欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
考虑到化学防治方法在田间病虫害防治问题上的不足,为了更好的结合机器视觉和生物防治方法,改善病虫害防治效果,针对棉蚜天敌图像的识别,提出了基于局部特征提取的昆虫识别方法。通过提取图像对象突出特征点,在特征点空间上比较特征点描述符的欧氏距离来度量不同类别对象之间的相似性。讨论了SIFT和SURF两种局部特征特征提取算子,实验结果表明,基于SURF算子的方法比基于SIFT算子的识别效果更好,识别速度更快,并具有很好的鲁棒性。
英文摘要：
Considering the lack of chemical control about the prevention and cure of insects in the field,for improving the effect of the prevention and cure with the combination of computer vision and biological control.Local features techniques,which measure the similarity of the different objects in interest-points space with Euclidean distance by extracting interest points of objects,are introduced to recognition of insects according to the recognition of the natural enemies of Aphis Gossypii Glover.Two methods of local features technique based on SIFT（scale invariant feature transform） and SURF（speeded up robust features） are discussed,and the experimental result show that the method based on SURF is more efficient,faster and well robust.
同期刊论文项目
同项目期刊论文
期刊信息
位置：