Machine Vision System for Quality Control Assessment of Bareroot Pine Seedlings

Abstract

Abstract A PC-based multiple camera machine vision system for measuring bareroot pine seedlings has been developed for operational data collection in a forest nursery. The system was used for quality control sampling in a forest nursery for 2 lifting seasons. In tests conducted during the first lifting season measuring 100 loblolly pine (Pinus taeda L.) seedlings, there was only moderate correlation on a seedling-by-seedling basis between manual measurements made by nursery personnel and machine vision measurements of root collar diameter, shoot height, and tap root length. Overall machine vision and manual measurement distribution results for seedling samples were in close agreement, however. A bud tip height measurement routine was added to the system after the first lifting season, and a second set of tests was conducted measuring 175 history plot seedlings. In these tests, manual measurements made by researchers correlated well with machine vision measurements of root collar diameter and tap root length. Overall sample distribution results were also in close agreement for root collar diameter, tap root length, and bud tip height. With the machine vision system, seedling samples can be measured by 1 person in approximately the same or less time than it takes 2 people to measure them manually. South. J. Appl. For. 21(2):90-96.