Image sample generation method and device, equipment and storage medium

The invention discloses an image sample generation method, apparatus and device, and a storage medium, which overcome the problem of difficult collection of machine vision image samples, and by using the method, a large number of positive sample images can be generated only by collecting a very small number of positive samples, and available negative samples can be generated according to various defect types learned in a data set. Meanwhile, the problem that an existing image sample generation method is poor in stability is solved, the network and the training method used in the method can accurately reserve features of product samples, and nearly real sample images are generated. 本发明公开了一种图像样本生成方法、装置、设备及存储介质，克服了机器视觉图像样本采集困难的问题，使用该方法可以仅采集极少量的正样本，即可生成大量的正样本图像，并根据数据集中学习的各种缺陷类型生成可用的负样本。同时，克服了现有图像样本生成方法稳定性差的问题，该方法使用的网络和训练方法可以准确保留产品样本的特征，生成近乎真实的样本图像。