글로버메뉴 바로가기 본문 바로가기 하단메뉴 바로가기

논문검색은 역시 페이퍼서치

> 한국농업기계학회 > 한국농업기계학회 학술발표논문집 > 26권 2호

Foreign Object Detection from Fresh-Cut Vegetables Using Near Infrared Hyperspectral Imaging Techniques

Foreign Object Detection from Fresh-Cut Vegetables Using Near Infrared Hyperspectral Imaging Techniques

( Salma Sultana Tunny ) , ( Byoung-kwan Cho )

- 발행기관 : 한국농업기계학회

- 발행년도 : 2021

- 간행물 : 한국농업기계학회 학술발표논문집, 26권 2호

- 페이지 : pp.147-147 ( 총 1 페이지 )


학술발표대회집, 워크숍 자료집 중 1,2 페이지 논문은 ‘요약’만 제공되는 경우가 있으니,

구매 전에 간행물명, 페이지 수 확인 부탁 드립니다.

1,000
논문제목
초록(외국어)
Foreign objects in vegetables and vegetable products are the primary source of customer complaints. However, there is not adequate research to detect the foreign materials from fresh-cut vegetables. Several studies have proved the necessity of developing an advanced technology to identify the foreign materials from fresh-cut vegetables. The present research has been done to develop a model using near infrared hyperspectral imaging (HSI) techniques to detect a wide range of possible foreign materials mixed with three different types of vegetables: cabbage, green onion, and carrot. Hyperspectral images of fresh-cut vegetables mixed with foreign objects were collected throughout the short wave infrared (SWIR) spectral region (1000 nm to 2500 nm) using a laboratory-based line scan HSI system in reflectance mode. The spectra of each foreign material and vegetable piece were collected and shuffled several times to avoid bias before applying the data to build a partial least square discriminant analysis (PLS-DA) model. 70% of the spectra were used for calibration, and the remaining 30% of data were used to validate the model. The model achieved more than 98% accuracy in both calibration and validation for identifying the foreign objects from all three types of vegetables. The obtained results showed that the near infrared hyperspectral imaging system with multivariate analysis could be an excellent choice for detecting the foreign objects from fresh-cut vegetables except for the black color foreign objects, which were similar in color with the background. This limitation might be solved by incorporating a RGB camera with the near infrared HSI system.

논문정보
  • - 주제 : 농학분야 > 농공학
  • - 발행기관 : 한국농업기계학회
  • - 간행물 : 한국농업기계학회 학술발표논문집, 26권 2호
  • - 발행년도 : 2021
  • - 페이지 : pp.147-147 ( 총 1 페이지 )
  • - UCI(KEPA) : I410-ECN-0102-2022-500-000880798
저널정보
  • - 주제 : 농학분야 > 농공학
  • - 성격 : 학술발표
  • - 간기 : 반년간
  • - 국내 등재 : -
  • - 해외 등재 : -
  • - ISSN :
  • - 수록범위 : 1996–2022
  • - 수록 논문수 : 4628