Identifying and Locating Surface Defects in Wood: Part of an Automated Lumber Processing System

Study of the Effect of Lighting Technology in Texture Classification Systems

“…The most popular ones are those known as Haralick features (Haralick et al, 1973), followed by the set of measures introduced by Conners et al (Conners & McMillin, 1983). Obviously, a GLCM cannot be computed directly from a colour image, but from a greyscale one, so a modification is needed to be able to process that kind of images.…”

“…The former methods extract a set of statistical properties from the spatial distribution of intensities of a texture. Common examples of this approach are the histogram method and the family of algorithms based on cooccurrence matrices (Conners & McMillin, 1983;Haralick et al, 1973). The latter techniques, on the other hand, compute a number of features obtained from the analysis of the local spectrum of the texture.…”

Development of an Automated Lumber Processing System

“…The problern of identifying defective areas of wood in hardwood lumber is very complex due to the presence of grain in the wood and the variability of the overall tone and texture of hardwood across species and even in the same species. Previous attempts [7,8] towards a vision system for ALPS, have employed a controlled environment to minimize these effects. A board is.viewed as several small disjoint rectangles and statistical measures (such as tone and texture of wood) are then used to classify each of these disjoint rectangles into one ofthe categories viz.…”

Real-Time Vision-Based System for Textile Fabric Inspection

“…The problem of web inspection is particularly important and complex, and the research in this field is wide open [3]. Web inspection systems are currently used for quality control in numerous production lines such as for cloths and fabrics [4][5][6][7][8][9][10][11][12][13][14][15], cable insulators [16], paper [17], plastic bags [18], strip steel and metals [19][20][21], wood [22,23] and leather [24].…”