Clustered citrus counting. For those regions where the
pixels were greater than 4/3 of the PsIC, an erosion and open
operation process was used to smoothen the edge of the regions
(Davies, 2005; Duda, Hart, & Stork, 2001), and the M8CCC
was used to distinguish the number of fruits (Fig. 1e). The
M8CCC could achieve the result similar to that obtained using
the LaplacianeGaussian segmentation algorithm method
while reduced a large number of data analysis process of the
software.
After processing through the above-mentioned steps, the
software counts the number of citrus (Fig. 1f). The estimation
result is displayed on the screen and stored in a database