation of images and to compress as much information as possible in it. The goal of image compression is to recreate original image with less number of bits and minimal data loss. For generating computer graphic images and compression of objects, it has been suggested that by storing images in the form of transformation instead of pixels lead to compression and can be achieved through fractal coding. In fractal image compression, encoding image blocks into fractal codes using iterated function system (IFS) takes large amount of time taken to compress it. A study of various meta-heuristics techniques, which are designed to solve complex problems approximately, has been conducted to improve upon computational time of fractal coding as well as compression ratio, while maintaining image visually. In this paper, using the property of pattern adaption of surroundings, cuttlefish optimisation algorithm is applied to minimise the time taken for fractal coding. Compression results have been compared with other meta-heuristic techniques, such as particle swarm optimisation and genetic algorithm, and has shown high compression ratio of approximately 31%.