HIGHLIGHTS
SUMMARY
The process of cloud removal can be regarded as a type of image reconstruction, which is heavily dependent on precise cloud detection. Generative adversarial networks have shown promise in cloud removal and surface information reconstruction, but they do not recognize when thin and thick clouds coexist, resulting in significant disparities between the output and original image. The authors propose the "SCMCNN" (Saliency Cloud Matting Convolutional Neural_Network) model that integrates deep learning techniques with image matting for remote sensing image cloud recognition, cloud opacity estimation, and cloud removal. The cloud-matting dataset outperforms the commonly . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.