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.

An automated eggplant grading system was introduced in the last paper. The system had three machine vision systems by use of NIR-color camera, 6 normal color cameras and 4 monochrome cameras. The first machine vision system was installed for checking fruits position and orientation on rotary trays, in which a fruit was put between two plates and was turned over, and which made all side fruit inspection possible. If fruit was not in center of the tray, it was necessary to return it back and to put it on the tray again. A CCD camera whose sensitivity was not only visible region but also infrared region was used in this system, because it was easy for infrared camera to detect black color eggplant fruit to discriminate from dark background.

The second machine vision system consisted of 6 color CCD cameras to inspect fruit color, size, shape, bruise and disease. The 3 color CCD cameras each were installed before and after turning over in a lane and connected to 2 PCs through 6 image grabber boards. Fruit length, average, max, and min diameters, area, apparent volume, fruit color, calyx color, fruit shape, degree of fruit bend, bruise number, bruise area, and so on were extracted from the images.

The third machine vision system consisted of 4 monochrome CCD cameras to check dullness of fruit surface, because dullness was an important index to evaluate fruit internal quality. 4 monochrome cameras were connected to a PC through 2 image grabber boards by occupying two channels each. From working results during a year of 2002-2003, it was observed that it was possible to detect many kinds of defects on eggplant fruit.