A similar image query method, device, system and storage medium
A similar image and query method technology, applied in the field of systems and storage media, similar image query methods, and devices, can solve the problems of several minutes or even tens of minutes and low query efficiency
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0027] figure 1 It is a schematic flow chart of a similar picture query method provided by Embodiment 1 of the present invention. This embodiment is applicable to the case where the coding information of the current picture is determined through the picture query model, and at least one of the current picture and the database server is determined according to the coding information. The similarity of a stored picture is used to determine the similarity of the current picture. This method can be executed by a similar picture query device, wherein the device can be implemented by software and / or hardware, and is generally integrated in the query server. For details, see figure 1 As shown, the method may include the following steps:
[0028] S110. Acquire the current picture, input the current picture into the pre-trained picture query model, and determine the encoding information of the current picture based on the feature information output by the picture query model.
[0029...
Embodiment 2
[0039] figure 2 It is a schematic flow chart of a similar image query method provided in Embodiment 2 of the present invention. The technical solution of this embodiment adds new steps on the basis of the foregoing embodiments. Optionally, the training method of the picture query model includes: creating an original deep residual network, wherein the original deep residual network includes a first extraction module and a second extraction module; inputting training original pictures into the In the original deep residual network, a predicted marked picture is obtained; according to the predicted marked picture and the training marked picture corresponding to the training original picture, the network parameters of the original deep residual network are adjusted to obtain the target deep residual network; the obtained The first extraction module in the target deep residual network is determined as the image query model. For the parts not described in detail in the method emb...
Embodiment 3
[0056] Figure 5 It is a schematic structural diagram of a device for searching for similar pictures provided by Embodiment 3 of the present invention. Such as Figure 5 As shown, the device includes: a coding information determining module 31 , a similarity determining module 32 and a similar picture determining module 33 .
[0057] Wherein, the coding information determination module 31 is used to obtain the current picture, input the current picture into the pre-trained picture query model, and determine the coding information of the current picture based on the feature extraction module of the picture query model;
[0058] A similarity determination module 32, configured to determine the similarity between the current picture and at least one stored picture in the database server based on the encoding information of the current picture;
[0059] A similar picture determining module 33, configured to determine a similar picture of the current picture from the stored pictu...
PUM
Abstract
Description
Claims
Application Information
- R&D Engineer
- R&D Manager
- IP Professional
- Industry Leading Data Capabilities
- Powerful AI technology
- Patent DNA Extraction
Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.
© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com