Index on fabric wrinkle using image processing technique

Image processing has been gradually growing on demand as a technique for evaluating and analyzing the gradation of fabric wrinkle. In this work, 150 samples of the Wrinkle Recovery Replica was captured digitally, processed and statistically analyzed to produce respective ranking index based on the ratio of black and white pixel elements, firstly introduced as Wrinkle Pixel Index (WPI_hz). There are five classes of fabric wrinkle gradation generated (WPI_hz-1, WPI_hz-2,WPI_hz-3,WPI_hz-4 and WPI_hz-5) which are based on group clustering of the acquired measured indices during image processing. In this experiment, the scope is limited to produce wrinkle index representing only for one tone color of wrinkled fabric samples. Finally, validation is done by comparing the experimental results with the ones that were already being classified by an expert observer prior before doing the image processing. The outcome has shown that this novice technique has achieved similarity result of 86.67% with the expert´s classification.