Image retrieval method based on VLAD (vector of locally aggregated descriptors) dual self-adaptation
An image retrieval and self-adaptive technology, applied in special data processing applications, instruments, electrical digital data processing, etc., to achieve high accuracy and strong adaptability
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment Construction
[0020] Below in conjunction with accompanying drawing and specific embodiment, the image retrieval method based on VLAD dual self-adaptation of the present invention is further described:
[0021] As shown in the figure, the present invention first uses the large-scale image database to be retrieved and the rough cluster centers to calculate and save the sum of all descriptors assigned to each cluster center and the number of descriptors; then use the saved Calculate the first adaptive clustering center of the data; use the sum of descriptors, the number of corresponding descriptors and the new clustering center again to recalculate the clustering center for each query image, and obtain VLAD; Finally, the VLAD is normalized twice, and the cosine distance is used to calculate the similarity distance between the query image and the image in the database to be retrieved. After sorting, the first N images are taken as the retrieval result image set.
[0022] Its specific implement...
PUM
Login to View More Abstract
Description
Claims
Application Information
Login to View More 