Application of spectral imaging for safety inspection of fresh cut vegetables. Issue 1 (March 2021)

• Record Type: Journal Article
• Authors: Cho, B K
• Is Part Of: IOP conference series. Volume 686:Issue 1(2021)
• Publication Date: 2021-03
• Subjects: Earth sciences -- Periodicals; Environmental sciences -- Congresses; Environmental sciences -- Periodicals; 550.5
• DOI: 10.1088/1755-1315/686/1/012001
• Languages: English
• ISSNs: 1755-1307

Abstract

Abstract: Food safety remains a critical issue and receiving increasing attention in both developed and developing countries. Fresh-cut vegetables may contain foreign objects such as small plastic pieces or biological contaminants (bugs and worms). In order to comply with requirements for product safety and maintain consumer confidence, there is a need for high-throughput, non-invasive technique for foreign objects detection in and real-time sorting of fresh-cuts. In this work, a liquid crystal tunable filter (LCTF) based spectral imaging system in the spectral range from 420 nm to 730 nm has developed and integrated with a conveyor unit. A software-interface was developed to collect the selected band images of moving samples and real time image processing for potential foreign materials detection on fresh cut vegetables. Spectral images throughout the whole spectral region were first collected for both fresh-cuts and foreign materials. The optimal wavebands for discriminating between foreign materials contaminated and sound fresh-cut were investigated. The optimal wavebands for each fresh-cut were selected and thus the developed system was used in multispectral imaging mode. Image processing algorithms were developed for detection of all potential foreign materials in fresh cut processing. The obtained results demonstrate that several kinds of foreign materials can be detected effectively despite being similar in color or nearly transparent. The ultimate advantage of a developed system is that it can be used either as in fluorescence mode or for collection of multispectral images in visible regions.