Design and Implementation of the Wood Knot Recognition System Based on Matlab GUI

Abstract

Wood knots are scaring formed by the death of branches in the trunk, which has a great influence on the quality of logs. To accurately detect the shape and area of wood knots is of great significance to the production and processing of wood. Wood images with live knots, dead knots, rotten knots and multiple knots are processed with the gray level transformation, Gaussian filtering which are carried out successively. Otsu algorithm is used for image thresholding, followed by erosion and dilation, remove connected domains, edge markers, number of knots, roundness detection, the ratio of the areas of the knots and whole image. The results show that the contour of knots can be detected accurately after processing. The number of knots, the roundness and area of knots are determined. The algorithms mentioned above are embedded into a Graphical User Interface(GUI) by using Matlab callback functions.