Online detection of bruises and soluble solid content of apple using near-infrared spectroscopy

The possibility of simultaneous online detection of apple bruises and soluble solids contents (SSC) was explored using the combination of near-infrared diffuse transmission spectrum technology and partial least squares method.
Visible near-infrared diffuse transmission spectra of bruised and normal apples were collected at a speed of five samples per second.
A partial least squares discriminant model for bruised apples was established using the energy spectrum; the prediction correlation coefficient (r_p) of the model was 0.92, the root mean square error of prediction (RMSEP) was 1.06, and the predictive misjudgment rate was 0. Meanwhile, the influence of bruised apples on the predictive ability of an SSC quantitative model of least squares was investigated; the root mean square error and the correlation coefficient (r_p) of the predicted model (RMSEP) were 0.54 and 0.87 °Brix, respectively.
The results show that the simultaneous online detection of apple bruises and SSC can be realized by using the combination of near-infrared diffuse transmission spectroscopy technology and the partial least squares method.

I International Apple Symposium

Xiaogang Jiang, Xudong Sun, Yande Liu

near infrared spectroscopy, bruising, diffuse transmission, apple, soluble solid contents, partial least squares

https://doi.org/10.17660/ActaHortic.2019.1261.30