High-dimensional image feature matching method and device

A high-dimensional image feature matching device, comprising:

• a memory;
• and
• a processor coupled to the memory, wherein the processor is configured to perform a high-dimensional image feature method on a basis of instructions stored in the memory, comprising:
• extracting a high-dimensional image feature of an image to be retrieve&
• dividing the high-dimensional image feature of the image to be retrieved into a plurality of low-dimensional image features;
• determining a nearest clustering center at each layer of low-dimensional image features of images in a database corresponding to each of the low-dimensional image features of the image to be retrieved, by comparing each of the low-dimensional image features of the image to be retrieved with clustering centers at each layer, wherein the determined nearest clustering center is nearest to the corresponding low-dimensional image feature;
• and
• determining a similarity of the low-dimensional image feature between the image to be retrieved and each of some images in the database according to the determined nearest clustering center, so that at least one high-dimensional image feature matching the high-dimensional image feature of the image to be retrieved is retrieved in the database.

A computer readable storage medium on which computer program instructions are stored, which when executed by a processor implement a high-dimensional image feature method, comprising:

• extracting a high-dimensional image feature of an image to be retrieved;
• dividing the high-dimensional image feature of the image to be retrieved into a plurality of low-dimensional image features;
• determining a nearest clustering center at each layer of low-dimensional image features of images in a database corresponding to each of the low-dimensional image features of the image to be retrieved, by comparing each of the low-dimensional image features of the image to be retrieved with clustering centers at each layer, wherein the determined nearest clustering center is nearest to the corresponding low-dimensional image feature;
• and
• determining a similarity of the low-dimensional image feature between the image to be retrieved and each of some images in the database according to the determined nearest clustering center, so that at least one high-dimensional image feature matching the high-dimensional image feature of the image to be retrieved is retrieved in the database.

The computer-readable storage medium according to 14, wherein determining a similarity of the low-dimensional image feature between the image to be retrieved and each of the images in the database according to the determined nearest clustering center, so that at least one high-dimensional image feature matching the high-dimensional image feature of the image to be retrieved is retrieved in the database comprises:

• determining a feature vector of each of the images associated with the determined nearest clustering center in the database and a feature vector of the image to be retrieved, by calculating the low-dimensional image features of the images in the database and the low-dimensional image features of the image to be retrieved, using a term frequency-inverse document frequency algorithm, and according to the determined nearest clustering center;
• determining a similarity score of the low-dimensional image feature between the image to be retrieved and each of the images associated with the determined nearest clustering center in the database, according to the distance between the feature vector of each of the images associated with the determined nearest clustering center in the database and the feature vector of the image to be retrieved;
• and
• obtaining N images with the highest score in the database using a quick select algorithm combined with an insertion sort algorithm, and making the high-dimensional image features of the N images with the highest score as features matching the high-dimensional image feature of the image to be retrieved.