Stochastic texture analysis for monitoring stochastic processes in industry

Abstract

Several continuous manufacturing processes use stochastic texture images for quality control and monitoring. Large amounts of pictorial data are acquired, providing both important information about the materials produced and about the manufacturing processes involved. However, it is often difficult to measure objectively the similarity among such images, or to discriminate between texture images of materials with distinct properties. The degree of discrimination required by industrial processes sometimes goes beyond the limits of human visual perception. This work presents a new method for multi-resolution stochastic texture analysis, interpretation and discrimination based on the wavelet transform. A multi-resolution distance measure for stochastic textures is proposed, and applications of our method in the non-woven textiles industry are reported. The conclusions include ideas for future work.