<title>CT image processing for hardwood log inspection</title>

Abstract

This study explores the application of digital image processing techniques to a machine vision system for log inspection in the forest products industry. This machine vision system uses the Computerized Tomography (CT) imaging to locate and identify internal defects in hardwood logs. To apply CT to these industrial vision problems requires efficient and robust image processing methods. Several image processing techniques are addressed in this paper: adaptive image smoothing, multi-threshold-based segmentation, morphologicaJ filtering, and 3-dimensional connetiveness labeling. Experimental results of these image processing techniques with CT images from two different wood species demonstrate the efficacy of the inspection system.

<title>Stochastic field-based object recognition in computer vision</title>

“…An efficient three-dimensional adaptive filtering method is described in [14] that can eliminate the unwanted ring structures while preserving other important image details, e.g., the presence of small checks. The filtered images are then segmented on an image-by-image basis, producing a number of regions each representing a potential defect such as splits, knots, barks, stains, and decays [14].…”

“…The proposed computer vision system is designed to be independent of wood species, and it is composed of two basic modules: a low level module and a high level module. The low level module performs such tasks as image filtering, segmentation, region detection and merging [14], and the high level module conducts defect recognition. The final product of the proposed vision system would be a data base that describes the defect distribution inside a log.…”

Automated analysis of CT images for the inspection of hardwood logs

“…More recently, Zhu, Conners, and Araman have described a knowledge-based vision system that is capable of locating, identifying and quantifying the internal defects of logs by analyzing CT image data [9,11,12]. The system is composed of three modules: a data acquisition unit, an image segmentation module, and a scene analysis module.…”

“…Several researchers have considered the use of x-ray computed tomography (CT) for this purpose, and have established the feasibility of defect detection using CT imagery [1,2,[8][9][10][11][12]. These researchers have employed texture-based techniques [10], image segmentation methods [11], and knowledge-based classification [9,12] to locate and classify defects. In most cases, image analysis has focused on a single two-dimensional (2D) CT slice, although in a few cases neighboring slices have been used for 3D filtering during preprocessing steps.…”

“…When a log is to be sawed, a cutting strategy must be determined which preserves large areas of clear wood on board faces. There is a strong incentive to perform a correct assessment, for both veneer and saw logs, since the economic return can improve considerably (possibly up to a factor of 10) when a correct decision is made [4,9,11].…”