Resumen

MORALES, Jorge  y  MORALES, Juan. Artificial Vision System for Failure Detection In Finished Fabric. RISTI [online]. 2025, n.58, pp.67-80.  Epub 30-Jun-2025. ISSN 1646-9895.  https://doi.org/10.17013/risti.58.67-80.

Fault detection in finished fabric is carried out through visual inspection by a person to determine the direct impact of various defects on parts of the fabric and thus determine its quality, referencing the international technical standard ASTM D 5430-07. This study proposes an Artificial Vision system for detecting the three main faults that can occur in fabrics: dirty thread, fabric drop, and stains. Using cameras, the system detects these faults in real-time through detection algorithms employing the Python program with cv2 libraries. Once a fault in the fabric is detected, the system displays the location of the anomaly on a screen. To validate the detection of faults, the comparison is made with the methods: LBP, HAAR and HOG, giving the best result for the detection of anomalies the LBP.

Palabras clave : Artificial Vision; Fabric; LBP; HAAR; HOG.

        · resumen en Español     · texto en Español     · Español ( pdf )