Combination of laser-light backscattering imaging and computer vision for rapid determination of oil palm fresh fruit bunches maturity

•The combination of backscattering imaging and computer vision system was applied.•The oil content and color values of oil palm FFB were analyzed.•The PCA and PLS models showed the great potential of the imaging techniques. The classification of oil palm fresh fruit bunches (FFB) in relation to the maturity level is an important aspect to determine the quality and productivity of the fruit. This study evaluated the utilization of combined computer vision and laser-light backscattering imaging in determining the oil content and color changes of oil palm FFB at different maturity levels i.e. unripe, ripe, and overripe. Red-green-blue (RGB) images referring to the computer vision and backscattering images of 90 oil palm FFB samples were acquired with 30 samples per each maturity level. Standard reference methods for oil content and color values (L*, a*, and b*) were determined in relation to the quality attributes of the oil palm FFB samples. Partial least squares (PLS) and principal component analysis (PCA) were used to analyze the quality changes of oil palm FFB based on its maturity levels. Multivariate algorithms such as linear discriminant analysis and quadratic discriminant analysis were applied to evaluate the classification performance based on the combined RGB and backscattering parameters. The combined optical techniques showed a good coefficient of determination (R2) of greater than 0.80 for both oil content and color values. The average classification accuracies were also higher than 85% in classifying oil palm FFB maturity. Hence, this work has demonstrated that combined computer vision and backscattering imaging systems could be useful as a non-destructive device for evaluating the classification of oil palm FFB.