Skip to main content
ARS Home » Research » Publications at this Location » Publication #113041
Title: DEVELOPMENT OF DETECTION METHODS FOR DEFECTS AND CONTAMINATIONS ON APPLE SURFACE USING HYPERSPECTRAL IMAGING TECHNIQUE

Author
item Mehl, Patrick
item Chen, Yud
item Kim, Moon

Submitted to: Transactions of the ASAE
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/31/2003
Publication Date: 1/1/2004
Citation: Mehl, P.M., Chen, Y.R., Kim, M.S. 2004. Development of detection methods for defects and contaminations on apple surface using hyperspectral imaging technique. Journal of Food Engineering. 61(1):67-81.

Interpretive Summary: Apples are one of the most important agricultural commodities in the U.S. market. The increasing occurrence of bacterial contamination on agricultural products has created a public concern about the presence of plant or human pathogens in human food. Safety/quality control should include all harvested apples for direct marketing and for industrial processing for juice or cider production or other apple product applications. This paper presents the development of a hyperspectral imaging system and demonstrate that the asymmetric second difference method is an excellent technique for the detection of defects and contaminations on surfaces of various apple cultivars. This method is not dependent on surfaces of various apple cultivars. This method is not dependent on the apple cultivars. It is simple enough that no multivariate analysis; such as, principal component analysis is needed, and computation is fast since only three wavelenghts are effectively used. These research results will b beneficial to FDA, FSIS, and AMS who are interested in techniques for rapid detection of contaminations, diseases, and defects on apple surface for safety and quality concerns. Researchers working on on-line systems for detection of contaminations and defects on apple surface will also be interested in the research results. This paper will also be interesting to equipment manufacturers of apple sorting machines or quality and safety.

Technical Abstract: A high spatial resolution hyperspectral imaging system (a spatial resolution of 0.5-1.0 mm) was developed for detection of defective and contaminated foods and agricultural products. Various methods are presented for the detection of defects and/or contaminations on surfaces of the following four apple cultivars: Red and Golden Delicious, Gala, and Fuji. Surface defects and contaminations studied included side rots, bruises, flyspecks, scabs and molds, black poxes, and fungal and soil contaminations. The differences in spectral responses within the 430-900 nm spectral range are analyzed using monochromatic images and second difference analysis methods to enable the sorting of intact and contaminated apples. An asymmetric second difference method is proposed as an improvement to the symmetric second difference method. An asymmetric second difference method with a chlorophyll absorption waveband of 685 nm and two bands in the near-infrared region is shown to provide an excellent detection of the defective and/or contaminated portions of apples, independent of the color and cultivars of apples. Also, the asymmetric second differences method is simple and does not require as much computation as it is required by other methods such as the multivariate analysis technique.