A Hypothetical Agent-Based Model Inspired by the Abstraction of Solitary Behavior in Tigers and Its Employment as a Chain Code for Compression

Abstract

In this paper, we design an agent-based modeling simulation that represents the solitary behavior in tigers and utilizes it in encoding image information. Our model mainly depends on converting the digital data to a virtual environment with paths classified based on the allocation of the data in the original image. Then, we introduce virtual tigers to the environment to begin the encoding process. Tiger agents are separated from each other, and the algorithm monitors their movements and keeps them away from each other. This separation in the virtual environment allows tigers to encode information that exists in different image parts. Additionally, tigers follow a relative movement style that encodes each tiger’s movement direction based on the previous one. This encoding approach allows particular movements that occur in different directions to be encoded in a similar way. After that, we apply Huffman coding to the chain of movements, the purpose of which is to reduce the size and have a new representation. The experimental findings reveal that we could obtain better results than leading standards in bi-level image compression, including JBIG family methods. Our outcomes strengthen the findings of previous studies that incorporated biological behaviors within agent-based modeling simulations and provide a new abstraction to be utilized in information processing research.