The difference between Al and traditional methods

Rate this post

Sorting device for apple, peach, nectarine, shebarang seed fruits based on color, size, detection of external defects and surface pests.

360 degree photography based on image processing equipped with artificial intelligence

Suitable for all kinds of fruits, especially apples, peaches, nectarines, shebarang, citrus fruits.

With detection capability

  • Color
  • Size (diameter)
  • Appearance defects (stains, scratches, fruit surface pests and other appearance characteristics)

Diagnosis examples:

Why artificial intelligence?

In general, two main methods are common in image processing based detection systems.

Classic method

In this method, the desired parameter is detected and extracted based on basic features such as color intensity.

Artificial intelligence

In this method, the appropriate neural network is selected and trained with a large number of examples. Based on its experience, the network detects and extracts the desired parameters.

Reasons for the superiority of artificial intelligence

  • In classical methods, many external defects cannot be detected or extracted, or they are confused with other parameters. For example, in apple, in color photo processing, no difference can be seen between the spots and the top and bottom of the apple in terms of color parameters. IR photography is used to solve this problem. This method is much more effective, but it still doesn’t show many pale spots.
  • Learning is one of the main features of artificial intelligence, which means that in this method you always have the possibility to upgrade or improve the detection system. If there is a need to detect a drop or a new feature, this is possible in a very short time.

How the device works:

  • First, the fruit is unloaded on the entrance conveyor and after initial processing (washing, inspection, etc.) it is lined up individually and enters the grader section.
  • Fruits are photographed individually and their appearance characteristics are extracted. Based on the settings of the machine operator, fruits with the same characteristics are removed from the line at the desired exit.
  • Example 1: The user can enter the settings menu in Hegan Sort Sib and choose apples exported to India. Then set the outputs as follows. (like the video above)
  • First output: Apples with defects such as blemishes above 10%
  • Second output: Apples with a weight between 100 and 150 grams and a red color above 80%
  • Third output: Apples with a weight between 100 and 150 grams and a red color of less than 80%
  • Fourth output: Apples with a weight between 150 and 200 grams and a red color above 80%
  • Fifth output: Apples with a weight between 150 and 200 grams and a red color of less than 80%
  • Output seven: Apples weighing more than 200 grams and red color above 80%
  • Output eight: Apples weighing more than 200 grams and red color less than 80%
Related posts

Leave a Reply

Your email address will not be published. Required fields are marked *

Table of Contents