<<HTML(<meta http-equiv="refresh" content="0;URL=../../CMV/Research/VisualInspection">)>>
<<HTML(<a href="#" onclick="history.back(); return false;">Back</a>)>>
Visual inspection research provides guidelines to identify attractive computing approaches, architectures, and algorithms for useful commercial systems. In practice, solutions from low-level image processing to even equipment installation and operating procedures are considered simultaneously. The roots of this expertise are in our industrial visual inspection studies in which we met extreme computational requirements already in the early 1980's, and we have contributed to the designs of several industrial systems.
One of the promising new uses for visual inspection is automated strength grading for sawn timber. The developed solution employs real-time feature extraction, classification, and Finite Element Method (FEM) combined into an adaptive learning scheme. The idea is to launch FEM based analysis whenever the region under inspection deviates from the ones modeled earlier. The 3-D model and element mesh needed in the FEM analysis are built from the data acquired during feature extraction, utilizing knowledge of the grain structure of wood. Various FEM analysis methods are surveyed, from almost complete reconstruction of the microstructure of wood to much faster, but less precise material parameter approaches.
Another example of the recent visual inspection application that takes also advantage of the GPU processing is the 3-D image reconstruction software developed in collaboration with VTT. The software constructs a surface map from several images of a sinusoidal patter projected onto a wooden or metallic sample. The method analyzes the displacements between the images, and computes the phase image using an iterative technique. All the required operations of the algorithm are computed in parallel using general purpose computations on GPUs.
Niskanen M & Silven O (2007)
Machine vision based lumber grain measurement.
Proc. IAPR Conference on Machine Vision Applications (MVA 2007), Tokyo, Japan, 408-411.
Silven O, Niskanen M & Kauppinen H (2003)
Wood inspection with non-supervised clustering.
Machine Vision and Applications 13(5-6):275-285.
Turtinen M, Pietikäinen M, Silven O, Mäenpää T & Niskanen M (2003)
Paper characterisation by texture using visualisation-based training.
The International Journal of Advanced Manufacturing Technology 22(11-12):890-898.
Niskanen M, Silvén O & Kauppinen H (2001)
Color and texture based wood inspection with non-supervised clustering.
Proc. 12th Scandinavian Conference on Image Analysis, June 11-14, Bergen, Norway, 336-342.