Image preprocessing method and device, defect detection method and device and computer equipment

The invention discloses an image preprocessing method and device, a defect detection method and device and computer equipment, and belongs to the field of image processing. Firstly, image samples collected by an unmanned aerial vehicle are obtained, then a target generative adversarial network is ob...

Full description

Saved in:
Bibliographic Details
Main Authors ZHOU SHUAI, YU HONG, XU ZHISONG
Format Patent
LanguageChinese
English
Published 20.12.2022
Subjects
Online AccessGet full text

Cover

Loading…
More Information
Summary:The invention discloses an image preprocessing method and device, a defect detection method and device and computer equipment, and belongs to the field of image processing. Firstly, image samples collected by an unmanned aerial vehicle are obtained, then a target generative adversarial network is obtained according to the image samples, and finally an image sample set is generated according to the target generative adversarial network. According to the scheme, the generative adversarial network is adopted to generate the image sample set, more image sample sets for training can be obtained, data overfitting is avoided, and it is guaranteed that the model obtained through training is good in detection effect. In addition, the generative adversarial network is obtained by training according to the image samples and the category features thereof, so that the obtained image sample set generated by the generative adversarial network has more categories, the problem that only a single category of samples can be gen
Bibliography:Application Number: CN202211180361