以同一品种同一批次玉米单倍体和多倍体个体种子为研究对象,基于近红外光谱定性分析技术,分析研究了用漫透射光谱测量方式采集光谱时,不同光照强度和光阑孔径得到的光谱曲线质量以及对玉米单倍体种子鉴别结果的影响.光谱数据经过预处理后,采用PLS+ OLDA特征提取算法,应用SVM建立玉米单倍体鉴别模型.分别统计多种不同实验条件所测数据建立的鉴别模型的正确识别率.结果表明,光阑孔径的大小影响光谱曲线信噪比,越接近玉米籽粒大小的孔径获得的光谱曲线越平滑;光照强度影响光谱曲线有效性,光照越强,波形峰谷差值越大,识别率越高.选择较高的光源强度和适合种子大小的光阑,可以得到高质量的光谱曲线,以此建立的玉米单倍体鉴别模型,识别率达到了100%.
Aiming at the same batches of the same variety corn seeds, this paper studied the influence factors about the spectral curve quality and the recognition rate of maize haploid identification based on NIR qualitative analysis technology. The factors were studied from the two aspects which were the light intensity and diaphragm aperture with the diffuse transmission spectrum measurement way. Partial least squares-discriminant analysis (PLS-OLDA) was used to compress the pretreated spectral data, and then the identification models were built based on Support Vector Machine (SVM). The measured data were recorded under different experimental conditions and the recognition correct rates were calculated. The results showed that the signal to noise ratio of light spectral curve is affected by the amount of light through the diaphragm, and the closer to the seed size the aperture diameter of the diaphragm is, the smoother spectrum would be got. The intensity of light source directly in