Skip to main content
ARS Home » Midwest Area » East Lansing, Michigan » Sugarbeet and Bean Research » Research » Publications at this Location » Publication #293333

Title: Hyperspectral imaging-based classification and wavebands selection for internal defect detection of pickling cucumbers

Author
item CEN, HAIYAN - Michigan State University
item Lu, Renfu
item ARIANA, DIWAN - Michigan State University
item Mendoza, Fernando

Submitted to: Food and Bioprocess Technology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/13/2013
Publication Date: 4/25/2014
Citation: Cen, H., Lu, R., Ariana, D.P., Mendoza, F. 2014. Hyperspectral imaging-based classification and wavebands selection for internal defect detection of pickling cucumbers. Food and Bioprocess Technology. 7:1689-1700.

Interpretive Summary: Internal defects like soft tissue, carpel separation and hollow center in pickling cucumbers result in inferior pickled products and lower profitability for the pickle industry. These defects are difficult to detect because they are hidden from the surface. The ARS laboratory in East Lansing has developed a hyperspectral imaging-based inspection system, which operates in both reflectance (400-700 nm) for surface quality inspection and transmittance (700-1000 nm) for internal defect detection. Hyperspectral imaging acquires three-dimensional spectral images that contain both spectral and spatial information. While the technique has shown to be effective for defects detection, it has been impeded for high-speed online inspection of pickling cucumbers and other horticultural products because of the need to process a huge quantity of data. This research was, therefore, aimed at identifying a few optimal wavebands from the hyperspectral image data for rapid, effective detection of internal defect of pickling cucumbers, so that the technique can be implemented for rapid online inspection of pickling cucumbers. Three hundred freshly harvested pickling cucumbers of ‘Journey’ variety were tested at two conveyor speeds, before and after they were treated with mechanical stress to induce severe and slight levels of internal damage. Two mathematical methods, called minimum redundancy-maximum relevance (MRMR) and principal component analysis (PCA), were used to identify the optimal wavebands for detecting internal damage in the pickling cucumbers. The MRMR method was better than the PCA method in the defect detection. The best two wavebands (837 nm and 887 nm) identified by MRMR resulted in 95% and 94% overall detection accuracy for the conveyor speeds of 85 mm/s and 165 mm/s, respectively, when the cucumbers were graded into the normal and defective classes. The overall detection accuracies for the three-class classification (i.e., slightly and severely damaged and normal) were 83% and 81%, respectively, for the conveyor speeds of 85 mm/s and 165 mm/s. The optimal wavebands identified from the research can be implemented for high-speed inspection of pickling cucumbers for internal defect.

Technical Abstract: Hyperspectral imaging is useful for detecting internal defect of pickling cucumbers. The technique, however, is not yet suitable for high-speed online implementation due to the challenges for analyzing large-scale hyperspectral images. This research was aimed to select the optimal wavebands from the hyperspectral image data, so that they can be deployed in either a hyper- or multi-spectral imaging-based inspection system for automatic detection of internal defect of pickling cucumbers. Hyperspectral reflectance (400-700 nm) and transmittance (700-1,000 nm) images were acquired, using a hyperspectral imaging system running at the conveyor speeds of 85 and 165 mm/s, respectively, for 300 ‘Journey’ pickling cucumbers before and after they were induced internal damage by mechanical load. Minimum redundancy-maximum relevance (MRMR) and principal component analysis (PCA) were used for the optimal wavebands selection. Discriminant analysis with Mahalanobis distances classifier was performed for the two-class (i.e., normal and defective) and three-class classifications (i.e., normal, slightly defective, and severely defective) using mean spectra and textural features from the region of interests in the spectral images. MRMR wavebands selection generally outperformed PCA in the classification performance. The two-band ratio of 887/837 nm from MRMR gave the best overall classification results with the accuracy of 95.1% and 94.2% at the conveyor speeds of 85 mm/s and 165 mm/s, respectively, for the two-class classification. The highest classification accuracies for the three-class classification based on the optimal two-band ratio of 887/837 nm were 82.8% and 81.3% at the conveyor speeds of 85 mm/s and 165 mm/s, respectively. The mean spectra-based classification achieved better results than the textural feature-based classification except in the three-class classification for the higher conveyor speed. The identified two-band ratio of 887/837 nm in transmittance mode could be applied for fast real-time internal defect detection of pickling cucumbers.