A combination of computer vision and laser-light backscattering imaging, oil palm fruit maturity is classified

Abstract

Laser-light backscattering imaging (LLBI) is an optical imaging technique that records the interaction of light with plant tissue, generating light reflection data relevant for assessing product quality. This research combines LLBI and an RGB-based computer vision system to non-destructively classify the maturity level of oil palm fresh fruit bunches (FFBs). This technique offers a faster, more cost-effective and more accurate solution than traditional methods. The analysis includes RGB imaging and light reflection, where parameters such as color intensity, principal axis length, and area are analyzed using image processing algorithms. Results showed that the combination of LLBI and a computer vision system significantly improved the accuracy of ripeness classification, with a strong correlation between imaging parameters and quality attributes such as oil content and color. This approach provides an important step in improving harvesting efficiency and the production of high-quality palm oil.