introduced automatic
apple sorting and grading using machine vision. In
this paper the intensity plane is converted to binary
image and calculated average intensity value, then
morphological filling is done to that image to
convert all the black pixels to white pixels assuming
that black pixels are defects, again average intensity
value is calculated. But not all the morphological
filled pixels are defects. A.N. Lorestani et al. [1]
introduced a fuzzy logic based decision support
system for grading of golden delicious apples by the
features such as color and size.
In the proposed method we have designed an
algorithm to automate the process for grading of one
variety of apple fruit, which is fast and efficient.
The paper is organized as follows: section 2
describes image acquisition and database collection,
and preprocessing, Which separates foreground and
background of the image. Section 3 discusses image
division, window elimination, feature extraction and
defect detection. Section 4 is about stem end and
calyx recognition which are natural parts of the
apple. In section 5 nearest neighbor classifier is
explained with Euclidean distance. Results are
discussed and output images are given in section 6.