Rapid and low-cost detection of moldy apple core based on an optical sensor system

Authors: LI, LONG - China Agricultural University; PENG, YANKUN - China Agricultural University; LI, YONGYU - China Agricultural University; YANG, CHENG - China Agricultural University; Chao, Kuanglin - Kevin Chao

Submitted to: Postharvest Biology and Technology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 12/30/2020
Publication Date: 1/20/2021
Citation: Li, L., Peng, Y., Li, Y., Yang, C., Chao, K. 2021. Rapid and low-cost detection of moldy apple core based on an optical sensor system. Postharvest Biology and Technology. https://doi.org/10.1016/j.postharvbio.2020.111276.
DOI: https://doi.org/10.1016/j.postharvbio.2020.111276

Interpretive Summary: Moldy core is a common internal defect of apples caused by fungal infection. Apples with moldy core not only reduces the quality of the fruit but also is a potential food safety risk. Thus, an accurate, rapid nondestructive method and device to detect moldy core is needed. A simple optical sensor system with seven discrete wavelengths was developed to detect moldy apple core. A simplified spectral data analysis method was proposed for moldy apple core detection and successfully validated. Compared with microbial testing (for fungal identity), spectral methods are more rapid, less complicated sampling and sample handling steps. Compared with traditional full spectral analysis methods, the new method minimized false negatives due to the high background noise, and thus achieved improvements in accuracy, sensitivity, and specificity. The optical system has the advantages of using lower cost instrumental devices facilitating its potential application in an online fruit sorting system, benefiting the fruit processing industry and consumers.

Technical Abstract: An optical sensor system for the detection of moldy apple core (Malus domestica Borkh.) was developed. The new optical sensing system uses only seven wavelengths at 425, 455, 515, 615, 660, 700 and 850 nm. Due to its very specific spectral properties, the traditional spectral processing method cannot be used to improve the spectral classification accuracy. Thus, spectral shape features (SSFs) (i.e., the spectral ratio (SR), spectral difference (SD) and normalized spectral ratio (NSR)) were applied to improve the model performance. Principal component analysis was performed prior to linear discriminant analysis modeling (LDA), which can eliminate the multicollinearity problems in the spectral datasets. After LDA models were established, Otsu’s method and the maximum entropy (ME) method were proposed for spectral qualitative analysis to determine the optimal threshold. The combination of the NSR, SD and SR, achieved the best prediction accuracy. For the calibration set, the accuracy was 98.5%, and the sensitivity and specificity were equal to 0.98 and 0.99, respectively. The discriminant accuracy was 95.8% for the independent validation set, and the sensitivity and specificity were equal to 0.97 and 0.95, respectively. The Otsu’s method resulted in higher prediction accuracy and was more suitable for threshold determination than the ME method. The optical sensor system is cost-effective and, combined with its high accuracy, results in a convenient and practical technology for rapid detection of moldy apple core (mold) damage on apples.