Skip to main content
ARS Home » Midwest Area » East Lansing, Michigan » Sugarbeet and Bean Research » Research » Publications at this Location » Publication #181424


item Noh, Hyun Kwon
item Lu, Renfu

Submitted to: ASAE Annual International Meeting
Publication Type: Proceedings
Publication Acceptance Date: 5/16/2005
Publication Date: 7/17/2005
Citation: Noh, H., Lu, R. 2005. Hyperspectral reflectance and fluorescence for assessing apple quality. ASAE Annual International Meeting. Paper No. 053069.

Interpretive Summary: The quality of apples directly affects the profitability of the fruit industry and consumer satisfaction. Currently, the fruit quality determination requires measuring multiple parameters including skin and flesh color, fruit firmness, sugar content, starch, acidity, ethylene production, etc. The methods/procedures for measuring these quality parameters are destructive, inefficient, and prone to operational error. Nondestructive sensing technology is not only useful for managing harvest time to maximize fruit postharvest quality, it is also critical for solving the inconsistent fruit quality problem that is still commonplace in the market. This research was to develop an optical technique of integrating hyperspectral reflectance and fluorescence for measuring fruit firmness, sugar content, and acidity of apples. Hyperspectral reflectance obtains both spectral and spatial information of the reflected light from an object, whereas fluorescence is a technique for measuring the light of longer wavelengths released from the fruit after it absorbed short-wavelength light, which is related to the physiochemical activities in the fruit. The results showed that fluorescence alone could not provide sufficient information for predicting apple quality. An integrated model of using both fluorescence and reflectance gave consistently better predictions of fruit quality parameters than the independent models of fluorescence and reflectance. This research demonstrated that integration of the two techniques improved our capability of measuring multiple quality attributes of apple fruit. Development of such an integrated system will be useful to the fruit growers and processors for better assessing the quality and maturity of apples and give the consumer greater confidence in the quality of apples.

Technical Abstract: Fluorescence and hyperspectral reflectance scattering have recently been researched for measuring fruit post-harvest quality and condition and they are promising for nondestructive detection of fruit quality. The objective of this research was to use hyperspectral reflectance and fluorescence for measuring apple fruit quality. A blue laser (408 nm) and a quartz tungsten halogen light were used as light sources for generating laser induced fluorescence and reflectance scattering in apples, respectively. The laser induced fluorescence and reflectance of ‘Golden Delicious’ apples were measured by using a hyperspectral imaging system. Fruit firmness and soluble solids and acid content were measured using standard destructive methods. Principal component analyses were performed to extract critical information from both hyperspectral reflectance and fluorescence data and this information was then related to fruit quality indexes. The fluorescence models had poorer predictions of the three quality indexes than the reflectance models. However, the prediction models of integrating fluorescence and reflectance performed consistently better than the individual models of either reflectance or fluorescence. The correlation coefficient for fruit firmness from the integrated model was 0.86 and the standard error was 6.97 N.