factor combinations used grayscale, histogram and line profile graph
of comparative analysis found that a component in the Lab and the
R-G color space under the fruit and the background factors have a
greater distinction In both cases the best interests of the kiwi fruit
split. In this study, a component selected Lab space.
(3) Fixed threshold method and the Great Law (OTSU) on the gray
image segmentation, can be a complete division of kiwifruit out of
which the law has good adaptability, but time-consuming.
Segmentation processing for binary image after the existence of
residue problems, this study used first to remove the image in the
small target, corrosion, corrosion of the image again to remove the
small target, and finally reverse-expansion methods to residues In
addition to, and achieved good results.
(4) Kiwifruit space for pose different processing methods used
to determine the appropriate goals of fruit contour, area, perimeter,
circularity, centroid, bounding rectangle parameters for the picking
robot guides the action.
Key Words：Kiwifruit; Target identification; Matlab; Lab Space;
Hough Transform
评论0
最新资源