An innovative chain coding mechanism for information processing and compression using a virtual bat-bug agent-based modeling simulation

The continuous changes in the size of data create new challenges to design new techniques to reduce its size and encode it in a way that changes its original representation. In this article, we develop a bat-bug agent-based modeling simulation for chain coding and employ it in compressing bi-level image information. The system consists of agents that are classified into static and dynamic depending on their movements. Bugs are considered static agents, and they are distributed over the virtual environment according to the allocation of pixels in the original image. On the other hand, bats are dynamic agent, and their role is to move around to consume bugs while the algorithm tracks their movements. Bats are designed in a way to move within certain boundaries to avoid crashing into each other. Bats employ specific movements that allow them to move in relative directions. Therefore, the frequency of their movements can follow a certain pattern that can help in further size reduction. In other words, the integration of relative movements into our design proved to be advantageous because there is an observable pattern of repeated movements, which allows getting higher compression results. Finally, arithmetic coding is applied to the final strings that represent the movements of bats while searching for bugs to eat. To assess the performance of the algorithm, we compared the findings against standardized benchmarks used in the image processing community: G3, G4, JBIG1, and JBIG2. The outcomes show that we could outperform all these benchmarks using all the images we used for testing. Additionally, we conducted a series of paired samples t-tests, and they revealed that the mean differences between our results and those obtained from other benchmarks are statistically significant.