Comparative study of integral image and normalized cross-correlation methods for defect detection on Batik klowong fabric

In 2009, UNESCO designated Batik as an essential element of Indonesia's cultural heritage, underscoring its cultural significance. Among the various Batik techniques, Batik tulis klowong, which employs wax outlines to form intricate motifs, is critical for preserving the authenticity and quality of traditional Batik. However, a major research gap exists in the form of inefficient and inconsistent manual defect detection, which hampers quality control in Batik production. This limitation highlights the need for a more reliable and automated inspection system, motivating the current study to develop a solution that addresses these challenges. This study presents an automated defect detection system utilizing normalized cross-correlation (NCC) and integral image techniques combined with adaptive thresholding. Focusing on the red color channel, the system's performance is evaluated using false positive rate (FPR), sensitivity, and accuracy. The results demonstrate that the integral image method surpasses NCC, achieving an FPR of 1.92, sensitivity of 86.18, and accuracy of 96.82, while significantly reducing processing time to 1.019 seconds compared to NCC's 6.116 seconds.These findings indicate that the integral image approach provides a more efficient and accurate solution for Batik tulis klowong quality assessment, benefiting the Batik industry by enhancing production processes, lowering operational costs, and contributing to the preservation of Indonesia's cultural heritage.