CAJ | 학술논문

针对农业领域植物病虫害检测问题，提出一种基于高清视频图像融合特征的支持向量机（ SVM）的检测方法，实现农业生产中植物病虫害的快速检测。对每幅植物叶片图像的颜色、HSV、纹理和方向梯度直方图四种特征采用基于特征包的多特征融合方法，形成特征向量，并利用SVM分类器进行训练分类。对单特征与融合特征的SVM分类器性能进行试验比较，所提出的方法具有较高的准确率。
침대농업영역식물병충해검측문제，제출일충기우고청시빈도상융합특정적지지향량궤（ SVM）적검측방법，실현농업생산중식물병충해적쾌속검측。대매폭식물협편도상적안색、HSV、문리화방향제도직방도사충특정채용기우특정포적다특정융합방법，형성특정향량，병이용SVM분류기진행훈련분류。대단특정여융합특정적SVM분류기성능진행시험비교，소제출적방법구유교고적준학솔。
For plant pests and diseases detection issue in agriculture field, we propose a detection method to realise the fast detection of plant pests and diseases in agricultural production, which is based on the SVM with the feature of high-definition video image fusion.For four kinds of features of each plant leaf image, the colour, HSV, texture and directional gradient histogram, the method adopts the bag of features-based multi-features fusion approach to form the eigenvector, and uses SVM classifier to train the classification.The method raised in the paper has higher accuracy rate, this is proved by the comparative test between the SVM classifiers with the function of mono-feature and of fusion feature.