Local target re-recognition method, device and system

A technology of re-recognition and target, applied in the field of image recognition, can solve the problems of low re-recognition accuracy, easy occlusion of pedestrians, and low matching accuracy.

Active Publication Date: 2019-01-15
BEIJING KUANGSHI TECH
View PDF6 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, in special scenes such as in front of the subway gate, in front of the checkout counter of an unmanned supermarket, and high-density crowds, pedestrians are easily blocked, causing only part of the pedestrian's body to be exposed in the image. This ima

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Local target re-recognition method, device and system
  • Local target re-recognition method, device and system
  • Local target re-recognition method, device and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0036] First, refer to figure 1 An example electronic device 100 for implementing a local target re-identification method, device and system according to an embodiment of the present invention will be described.

[0037] Such as figure 1 Shown is a schematic structural diagram of an electronic device. The electronic device 100 includes one or more processors 102, one or more storage devices 104, an input device 106, an output device 108, and an image acquisition device 110. These components pass through a bus system 112 and / or other forms of connection mechanisms (not shown). It should be noted that figure 1 The components and structure of the electronic device 100 shown are only exemplary, not limiting, and the electronic device may also have other components and structures as required.

[0038] The processor 102 can be implemented in at least one hardware form of a digital signal processor (DSP), a field programmable gate array (FPGA), and a programmable logic array (PLA)...

Embodiment 2

[0045] see figure 2 The flow chart of a local target re-identification method is shown, the method can be executed by the electronic device provided in the foregoing embodiment, and the method specifically includes the following steps:

[0046] Step S202, acquiring a whole-body image of an existing target and a partial image of the target to be identified. Wherein, the target mentioned in this embodiment may be a pedestrian, and of course it may also be other categories such as a vehicle, an animal, or an object, which is not limited here. The whole body image shows all the body features of the target, and the partial image only shows some body features of the target.

[0047] The whole-body image of an existing target can be obtained directly from the image base library, or can be obtained from a camera arranged at a designated position. It can be understood that this embodiment can obtain one or more whole-body images of different existing targets. The partial image of th...

Embodiment 3

[0086] For the partial target re-identification method provided in Embodiment 2, the embodiment of the present invention provides a partial target re-identification device, see Figure 7 A structural block diagram of a local target re-identification device shown, the device includes the following modules:

[0087] An image acquisition module 702, configured to acquire a whole-body image of an existing target and a partial image of the target to be identified;

[0088] The model processing module 704 is used to input the whole-body image and the partial image to the pre-trained local object re-identification model, so that the partial object re-identification model performs affine transformation processing on the whole-body image based on the partial image, and obtains the affine transformation of the whole-body image image, and determine the feature spacing between the affine image and the local image;

[0089] The target matching module 706 is configured to determine whether...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention provides a local target re-recognition method, a device and a system, which relate to the technical field of image recognition. The method comprises the following steps: obtaining a whole-body image of an existing target and a local image of a target to be recognized; obtaining a local image of a target to be recognized; obtaining a whole-body image of an existing target. The whole body image and the local image are input to the pre-trained local target re-recognition model, so that the local target re-recognition model performs affine transformation on the whole body image basedon the local image, obtains the affine image of the whole body image, and determines the feature distance between the affine image and the local image; whether the target to be recognized matches with the existing target is judged based on the feature spacing. The invention can effectively improve the matching accuracy of the local image and the whole body image, thereby improving the re-identification accuracy.

Description

technical field [0001] The present invention relates to the technical field of image recognition, in particular to a local object re-identification method, device and system. Background technique [0002] In many applications of video structuring, object analysis is very important, especially the identification of objects plays a central role in many fields such as security and video retrieval. Currently, there are identity re-identification (Reid) technologies such as person re-identification and vehicle re-identification. Taking pedestrian re-identification as an example, this technology can find the person in the images of pedestrians collected by the city's huge camera network based on an image of the pedestrian to be tracked. However, in special scenes such as in front of the subway gate, in front of the checkout counter of an unmanned supermarket, and high-density crowds, pedestrians are easily blocked, causing only part of the pedestrian's body to be exposed in the i...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
IPC IPC(8): G06K9/00G06K9/62
CPCG06V40/103G06V40/10G06F18/22G06F18/214
Inventor 罗浩张弛
Owner BEIJING KUANGSHI TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap