A Computer Vision System forLocating and Identifying Internal Log Defects Using CT Imagery

Abstract

A number of researchers have shown the ability of magnetic resonance imaging (MRI) and computer tomography (CT) imaging to detect internal defects in logs. However, if these devices are ever to play a role in the forest products industry, automatic methods for analyzing data from these devices must be developed. This paper reports research aimed at developing a computer vision system for locating and identifying internal defects in hardwood logs using CT imagery. This vision system can conceptually be divided into three components: a CT scanner based image acquisition system, a low-level module for image segmentation, and a high-level module for defect recognition. The processing steps involved in this vision system include CT data collection, image segmentation, three-dimensional volume growing, and a rule-based expert system for defect recognition. To date progress has been made on all these processing steps though the work on developing a rule-based expert system is really just getting underway. Experimental data is presented to show the progress that has been made on this vision system development.

Citation

Zhu, Dongping; Conners, Richard W.; Lamb, Frederick; Araman, Philip A. 1991. A Computer Vision System forLocating and Identifying Internal Log Defects Using CT Imagery. Proceedings, 4th International Conference on Scanning Technology in the Wood Industry. pp. 1-13.