Skip to main content
ARS Home » Midwest Area » East Lansing, Michigan » Sugarbeet and Bean Research » Research » Publications at this Location » Publication #188225
Title: HYPERSPECTRAL REFLECTANCE IMAGING FOR DETECTION OF BRUISES ON PICKLING CUCUMBERS

Author
item ARIANA, DIWAN - USDA-FAS
item Lu, Renfu
item GUYER, D - MICHIGAN ST UNIVERSITY

Submitted to: Computers and Electronics in Agriculture
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 5/4/2006
Publication Date: 6/26/2006
Citation: Ariana, D., Lu, R., Guyer, D.E. 2006. Hyperspectral reflectance imaging for detection of bruises on pickling cucumbers. Computers and Electronics in Agriculture. 53(1):60-70.

Interpretive Summary: Mechanical injuries to cucumber fruit may occur during mechanical harvest, transport, and handling. During these operations, care should be taken to minimize losses due to bruising. Mechanical injury often causes hidden internal physical damage to cucumber fruit that is invisible or difficult to detect by human inspectors. The injured cucumbers that passed inspection may lead to increased bloating during brining, causing major economic losses to the pickle processor. This research investigated the potential of hyperspectral imaging in the near-infrared (NIR) region that is invisible to human eyes for detecting bruises on pickling cucumbers. An NIR hyperspectral imaging system was developed to capture both spatial and spectral information of the reflected light from cucumber samples that were bruised by dropping and rolling, which frequently occur during mechanical harvesting and postharvest handling. Computer algorithms were developed to process hyperspectral images for identifying bruised areas from the individual cucumbers. The near-infrared hyperspectral imaging system effectively detected bruises in freshly harvested cucumbers with an accuracy rate of greater than 90%. Effective wavelengths and their combinations were identified, which will be useful in further development of a rapid, automated imaging system for bruise inspection in cucumbers. Automated detection of mechanically injured cucumbers through hyperspectral or multispectral imaging will help the industry provide high quality pickle products and reduce economic losses.

Technical Abstract: Mechanical injury often causes hidden internal damage to pickling cucumbers, which is difficult to detect in visual inspection. Bruised pickling cucumbers lower the quality of pickled products and can incur economic losses to the processor. A near-infrared hyperspectral imaging system was developed to capture hyperspectral images from freshly harvested pickling cucumbers in the spectral region of 900 nm - 1700 nm. The system consisted of an imaging spectrograph attached to an InGaAs camera with line-light fiber bundles as an illumination source. Experiments were conducted to acquire hyperspectral images of pickling cucumbers during the six-day time period after bruising from dropping or rolling under load which simulated damage caused by mechanical harvesting and handling systems. Results showed that reflectance spectra from the regions of interest (ROI) for bruised areas were consistently lower than those of normal areas. A large spectral difference between normal and bruised tissue occurred in the 950 nm - 1350 nm region. Bruise spectra changed over time from lower reflectance to higher reflectance. Band difference, band ratio, and principal component analysis were able to segregate bruised areas from normal areas with a detection accuracy of greater than 90%. Detection accuracies were lower as time progressed, which was attributed to the self-healing of the bruised tissue after mechanical injury. This research demonstrated that NIR hyperspectral imaging is useful for detecting bruises on pickling cucumbers. With the selected wavelengths identified, an efficient imaging system can be developed for rapid, real-time detection of bruised pickling cucumbers.