SHORT NOTES: CAN THE PHYSICAL PROPERTIES OF WOOD SAMPLES BE PREDICTED FROM PHOTOGRAPHS DISPLAYED ON A MONITOR?

Abstract

This study focused on evaluatingthe physical properties of wood from photographs displayed on a monitor. Sample photos of 475 hardwoods and their physical information were collected from a wood database. R, G, and B values were extracted from the wood photos using color picker software.Statistical techniques such as Pearson's correlation coefficient and multiple regression analysis were applied to investigate relationships between wood color and physical properties. From results of Pearson's correlation coefficient, R, G, and B values were most affected by specific gravity. In a multiple regression analysis, tree size, specific gravity, and modulus of rupture (MOR) were significant in the positive (+) direction by color (R, G, and B). On the other hand, modulus of elasticity (MOE) was significant in the negative (-) direction at the 1% level by color. The specific gravity of wood had the most significant effect on R, G, and B values in multiple regression analysis. In conclusion, the color and specific gravity of wood were related closely. Additionally, it is possible to predict thephysical properties of wood from the R, G, and B values of a wood sample photograph displayed on a monitor. These results could provide useful information for wood researchers as well as wood exporters and importers.