Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

The objective of this research was to develop an automatic outward-appearance features measurement system for grafted tomato seedlings. The system employs a set of image acquisition and processing instruments that incorporate image analysis technology. Computer software packages LabVIEW 7.0 and Vision Assistant 7.0 were used to analyze the measured seedling feature information, which included the width and height of the rootstock/scion, the width and the height of the grafting-tubes, the whole grafted seedling height, number of leaves, and the canopy area. By monitoring and analyzing this feature information, the growth condition and the physiological status of the seedlings can be understood better and controlled. This research was targeted at grafted tomato seedlings that used "EG-203" eggplant as the rootstock and "Asian No. 6" tomato as the scion. The experimental results showed that the measurement error for the width and the height of both the rootstock and the scion was less than 9%. For leaf area measurement, a LI-COR LI-3100 leaf area meter was used to give a comparative accuracy of the system. The results showed that the average error for LI-3100 measurements was 3.09% with a standard deviation of 2.93, while these values for the developed system was 3.64% and 2.03, respectively; thus, there was no significant difference between the leaf area measured by LI-3100 and the developed system.