Skip to main content
ARS Home » Midwest Area » East Lansing, Michigan » Sugarbeet and Bean Research » Research » Publications at this Location » Publication #295576
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: ASABE Annual International Meeting
Publication Type: Proceedings
Publication Acceptance Date: 6/24/2013
Publication Date: 7/21/2013
Citation: Cen, H., Lu, R., Ariana, D.P., Mendoza, F. 2013. Hyperspectral imaging-based classification and wavebands selection for internal defect detection of pickling cucumbers. In: Proceedings of the American Society of Agricultural and Biological Engineers Annual International Meeting, July 21-24, Kansas City, Missouri. Paper No. 13-1671220.

Interpretive Summary:

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 an in-house developed hyperspectral imaging system running at two conveyor speeds of 85 and 165 mm/s, 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 (energy and variance) from the region of interests in the spectral images at selected waveband ratios. 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 overall classification accuracies for all selected waveband ratios at the low conveyor speed were slightly higher than those at the higher conveyor speed, since the low speed resulted in more scan lines, thus higher spatial-resolution hyperspectral images. 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.