) C2a ) lc C ul al tceu la th te t m he an m s ean s and atnhdetshtea nd s a ta rd n dard t he im th aegiemsaignedsaitnabdaasteab to as geettothgeet th im e ag iemsam ge ostmco lo sstec lose : (13) (13) (12) (14) textu 0.5.

doi:10.1201/b18660-90

Chapter

) C2a ) lc C ul al tceu la th te t m he an m s ean s and atnhdetshtea nd s a ta rd n dard t he im th aegiemsaignedsaitnabdaasteab to as geettothgeet th im e ag iemsam ge ostmco lo sstec lose

Chapter

ABSTRACT

The system extracts an image randomly as a retrieval example. First, the system calculates the feature vector of the image extracted; then the feature vector matches with the feature vectors of all images. Finally, return the images top 12 and range them from the top down, from the left right by the similarity. The experimental results is shown in Figure 2.