Automatic Detection of Defects on Birch Wood Boards

Abstract

In a factory which produces veneer boards from birch wood, grading of the boards into different quality categories is usually part of the production process. To improve the efficiency of grading, attempts are being made to automate it using automated visual inspection (AVI). Integral to the process of inspection is segmentation. Segmentation is the part of the AVI process concerned with separating clear wood and defective areas in the image. This paper describes a system that is used to segment the images of birch wood boards. The system consists of four modules which are called global adaptive thresholding, multi-level thresholding, row-by-row adaptive thresholding and vertical profiling. The paper gives details of the four modules and presents the results obtained in segmenting images of a large sample of birch wood boards.