Location: Sugarbeet and Bean Research
Title: DETECTION OF INTERNAL DEFECT IN PICKLING CUCUMBERS USING HYPERSPECTRAL TRANSMITTANCE IMAGING Authors
|Ariana, Diwan - USDA-FAS|
Submitted to: Transactions of the ASABE
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: February 28, 2008
Publication Date: April 1, 2008
Citation: Ariana, D., Lu, R. 2008. Detection of internal defect in pickling cucumbers using hyperspectral transmittance imaging. Transactions of the ASABE. 51(2):705-713. Interpretive Summary: Cucumbers are susceptible to damage during fruit growth, harvest, transport, and processing. The presence of defects, either external or internal, can lead to rejection because consumers demand quality products. Typically, quality control measures involve evaluations of freshly harvested cucumbers, brine stock, and pickled products. External defects such as broken, smashed, and cuts are relatively easy to identify and these defective cucumbers can be discarded during the sorting/grading process. However, internal defects such as carpel separation or hollow cucumbers are difficult to detect by human inspectors without cutting open the fruit, which prohibits evaluation of individual cucumbers. Therefore, a rapid and nondestructive method to evaluate internal defects would be valuable to the pickling cucumber industry. In this research, a hyperspectral transmittance imaging technique was developed for detecting internal defect of cucumbers in the form of carpel separation or hollow cucumbers. Hyperspectral imaging provides both spectral and spatial information about the sample, which is especially useful for identifying difficult-to-detect defects or quality attributes. With transmittance mode, light can have better interaction with interior fruit tissue, thus allowing for assessment of internal quality/defects of cucumbers. Experiments were performed on normal and defective fruit for two cucumber varieties with the hyperspectral imaging system over the visible and near-infrared region of 450-950 nm during the six-day period. Two image processing methods were used and compared for classifying defective and normal cucumbers. The hyperspectral transmittance imaging technique achieved overall classification accuracies of approximately 90% for defective cucumbers and 98% for normal cucumbers. The technique is fast and nondestructive and, hence, it would be useful for online sorting and grading of pickling cucumbers to ensure consistent, high quality pickled products for the consumer.
Technical Abstract: Internal quality is an important aspect in the quality control and assurance of pickled products. A rapid and nondestructive method for internal defect detection would be of value to the pickle industry. A hyperspectral transmittance imaging technique was developed to detect internal defect in the form of carpel suture separation or hollow cucumbers resulting from dropping and rolling under load. Hyperspectral transmittance images were collected from ‘Journey’ and ‘Vlaspik’ cucumbers over the six day period after they were subjected to mechanical stress. Partial least squares discriminant analysis (PLS-DA) was performed on mean and standard deviation spectra extracted from the hyperspectral transmittance images for classifying cucumber samples into defective or normal classes. An image segmentation method, based on comparison of unknown samples with a mean hyperspectral image of normal cucumbers, was applied to classify pixels into normal or defective class. Transmittance spectra of defective and normal cucumbers were similar in shape but higher in magnitude for the defective cucumbers. Transmittance values for both defective and normal cucumbers were higher in the near-infrared range of 700-950 nm than that of the visible range (450-700 nm). Average classification accuracies of 90.2%, 98.7%, and 95.4% were achieved using PLS-DA, whereas accuracies of 89.1%, 94.6%, and 90.5% were achieved using the image segmentation method for Journey, Vlaspik, and mixed variety model respectively. Hyperspectral transmittance imaging technique has the potential for nondestructive and rapid detection of internal defect in pickling cucumbers.