Abstract:
This article describes about assessing the quality of Nam-
Dok-Mai mangoes using image processing methods for help farmers or
users to assess the quality of the mangoes and identify their sizes using
the images captured from digital camera. The operation of the program
starts by converting the color image of the mango into a binary image.
Next, dilation and erosion method are used to reduce noise in the image.
The program then compares the image with the templates in the library
to classify the mangoes by the sizes and inspects the speckles. Finally,
the color compositions of the image is analyzed. All the analyzed data
will be taken into account in determining the quality of the mangoes.
Form 36 tests, an image each, it was found that the accuracy was 86 percent.
The errors are caused by the error of color reading and the orientation of
mangoes which is not in the correct position