Rapeseed Variety Recognition Based on Hyperspectral Feature Fusion

As an important oil crop, rapeseed contributes to the food security of world. In recent years, agronomists have cultivated many new varieties, which has increased human nutritional needs. Variety recognition is great importance for yield improvement and quality breeding. view low efficiency damage traditional methods, in this paper, we develop a noninvasive model varieties based on hyperspectral feature fusion. Three types image features, namely, multifractal feature, color characteristics, trilateral parameters, are fused together identify 11 species. An optimal selected using simple rule, then three kinds features fused. The support vector machine kernel method employed as classifier. average rate reaches 96.35% 93.71% distinguishing two species species, respectively. abundance test demonstrates that our possesses robustness. high almost independent number modeling samples classifiers. This result can provide some practical experience guidance rapid varieties.