Mobile visual retrieval system based on key feature descriptor selection

Abstract

The invention discloses a mobile visual retrieval system based on key feature descriptor selection, and belongs to the field of mobile visual retrieval. The mobile visual retrieval system based on key feature descriptor selection is characterized by utilizing an affinity propagation clustering method to perform key feature descriptor selection on the feature descriptors of images to be queried, using a method of transmitting and then weighing and matching the key feature descriptors, reducing data transfer volume through key feature descriptor selection to reduce network transmission time delay and further achieving the aim of improving user experience. Meanwhile, the method can effectively eliminate noise point data of the images to be queried, thereby reducing the data transfer volume and effectively improving the retrieval accuracy at the same time.

Description

technical field [0001] The invention belongs to the field of mobile vision and relates to a mobile vision retrieval system based on key feature descriptor selection. It involves a method of selecting key feature descriptors for image features by using an affinity propagation (Affinity Propagation, AP) clustering method, and transmitting key feature descriptors and performing weighted image matching. Background technique [0002] With the rapid development of software and hardware technology, mobile visual retrieval applications have developed rapidly. The ever-increasing number of users and the improvement of users' requirements for service quality have challenged the traditional mobile visual retrieval service based on the client / server model. In the traditional client / server mode, the user transmits the images taken by mobile clients such as mobile phones to the server through the wireless network, and then the server retrieves the images to return the relevant informatio...

AI Technical Summary

Problems solved by technology

On the one hand, when there are a large number of feature points in the query image, which also corresponds to a large number of feature descriptors, the amount of transmitted data in this way is still high, and the resulting network delay still needs to be reduced. On the other hand, the query image The importance of each feature point in the feature points is different, and there are some noise points among them, which will reduce the accuracy of image retrieval. How to avoid transmitting useless noise point feature descriptors is another problem worth studying.

Images