Identification Of Potato Varieties Based On Near Infrared Spectroscopy And Detection Of Dry Matter Content

Based on the food supply and demand situation, after rice, wheat, corn, potato we called our new fourth largest staple food. In recent years, our in-depth discussion of the potato staple food of strategic significance, and actively promote the development of the potato industry, therefore, we need to be more scientific and more efficient ways to strictly control every aspect of production. Discrimination on potato varieties and quality testing, the majority are still using traditional laboratory methods, this method is not suitable for real-time analysis of the production process large quantities of samples, and there is destructive. New ways to improve the detection efficiency of potato, are of great significance.NIR spectroscopy is China has gradually developed an analysis technique for nearly two decades, it is efficient, non-destructive analysis of the characteristics are widely used in many fields. According to the current industrial demand of potato, this paper has carried out the research of potato quality detection based on near infrared spectrum analysis:1.The potato(Solanum tuberosum. L) is one of the most important carbohydrate food crops in the China ranking fourth after rice, wheat and maize, and plays a significant role in national economy. There are many varieties of potato, and for each potato variety, the quality such as physical sensory property and chemical components, differ drastically, different potato varieties are suitable for different utilization. Thus, rapid and nondestructive identification of potato cultivars plays an important role in the better use of varieties. Near infrared(NIR) spectroscopy has raised a lot of interest in the classification and identification of agricultural products because it is a rapid and non-invasive analytical technique. In this study, a rapid visible(VIS) and near infrared(NIR) spectroscopic system explored as a tool to measure the diffuse spectroscopy of three different species of potatoes. 352 potato samples(Sample A 142, Sample B 84, Sample C 126) from different sites in Heilongjiang province of China, which were obtained from peddlers market, were randomly divided into two sets at random: Calibration Set and Prediction Set, with 307 samples and 45 samples respectively. The potatoes in the Calibration Set were tested by visible-near infrared spectroscopy method. The spectral data obtained from this test were analyzed by near infrared spectral technology, along with data processing algorithm, i.e., Savitzky-Golay(S-G) smoothing and multiplicative scatter correction(MSC). The spectra data was firstly transformed by multiplicative scatter correction(MSC) to compensate for additive and/or multiplicative effects. In order to reduce the noise components from a raw spectroscopic data set, Savitzky-Golay smoothing and differentiation filter method were introduced. It was proved that, with the soothing segment size of 9, many high frequency noises components can be eliminated. Based on the following analysis by principal component analysis(PCA), partial least square(PLS) regression and back propagation artificial neural network(BP-ANN), a near infrared discrimination model was established. The results obtained from the partial least squares(PLS) analysis showed a positive cumulate reliability of more than 96% for the first four components. The clustering effect was also getting better. After that, twenty absorption peaks extracted from the first four principal components were applied as BP neural network inputs, and a three layers BP neural network [20(input)- 12(implicit)- 3(output)] was constructed, upon which the recognition accuracy of potato varieties for those Prediction Set samples reaches 100%. As a result, the model established in this study can rapidly and accurately identify potato varieties without any destruction, which provides a new way for potato quality detection and variety identification.2.The utility of short-wavelength near-infrared spectroscopy(SWNIS) was assessed as a means of estimating the dry matter concentration(DMC) of potato tubers. In this paper, an attempt has been made to evaluate potato DMC by non-destructive potato samples. A comparative research method has been introduced, in which both sliced and intact tubers of potatoes are employed to assess their DMC by using SWNIS. A total of 207 potato tubers were subdivided into two groups: the first group(n = 115) was longitudinally cut into around 20-mm-thick slices in the middle part of each tubers, the second group(n = 92) was complete potato samples. All those samples were then tested by SWNIS. The spectral results of sliced potato samples were pretreated by the Savitzky-Golay(S-G) method to increase the signal-to-noise ratio without too much distorting of the original signals. Based on the local principle of maxima and minima, and also the vibration of C-H and O-H bonds, five characteristic wavelengths were selected to build the model, with its determinate coefficient(R2) of 0.9416 and its standard deviation(RMSE) of 3.91. Similarly, the spectral results of complete potato samples were processed by derivative S-G method after pretreated by Multiplicative Scatter Correction(MSC). Also five characteristic wavelengths of better linear relations were chosen to build the model, with its R2 of 0.8475 and its RMSE of 4.07. The comparative research results suggest, although not 100% correlative, the non-destructive detection of potato quality by SWNIS might also be a practical method for bulk potato testing of DMC.