recently i wrote one php spider progrom just for downloading images , then i find i have had download to many duplicated images
as one poor student , i got to find the way clean the copy of them .
this is what i found by digging web for related Pin Point The Same Images
while wrriting and digging the main point move from Pin point same Images to Transform images into Linear/Static/Geomatry problems
Structure
Algorithm | Base theory | Advance Apply |
---|---|---|
Perceptual hash algorithm | fingerprint | pHash / SIFT |
Color histogram | transform | Otsu’s method |
PPMCC | linear dependence | |
TF-IDF | statistics dependence | |
Cosine similarity | geomatry dependence |
Info
1, Perceptual hash algorithm
2, Color histogram
3, Pearson product-moment correlation coefficient
4, Term frequency–Inverse document frequency
5, Cosine similarity
update 2016/05/20