Skip to main content
ARS Home » Northeast Area » Beltsville, Maryland (BARC) » Beltsville Agricultural Research Center » Environmental Microbial & Food Safety Laboratory » Research » Publications at this Location » Publication #295067
Title: Detection algorithm for cracks on the surface of tomatoes using Multispectral Vis/NIR Reflectance Imagery

Author
item JUNG, DANHEE - Chungnam National University
item Kim, Moon
item Chao, Kuanglin - Kevin Chao
item HASEGAWA, MASUMI - Kyushu University
item LEE, HOONSOO - Chungnam National University
item LEE, HOYOUNG - Seoul National University
item CHO, BYOUNG-KWAN - Chungnam National University

Submitted to: Biosystems Engineering
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 8/16/2013
Publication Date: 9/5/2013
Citation: Jung, D., Kim, M.S., Chao, K., Hasegawa, M., Lee, H., Lee, H., Cho, B. 2013. Detection algorithm for cracks on the surface of tomatoes using Multispectral Vis/NIR Reflectance Imagery. Biosystems Engineering. 38(3):199-207.

Interpretive Summary: Growth cracks on tomatoes have been known to be a pathway for internalization of pathogenic bacteria. In this study, ARS researchers investigated the use of spectral imaging techniques to detect tomatoes with cuticle cracks. The results showed that spectral imaging methods could be used to identify cracks on tomato surfaces with a 91% successful classification rate. We demonstrated that spectral imaging is feasible for the classification of defective (cracked) tomatoes. However, further investigation is needed to improve the tomato crack detection accuracy rates. The information provided in this investigation is useful to food engineers and fruits processing industries.

Technical Abstract: Tomatoes, an important agricultural product in fresh-cut markets, are sometimes a source of foodborne illness, mainly Salmonella spp. Growth cracks on tomatoes can be a pathway for bacteria, so its detection prior to consumption is important for public health. In this study, multispectral Visible/Near-Infrared (NIR) reflectance imaging techniques were used to determine optimal wavebands for the classification of defect tomatoes. The results showed that two optimal wavelengths, 713.8 nm and 718.6 nm, could be used to identify cracked spots on tomato surfaces with a correct classification rate of 91.1%. The result indicates that multispectral reflectance imaging with optimized wavebands from hyperspectral images is an effective technique for the classification of defective tomatoes. Although it can be susceptible to specular interference, the multispectral reflectance imaging is an appropriate method for commercial applications because it is faster and much less expensive than Near-Infrared or fluorescence imaging techniques.