Target self-adaptive hiding method based on feature space statistical information mapping
The invention provides a target self-adaptive hiding method based on feature space statistical information mapping. The method comprises: using a deep convolution learning mode, on the basis of target and background region division by using significance target detection, respectively obtaining a tar...
Gespeichert in:
Hauptverfasser: | , , , , , , , , |
---|---|
Format: | Patent |
Sprache: | chi ; eng |
Schlagworte: | |
Online-Zugang: | Volltext bestellen |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | The invention provides a target self-adaptive hiding method based on feature space statistical information mapping. The method comprises: using a deep convolution learning mode, on the basis of target and background region division by using significance target detection, respectively obtaining a target feature space and feature space statistical information representing background style features, establishing a mapping network of the feature space statistical information between a target image and a background region image, the target image having the feature space statistical information of the background region image, and obtaining a target self-adaptive hidden image through boundary fusion. According to the method, the mapping network about the characteristic space statistical information between the target image and the background area image is established, the task target of target self-adaptive hiding is carried out, a corresponding solution is provided for the task target, a good result is obtained, an |
---|