Seed vigor detection of sweet corn by near infrared spectroscopy under high temperature stress

The purpose of this study was to establish a new method for rapid and nondestructive detection of sweet corn seed vigor (including germination percentage, germination index and vigor index) by using near infrared spectroscopy. Eight sweet corn varieties were used as test materials and the seeds of each variety were divided into 5 sets with total of 40 samples. All samples were oven-dried at 50℃, then one set of the sweet corn seeds of each variety was taken out at 1 day, 2 days, 3 days, 4 days and 5 days after drying for the near infrared spectroscopy detection and investigation of biological index. A quantitative model of maize correlation index was established by using the partial least square regression (PLSR) method. The results showed that the correlation coefficient for calibration (R_c), correlation coefficient for prediction (R_p), correlation coefficient for cross validation (R_cv), root mean square error for calibration (RMSEC), root mean square error for prediction (RMSEP) and root mean square error for cross validation (RMSECV) were 0.8952, 0.8605, 0.8503, 6.6490, 6.6270, 7.9091 for germination percentage, respectively. The parameters of PLSR model of germination index, Rc, Rp, Rcv, RMSEC, RMSEP and RMSECV were 0.9710, 0.9366, 0.9427, 7.4373, 24.7921 and 10.5003, respectively. The relative parameters of the vigor index were 0.9109, 0.8962, 0.8711, 0.8674, 0.9489 and 1.0393, respectively. The method was simple, fast, accurate and reliable, and was suitable for rapid and nondestructive evaluation of sweet corn germination rate, germination index, vigor index and other seed germination indexes. This study provides a new rapid method for high vigor corn seeds screening with high in the future. (authors)