Ammonia (NH 3) is the main preservative that is added to field and concentrated latices to prevent the deterioration of properties and, in serious cases, coagulation of latex. Almost all factories monitor NH 3 content or alkalinity during processing, as it is an important parameter for trading purposes. Alkalinity is determined by the standard analytical method, acid-based titration, as detailed in ISO 125:2011(E) Natural Rubber Latex Concentrate-Determination of Alkalinity. This method requires a skilled analyst and also the use of chemicals. In addition, the titration involves the subjective determination of the end-point, which may vary with each analyst. Near infrared (NIR) spectroscopy, a rapid, non-chemical and repeatable method, was used in this work. Calibration equations for predicting alkalinity were constructed from the relationship between the absorbance spectra of latex (measured using a portable NIR and a Fourier transform (FT)-NIR spectrometer) and the alkalinity content of the latex. It was found that the best equation obtained using the portable and the FT-NIR spectrometer could be used to predict alkalinity in latex with a coefficient of determination, standard error of prediction and ratio of standard error of validation to the standard deviation of 0.63, 0.101% and 1.62, and 0.97, 0.027% and 6.07, respectively. From the statistics testing for performance measurement as detailed in ISO12099:2010, NIR-predicted values were no different from actual values at the 95% confident interval. Moreover, the best equation obtained from the more reliable calibration achieved using the FT-NIR spectrometer is attributed to the longer wavelength range.

Starch content is an important parameter indicating the state of harvest maturity of fresh cassava root. Nowadays, the methods used for estimating the starch content in the field are the measurement of root weight, size, or snapping force. These methods are simple but the results are rather incorrect. For this reason, a developed portable visible and near-infrared spectrometer­(350–1050 nm) was used to estimate rapidly and nondestructively starch content in fresh cassava root. The best starch prediction model received from the full wavelength region was able to predict the starch content with a correlation coefficient of prediction (r p) of 0.825, standard errors of prediction of 2.502%, and bias of −0.115%. Moreover, the predicted values were not significantly different from the actual values obtained from the standard method at 95% confidence intervals. It was also noted that the top position of the root was a good representative for starch prediction. In addition, this position was easy to be measured in the field before harvesting.

Short-wavelength near infrared spectra in the interactance mode were collected from intact cassava roots and cassava flesh, using two portable spectrometers for the spectral regions of 720–1050 and 850–1150 nm, respectively. All starch prediction models were developed using the partial least squares regression. Good prediction performance was obtained from the cassava flesh (cross-section cut root) measurement with a correlation of prediction (r p) of 0.917 and standard error of prediction (SEP) of 1.73%, for both spectrometers. For the intact root, the prediction models were satisfactorily accurate with r p values of 0.687 and 0.772 and SEP of 3.151 and 2.803%, respectively. Moreover, the performance measurement of all optimum models was also evaluated according to ISO 12099:2017­(E). The results showed that the predicted values were not significantly different from the actual values obtained from the standard method at 95% confidence intervals. These results showed the feasibility of using portable spectrometers to predict the starch content of fresh cassava roots.

Volatile Fatty Acid number (VFA no.) is one of the parameters indicating the state of quality of Para rubber latex at that particular time. Most factories analyze this parameter using standard analytical method as in ISO 506:1992(E). Nevertheless, this procedure is complicated, chemical and time consuming, as well as skilled analyst required. Therefore, near infrared (NIR) spectroscopy which is rapid, accurate and nonchemicals method was applied to determine the VFA no. in¯eld latex and concentrated latex based on quanti¯cation and discriminant model. The best calibration equation was obtained from standard normal variate (SNV) spectra in the region of 6109.7-5770.3, 4613.1-4242.9 cm À1 with R ¼ 0:832, SECV ¼ 0:036 and no bias. From the performance check, statistically it was found that SECV and bias were low enough for practical acceptance and the predicted VFA no. was not di®erent signi¯cantly from actual VFA no. at 95% con¯dence intervals. In addition, discriminant model was developed to separate good quality latex from the deteriorated latex using VFA no. at 0.06 as standard as in ISO 2004ISO :2010. The discriminant model can be used to screen the latex with overall accuracy of 91.86% in validation set.

Watercore and sugar content are internal qualities which are impossible for exterior determination. Therefore the aims of this study were to develop models for nondestructive detection of watercore and predicting sugar content in pear using Near Infrared Spectroscopy (NIR) technique. A total of 93 samples of Asian pear variety \SH-078" were used. For sugar content, spectrum of each fruit was measured in the short wavelength region (700-1100 nm) in the re°ection mode and the¯rst derivative of spectra were then correlated with the sugar content in juice determined by digital refractometer. Prediction equation was performed by multiple linear regression. The result showed Standard Error of Prediction ðSEPÞ ¼ 0:58 Bx, and Bias ¼ 0:11. The result from t-test showed that sugar content predicted by NIR was not signi¯cantly di®erent from the value analyzed by refractometer at 95% con¯dence. For watercore disorder, NIR measurement was performed over the short wavelength range (700-850 nm) in the transmission mode. The¯rst derivative spectra were correlated with internal qualities. Then principle component analysis (PCA) and partial least squares discriminant analysis (PLSDA) were used to perform discrimination models. The accuracy of the PCA model was greater than the PLSDA one. The scores from PC1 were separated into two boundaries, one predicted rejected pears with 100% classi¯cation accuracy, and the other was accepted pears with 92% accuracy. The high accuracy of sugar content determining and watercore detecting by NIR reveal the high e±ciency of NIR technique for detecting other internal qualities of fruit in the future.