Applicability of Table Reasoning Method for Fabric Defect Evaluation

Abstract

Fabric inspectors can recognize the fabric defects because they can observe not only the light and darkness but also the brilliance or gloss on the fabric surface, So, in order to distinguish the fabric defects by machine, it is also necessary to deal with such information as gathered by men. In this paper, a new method was researched to distinguish the fabric defects. We shined laser light on fabric surface and measured the diffused light over 360 degrees to create a reflected pattern.There are many naps and twists on the fabric surface irregularly, so the distribution of diffused light will be changed when the incident angle of laser is changed. If the light is shined on a fabric surface at an appropriate angle to detect the signal, the information about the fabric defect could be taken at its maximum. In our experiment, the most appropriate angle of incidence is 25 degrees and the most appropriate angle of measurement is 75 degrees.By using a pinhole and a filter in the optical path, we normally came to be able to detect the signal even if not in drakroom. And when the pinhole diameter was 0.7mm, we detected the signal with averaging the rippl of the noise of the thread naps and twists, and without losing the defect signal. Here the measured area (S) is given by the following equation. S=πa2/4cosθ Where a is the pinhole diameter, and θ is the angle of measurement.In this experiment, we found that the reflected pattern was dependent on the fabric construction. The normal pattern was symmetric, while abnormal one was distorted and the distortion of the reflected pattern was dependent on the kind of defect. The similar reflected patterns can be detected in the same sort of defect, but the identical one practically does not appear again even in the same sort of defect.Table reasoning method was introduced in order to recognize the fabric defect with reflected pattern. Firstly twelve characteristic parameters were smapled from the reflected pattern to investigate the relations between reflected pattern and fabric defect, then an expert system was built to identify the fabric defects.