Biological characteristic storage optimization method and device, electronic equipment and storage medium

Abstract

The invention provides a biological feature storage optimization method and device, electronic equipment and a storage medium, belongs to the field of information storage, and aims at solving the problem that the face retrieval speed is low when the data size is large. The method comprises the following steps: classifying biological characteristic data to be stored according to biological characteristic library fields to obtain classified biological characteristic data; partitioning a storage medium according to the biological feature library field; and according to a set rule, performing corresponding processing on the classified biological characteristic data, and writing a processed result into a corresponding partition. The device comprises a classification module, a partition module and a write-in module. According to the optimization method and device provided by the invention, the retrieval and query speed is increased, the user experience is improved, the disk storage space is reduced, the total storage amount is increased, and the hardware resource cost is saved.

Description

technical field

[0001] The present application belongs to the field of information storage, and in particular relates to a biological feature storage optimization method, device, electronic equipment and storage medium. Background technique

[0002] With the continuous development of face recognition technology, the commercial use of face recognition is becoming more and more extensive, which provides convenience to our daily life. However, with the continuous expansion of user groups, the diversification of usage scenarios, and the Accumulation has put forward relatively high requirements for the retrieval performance of face recognition services.

[0003] At present, in the existing technology, when the amount of data in the database is getting larger and larger, the face retrieval speed is slow and cannot meet the requirements. If you want to accommodate a large amount of data in the database, you need to increase the storage space of the hardware disk. Such a solution wi...

Examples

Embodiment 1

[0056] Take the data storage of the attendance machine of a company as an example. The company has about 1,000 employees. The attendance machine is an attendance machine that uses fingerprints as the basis for identification. The storage table includes: biometric field and biometric data, and the biometric field includes: fingerprint picture, job number, department name, attendance record;

[0057] Classify the feature storage table according to fingerprint pictures, job numbers, department names, and attendance records;

[0058] The storage medium is partitioned according to the biometric feature database field, and its storage medium partition is also: fingerprint picture, job number, department name, attendance record, and the partition method adopts Range partition;

[0059] According to the set rules, the classified biometric data are processed accordingly, and the processed results are written into corresponding partitions. The setting rules in this embodiment include: ...

Embodiment 2

[0064] This embodiment takes the data storage of the medical image query system as an example. A general hospital has a large amount of medical image data every day, such as CT images, 3D color Doppler ultrasound, X-ray chest films, etc., and may store many pictures at the same time , this large amount of picture data storage will bring huge pressure to the hardware system. If no measures are taken, the hardware storage cost will be very expensive. This embodiment will use the first rule, the second rule, and the third rule. Azimuth improves storage capacity, reduces storage space, reduces hardware costs, and can increase retrieval speed.

[0065] First extract the feature storage table of the medical image query system, and classify the feature storage table according to the fields of the biometric database. The feature storage table includes: biometric fields and biometric data. The biometric fields described in this embodiment include but are not limited to : CT image, 3D c...

Embodiment 3

[0074] This embodiment takes the data storage of the face recognition access control system as an example. There are more than 1,000 permanent residents in a community, and the owners need to enter the faces of themselves and their family members into the database of the face recognition access control system. When entering the community, the face recognition access control system will match the current face data of the owner with the face data saved in the database. If the match is successful, the gate will be opened for the owner to allow the owner to pass. If successful, an alarm is issued and the owner is not allowed to go along. Its characteristic is that only facial feature data is frequently used data, while other data has been entered into the system, but generally does not need to be called, and its retrieval speed needs to be considered to meet real-time needs.

[0075] Firstly, the feature storage table is retrieved from the face recognition access control system, a...

technical field

[0001] The present application belongs to the field of information storage, and in particular relates to a biological feature storage optimization method, device, electronic equipment and storage medium. Background technique

[0002] With the continuous development of face recognition technology, the commercial use of face recognition is becoming more and more extensive, which provides convenience to our daily life. However, with the continuous expansion of user groups, the diversification of usage scenarios, and the Accumulation has put forward relatively high requirements for the retrieval performance of face recognition services.

[0003] At present, in the existing technology, when the amount of data in the database is getting larger and larger, the face retrieval speed is slow and cannot meet the requirements. If you want to accommodate a large amount of data in the database, you need to increase the storage space of the hardware disk. Such a solution wi...

Examples

Embodiment 1

[0056] Take the data storage of the attendance machine of a company as an example. The company has about 1,000 employees. The attendance machine is an attendance machine that uses fingerprints as the basis for identification. The storage table includes: biometric field and biometric data, and the biometric field includes: fingerprint picture, job number, department name, attendance record;

[0057] Classify the feature storage table according to fingerprint pictures, job numbers, department names, and attendance records;

[0058] The storage medium is partitioned according to the biometric feature database field, and its storage medium partition is also: fingerprint picture, job number, department name, attendance record, and the partition method adopts Range partition;

[0059] According to the set rules, the classified biometric data are processed accordingly, and the processed results are written into corresponding partitions. The setting rules in this embodiment include: ...

Embodiment 2

[0064] This embodiment takes the data storage of the medical image query system as an example. A general hospital has a large amount of medical image data every day, such as CT images, 3D color Doppler ultrasound, X-ray chest films, etc., and may store many pictures at the same time , this large amount of picture data storage will bring huge pressure to the hardware system. If no measures are taken, the hardware storage cost will be very expensive. This embodiment will use the first rule, the second rule, and the third rule. Azimuth improves storage capacity, reduces storage space, reduces hardware costs, and can increase retrieval speed.

[0065] First extract the feature storage table of the medical image query system, and classify the feature storage table according to the fields of the biometric database. The feature storage table includes: biometric fields and biometric data. The biometric fields described in this embodiment include but are not limited to : CT image, 3D c...

Embodiment 3

[0074] This embodiment takes the data storage of the face recognition access control system as an example. There are more than 1,000 permanent residents in a community, and the owners need to enter the faces of themselves and their family members into the database of the face recognition access control system. When entering the community, the face recognition access control system will match the current face data of the owner with the face data saved in the database. If the match is successful, the gate will be opened for the owner to allow the owner to pass. If successful, an alarm is issued and the owner is not allowed to go along. Its characteristic is that only facial feature data is frequently used data, while other data has been entered into the system, but generally does not need to be called, and its retrieval speed needs to be considered to meet real-time needs.

[0075] Firstly, the feature storage table is retrieved from the face recognition access control system, a...

Images