Identification of early decayed oranges using structured-illumination reflectance imaging coupled with fast demodulation and improved image processing algorithms

Authors: LI, JIANGBO - Beijing Academy Of Agricultural Sciences; LU, YUZHEN - Michigan State University; Lu, Renfu

Submitted to: Postharvest Biology and Technology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 10/18/2023
Publication Date: 10/27/2023
Citation: Li, J., Lu, Y., Lu, R. 2023. Identification of early decayed oranges using structured-illumination reflectance imaging coupled with fast demodulation and improved image processing algorithms. Postharvest Biology and Technology. 207. Article 112627. https://doi.org/10.1016/j.postharvbio.2023.112627.
DOI: https://doi.org/10.1016/j.postharvbio.2023.112627

Interpretive Summary: Orange fruit are susceptible to pathogen invasion during postharvest handling. Early detection and removal of pathogen-infected orange fruit is critical to minimizing potential devastating economic loss to growers and packers. However, the fruit infected with such pathogens as Penicillium digitatum fungus, the most serious and devastating pathogen for orange fruit, usually do not display visual symptoms at the early stage, which makes it difficult to detect using conventional imaging techniques. Structured-illumination reflectance imaging (SIRI) is an emerging imaging technique, which applies special patterns of illumination to the imaging object. The technique has shown enhanced capabilities for detecting subsurface defects of horticultural products. This study was aimed at exploring a faster imaging approach and effective image processing algorithms for applying the SIRI technique for early detection of decayed orange fruit. Phase-shifted pattern images were acquired, using a specially designed SIRI system, from individual sound and infected oranges inoculated with Penicillium digitatum fungus. Different image demodulation, enhancement and segmentation algorithms were applied to the acquired SIRI images. Three image processing methodologies based on single and two acquired phase-shifted images for each sample were proposed for identification of early decayed orange fruit. Results showed that all three methodologies achieved 95% or higher accuracies for segregating decayed fruit from sound ones. This study demonstrated that it is technically feasible to use one or two SIRI images, coupled with the developed image enhancement and processing algorithms, for effective identification of early decayed fruit. This would make the technique possible for future online, real-time inspection of early decayed orange fruit.

Technical Abstract: Effective detection of decayed citrus in the early stage is challenging because there are no or few visual symptoms. Structured-illumination reflectance imaging (SIRI) has been proven effective for enhanced detection of subsurface defects in fruit. Amplitude component (AC) images retrieved from the original SIRI patterned images are useful for defect detection, but generally require acquiring three phase-shifted pattern images, which limits the imaging and detection speed. Moreover, the AC images may also suffer from noticeable uneven brightness due to fruit curvature, which affects the identification of decayed areas. This study was therefore aimed to explore a faster methodology, based on SIRI technology, for identification of early decayed oranges. Pattern images were acquired, using three phase-shifted sinusoidal illumination patterns at the wavelength of 810 nm and a spatial frequency of 0.20 cycles mm-1, from the orange samples infected with Penicillium digitatum fungus, the most serious and devastating pathogen for orange fruit. Two-dimensional spiral phase transform was used to obtain AC images from one or two pattern images. The acquired AC images were then processed by using simple brightness adjustment and integral image-based fast average filtering for brightness correction, and improved watershed algorithm and global threshold for segmentation of decayed areas. Three different combinations of these image processing procedures for single and two pattern images were proposed to distinguish decayed oranges from sound ones. The three methodologies all achieved high overall identification rates of 97.5%, 95.0% and 95.3%, when the stem-end effect was also considered. This study showed that accurate detection of early decayed orange fruit can be achieved by using one or two phase-shifted pattern images, which would be beneficial for real-time implementation of the technique.