Abstract: Seeds motion images during the sowing process were obtained using high-speed camera system. Image quality was enhanced effectively by means of the image processing which including the background elimination, image filter, image sharpening and binary processing. Detection method for precision seeder was established according to the analysis of area and characteristic values of centroid position of the seeds. Precision and sowing uniformity tests were carried out with different working parameters. The results of comparison between high-speed camera and manual detection showed that the relative error of seeding qualified index was less than 1% and coefficient of variation was less than 3%. The method presented in the paper is adapted to the accuracy requirements for performance detection of precision seeder.