Online Measurement of Outline Size for Pinus densiflora Dimension Lumber: Maximizing Lumber Recovery by Minimizing Enclosure Rectangle Fitting Area

Abstract

This paper proposes a non-contact, online, rapid, and non-destructive measurement method of Pinus densiflora S.et Z. (Pinus densiflora) dimension lumber based on an algorithm of maximizing lumber recovery by minimizing the enclosure rectangle fitting area. The method takes the full influence of multiple factors, such as the difficulty in measuring large-size dimension lumber, conveying deflection, etc. into account. Image splicing, object and background segmentation, and lumber size measurements are carried out. Compared to other current algorithms, it can decrease the measurement errors and improve measurement efficiency. The problems of slow image segmentation, large-size lumber measurement, and conveying deflection are solved. Through measurement tests and error analysis, the measurement accuracy of the lumber size can reach 0.8 mm/m. This method meets the requirements of high-speed lumber production and lays the foundation for the development of a lumber detection and evaluation system.