Image features extraction and Image similarity comparison.
Project description:
Purpose: Perform Image features extraction; Judgement 3 kinds of similarities between two photos.
- Pixel-level similarity: Each corresponding pixel value of the two images is exactly equal, which is directly manifested by the fact that the two image files are identical in terms of binary content.
- Global visual similarity: After two pixel-level similar images are scaled and compressed respectively, their corresponding pixel values also change to some extent due to scaling or compression, but remain visually identical.
- Feature similarity: Determines whether the features are matching
Output: Print “T” or “F” for above similarity in each level. The interpretation of content we will print show in below rows:
Output; Pixel-level; Global visual; Feature matching
① TTT: Similar in Pixel-level; Similar in Global visual; Similar in Feature matching;
② FTT: Dissimilar in Pixel-level; Similar in Global visual; Similar in Feature matching;
③ FFT: Dissimilar in Pixel-level; Dissimilar in Global visual; Similar in Feature matching;
④ FFF: Dissimilar in Pixel-level; Dissimilar in Global visual; Dissimilar in Feature matching;