ASABE Technical Library - Abstract

Click on “Download PDF” for the PDF version or on the title for the HTML version.
If you are not an ASABE member or if your employer has not arranged for access to the full-text, view access and subscription options

NIR-Based Sensing Coupled with Physical/Color Features to Identify Brix Level of Florida Citrus

Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan www.asabe.org

Citation:  Paper number  026037,  2002 ASAE Annual Meeting . (doi: 10.13031/2013.9399) @2002
Authors:   William M. Miller, Manuela Zude
Keywords:   Citrus, fruit quality, instrumentation

Preliminary tests were undertaken to assess the accuracy of near infra-red (NIR) technologies to measure Brix level of intact Florida grapefruit and Honey tangerines. Two units were utilized: (1) an on-line prototype unit (Mitsui QScope, Sumitomo Metal Mining Co., Tokyo, Japan) and (2) an hand-held unit developed for general fresh fruit quality assessment (Inst. of Agric. Engr.- Bornim, Potsdam, Germany). The on-line tests were conducted at a rate of ~5 fruit per second. Preconditioned Interior white grapefruit were tested at two temperatures, 5 and 20EC while commercial Indian River red grapefruit were tested at ambient conditions. Comparative measurements were obtained for both on-line and hand-held NIR units with the fruit subsequently juiced to obtain a refractive index measurement of the fruits Brix level. Linear regression relationships were developed between the non-destructive NIR techniques and the laboratory Brix measurements. The highest correlation, r2 = 0.67**, was found for Indian River red grapefruit of the smaller size of fruit tested (9.5-11.0 cm diameter). Percent correct classification analysis was ascertained at 9EBrix for Interior grapefruit and 10EBrix for Indian River red grapefruit. Higher classification rates were obtained for ambient temperature tests at ~20EC and for the smaller size of Indian River red grapefruit. A neural network (NN) model with combined inputs of physical and color attributes, plus predicted Brix, was evaluated. Corresponding Brix data for Honey tangerine also were obtained utilizing a handheld NIR probe. Accuracy and repeatability were measured for grapefruit only in a second year test program. Percent correct classification averaged ~90% for 10oBrix and ~80% for a 11oBrix setpoint.

(Download PDF)    (Export to EndNotes)