Development of a Continuous Online Detecting System for Fruits Using Near Infrared Technology

A continuous online NIR inspection system in cooperation with light source, detection unit, reference calibration unit, and controlling unit was developed for detecting internal quality of a fruit. For the system, the spectrum measurement, reference calibration, and sugar content calculation can be automatically controlled by its operation software which developed with LabVIEW graphical programming language. The developed prototype system was used to measure the sugar content of Waxapples and Jujubes online. As well, the prediction model was established by Pentium III (933M Hz) computer in conveying speed of 40 mm/sec. The detection results demonstrated a good prediction capability from the prediction model. Moreover, the value of R2 is higher than 0.850 and the value of SEC is lower than 0.713, respectively. It indicated that the correct prediction percentage is over 84.4% within ±1 °Brix error. It point out a reliable inspection system specific dealing with the agricultural engineering aspect was developed in this study.