Performance evaluation of SIFT against common image deformations on iban plaited mat motif images

Decorative plaited mat is one of the many examples rich plait work often seen on Borneo handicraft products. The mats are decorated with simple and complex motif designs; each has its own special meaning taboos. designs used as a reflection environment traditional beliefs in Iban community. In line efforts from UNESCO’s Sarawak Government’s, digitization, use IR4.0 technologies to preserve promote this cultural heritage encouraged. Towards end goal, we present novel image dataset containing 10 classes. motifs made diagonal symmetrical shapes, well geometric non-geometric patterns. Classification’s accuracy using scale-invariant feature transform (SIFT) features was evaluated against 6 common deformations: zoom+rotation, viewpoint, blur, JPEG compression, scale illumination, across multiple threshold values. Varying degrees deformation were applied digitally cleaned (and cropped) class. We RANSAC remove outliers noisy SIFT matching result. optimal value 2.0e-2 reported 100.0% for change zoom+rotation set.