Photo classification method
A classification method, photo technology, applied in still image data clustering/classification, still image data retrieval, instruments, etc., can solve problems such as no way to classify
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0054] Embodiment 1, there are 100 old photos, and 900 digital photos, first remake the old photos, obtain 100 remade photos, modify the shooting time in the 100 remade photos to the old photo shooting time, get 100 A remake photo with information, and then put 100 remake photos with shooting time information and the remaining 900 digital photos into the photo library, and compare the similarity of the avatar pictures in the high-quality library through the first method, that is First establish a face database, compare the features of the avatar pictures in the high-quality library with the feature combinations of the avatar pictures in the face database to determine whether they are the same person, and set the features of the avatar pictures in the high-quality library to be compared with the feature combinations of the face database. If the similarity of the feature combination of the avatar picture in the high-quality database is ≥80%, it is considered to be the same person...
Embodiment 2
[0061] Embodiment 2, there are 100 old photos, and 900 digital photos, first remake the old photos, obtain 100 remade photos, modify the shooting time in the 100 remade photos to the shooting time of the old photos, get 100 remake photos with shooting time information, then put 100 remake photos with shooting time information and the remaining 900 digital photos into the photo library, and use the first method to compare the similarity of the avatar pictures in the high-quality library Comparison, that is to compare the avatar pictures in the high-quality library with each other in turn to determine whether they are the same person, and set the similarity ≥ 80%, then it is considered the same person, set the similarity < 80%, and ≥ 50, then It is considered that they may be the same person, and if the similarity is set to <50%, then it is determined that they are not the same person; when the face database is compared with the corresponding similar database, if the number of av...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 
