论文部分内容阅读
针对稻种发芽率传统检测方法周期长,近红外光谱检测技术等无损检测方法受稻种自然颜色及含水量影响大的问题,通过连续偏振光谱结合嵌入型灰色神经网络(IGNN)的方法建立稻种发芽率预测模型。对检测连续偏振光谱运用经典模式分解(EMD)和小波包变换进行去噪处理,根据去噪效果选择EMD去噪。利用主成分分析(PCA)提取去噪后的连续偏振光谱特征,结合偏最小二乘法回归(PLSR)、反向传播神经网络(BPNN)、径向基神经网络(RBFNN)和IGNN分别构建稻种发芽率预测模型,建模结果显示10 min检测时间点IGNN预测模型精度最高,预测集相关系数RP=0.985,预测集均方根误差(RMSEP)为0.771。研究结果表明基于连续偏振光谱技术结合嵌入型灰色神经网络的方法实现稻种发芽率快速无损检测是可行的且精度较高。
In view of the problem that the traditional detection methods of rice seed germination rate are long-term and near-infrared spectroscopy and other non-destructive testing methods are greatly affected by the natural color and water content of the rice varieties, rice is established through continuous polarization spectroscopy combined with an embedded gray neural network (IGNN) Seed germination rate prediction model. The detection of continuous polarized spectrum using classical pattern decomposition (EMD) and wavelet packet transform to denoise, according to denoising effect select EMD denoising. Principal component analysis (PCA) was used to extract continuously polarized spectral features after de-noising, combined with partial least squares regression (PLSR), back propagation neural network (BPNN), RBFNN and IGNN The results showed that the accuracy of IGNN prediction model was the highest at 10 min detection time, the correlation coefficient of prediction set was RP = 0.985, and the root mean square error of prediction set (RMSEP) was 0.771. The results show that it is feasible and accurate to realize fast and nondestructive detection of rice seed germination rate based on continuous polarization spectroscopy combined with embedded gray neural network.