The Design of the Divider Door for the Orange Grading Machine Based on the Diameter and Color Uses Image Processing

Abstract

Post-harvest handling will affect the quality of agricultural commodities. One of the main post-harvest activities of orange fruit is grading. Grading is an activity to separate or classify orange fruit based on certain criteria such as size, weight, and color according to the applicable quality standard. This grading activity is important because it will determine the next handling process. Generally, orange grading activities are done manually, this was chosen because the process can be faster, but the manual grading has weakness e.g less efficient and requires more man power. Using the man power in determining product quality are based on experience, guesswork and tend to be subjective because it is in accordance with the perceptions of human being. A grading system and machines that use controls were developed to obtain a more uniform product based on color and size is an alternative to overcome this problem, . Nowadays, the grading process has been developed with an automation system using electronic and mechanical devices so that the results are more effective and accurate. The grading process of orange fruit with a divider door that uses image processing application and a microcontroller will have more precise and objective capability. It will be more profitable than manual grading in the future life. The aim of this research is to design a machine grading system based on the Arduino microcontroller. The grading machine uses an image processing application that reads the diameter value and skin color of the orange, Arduino microcontroller is used to separate oranges based on its quality class using a divider door connected to a stepper motor. The design of the orange grading machine has a size of 1.35 cm × 74 cm and has 4 grading classes, i.e class A, class B, class C and class D. The design of the divider door for the orange grading machine has 2 trapezoid-shaped doors, the length of the first door is 30 cm, the width of the top is 26 cm and the width of the bottom is 10 cm. The length of the second door is 30 cm, the width of the top is 22 cm and the width of the bottom is 10 cm. The percentage of suitability of the results of the grading machine with a dividing door compared to manual grading of oranges is 97%. Image processing error percentage is 14%. The capacity of the orange grading machine model with a dividing doors is 300 items / hour and the efficiency value of the orange grading machine model is 55.66%.