首页>
外国专利>
HYPERSPECTRAL IMAGE CLASSIFICATION METHOD AND RELATED DEVICE
HYPERSPECTRAL IMAGE CLASSIFICATION METHOD AND RELATED DEVICE
展开▼
机译:高光谱图像分类方法及相关设备
展开▼
页面导航
摘要
著录项
相似文献
摘要
Embodiments of the present application disclose a hyperspectral image classification method and a related device. The method comprises: first determining a training sample set and a sample set to be classified of a target hyperspectral image, then using a class ablation strategy, and generating K training subsets by means of the training sample set; selecting, by using the K training subsets and a preset selection strategy, a second preset number of pixel points from the sample set to be classified, adding the pixel points into the training sample set to update the training sample set, and updating the sample set to be classified; and finally, performing model training by using the updated training sample set so as to obtain a first image classification model, and predicting, by using the first image classification model, the updated sample set to be classified so as to obtain first classification prediction information of each sample to be classified, thereby realizing ground object classification of the target hyperspectral image. The target hyperspectral image is processed by means of multiple views, such that the accuracy of small sample classification can be effectively enhanced; and an active learning method based on class ablation can be adaptive to an inputted target hyperspectral image.
展开▼