Criminisi image restoration method based on threshold boundary and region division
The invention discloses a Criminisi image restoration method based on threshold demarcation and region division, which is characterized by comprising the following steps: firstly, improving a priority function through a threshold demarcation mode, then introducing a Jaccard distance as a constraint...
Saved in:
Main Authors | , , , , , |
---|---|
Format | Patent |
Language | Chinese English |
Published |
30.11.2021
|
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | The invention discloses a Criminisi image restoration method based on threshold demarcation and region division, which is characterized by comprising the following steps: firstly, improving a priority function through a threshold demarcation mode, then introducing a Jaccard distance as a constraint condition of a matching criterion, and finally, optimizing a search strategy through a local segmentation mode, so as to effectively solve the problems that a confidence item rapidly tends to zero in iteration and a search strategy is complex. According to the method, through the improved threshold boundary priority function, the problems that the image restoration sequence is unreasonable and the final restoration effect is poor due to the fact that the confidence coefficient item tends to zero after multiple iterations of a traditional Criminsi algorithm can be effectively solved, and local texture information of the image can be better restored by using the size of the self-adaptive sample block.
本发明公开了一种基于阈值分界和 |
---|---|
Bibliography: | Application Number: CN202110816774 |