Material quality assessment of silf fibroin nonofibers based on swarm intelligence MACHADO, B. B. ; GONÇALVES, W. N. ; BRUNO, O. M. International Conference on Mathematical Modeling in Physical Sciences, 2012, p. 241.
Nanofibers have been widely used in biomedical applications due to their high capacity to regenerate bones and tissues. However, evaluating the suitability of a new material relies strongly on its physical surface properties, often produced with irregular structures. In this paper we present the mathematical modeling of nanofibers by using a swarm intelligent system. The non-regularity on nanofibers leads several traditional methods to fail, especially those based purely on quantitative statistical measures. Here, we propose a novel method for texture analysis based on the artificial crawler model. A population of artificial crawlers is initially considered and then employed on the image. Each image is mapped into a 3D surface, which the z-axis corresponds to the gray values. As feature vector, we compute measures extracted of: (i) the evolution process, characterized by the energy of each alive artificial crawler, and (ii) the interaction strategy of the walking of artificial crawlers in higher and lower intensities. Experiments are evaluated on images of silk fibroin fibers with different concentrations of glycerol, from 2.5% up to 10%, providing worse or better mechanical properties. Experimental results reveal that the proposed approach outperforms the state-of-the-art method. The results support the idea that our method can be successfully used for quality assessment of other nanomaterials.