Development of sensing platform for fruit sorting

Authors: LEE, AHYEONG - National Institute Of Agricultural Sciences (RDA); Baek, Insuck; Kim, Moon; KIM, JINSE - National Institute Of Agricultural Sciences (RDA); HONG, SUKJU - National Institute Of Agricultural Sciences (RDA)

Submitted to: ASABE Annual International Meeting
Publication Type: Abstract Only
Publication Acceptance Date: 7/25/2024
Publication Date: N/A
Citation: N/A

Interpretive Summary:

Technical Abstract: The quality of fruits is a crucial determinant for consumers in making purchasing decisions. Various countries have established sorting criteria to eliminate defect fruits and grade normal fruits. For automated fruit sorting technology development, research utilizing optical technologies has been conducted. Existing studies mainly utilize RGB imaging for discriminating external defects which are visually distinguishable such as cuts and scab. In cases of bruises where defects resemble the color of normal areas, hyperspectral imaging is usually employed. This study aims to comprise hardware and develop software capable of simultaneously capturing RGB and hyperspectral images for fruit sorting. The system consists of a stage, light sources, and cameras (Hyperspectral, and RGB). Light sources include halogen and UV for hyperspectral imaging, and white LEDs for RGB imaging. The stage, movable along the y-axis, allows controlled adjustments of movement speed through software. The acquired hyperspectral and RGB images from the developed system will be used to develop a model for detecting defects in fruits.