Prediction and animation of dynamical behavior of color diffusion in water using 2-D tightly coupled neural cellular network

Unreliable accuracy and costly computation used to be major problems when the technique of differential equations was implemented for predicting the behavior of a dynamical system. But now these constraints have been successfully overcome by introducing 2-D tightly coupled neural cellular network to learn about the diffusion characteristics of colored liquid dropped onto water surface. Afterwards, it is allowed to predict the diffusion characteristic by generating its own diffusion image. Finally the generated image is compared with the actual one in term of value of cosine, and the result indicates that the accuracy of prediction is higher than 90 percent. Furthermore, the creation of animation also requires significantly lower computational cost compared to the recent methods.