Locating and identifying sound knots and dead knots on sugi by the rule-based color vision system

Abstract

Knowledge of the occurrence of sound and dead knots on the surface of sugi is important for the classification and application of the material. This study examined a color vision system for detecting sound and dead knots on sugi. The system can be conceptually divided into three components: a CCD-camera scanning system, an image-segmenting module, and a rule-based defect identifying module. The results showed that the potential defect regions could be located by Otsu's threshold algorithm in conjunction with t-test analysis. The accuracies of locating sound knots and dead knots were 92.6% and 97.1%, respectively. The rule-based approach was used to identify sound and dead knots and the identifying accuracies for sound knots and dead knots were 92.0% and 94.1%, respectively. The overall detection accuracy of the system was 87.6%. The results indicated that the rule-based color vision system is an efficient means of detecting sound knots and dead knots on sugi.