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.

A testing approach for seeding precision was developed in this paper, which is an integrated technology of machine vision, pattern recognition, and automatic control. A machine vision based test-bed was developed for performance tests of grain seeders and a corresponding software package was compiled to capture the images of the deposited seeds, to segment the seeds from the background of the image, and to calculate the spacing between two seeds after precision seeding, the number of seeds per length after drill seeding, and the distance between hills and the number of seeds per hill after hill-drop seeding. A special illumination system was constructed to light the scenes under the cameras. A special designed image splicing algorithm was enabled to eliminate overlapped area of two adjacent images in sequence. The testbed has been proved reliable and accurate from quantity of practical tests.