Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Image fusion quality detection method and device

A quality inspection method and image fusion technology, applied in the field of image processing, can solve problems such as the failure to track people and the inability to accurately understand the situation of the person being photographed, and achieve a good positioning effect

Active Publication Date: 2020-08-07
深圳市车宝信息科技有限公司
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, artificial inference cannot accurately understand the situation of the subject to a large extent, which leads to the failure of character tracking

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
  • Image fusion quality detection method and device
  • Image fusion quality detection method and device
  • Image fusion quality detection method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0043] Embodiment 1 of the present application provides an image fusion quality detection method, such as figure 1 shown, including:

[0044] Step 110, searching for the image frame of the tracked person from each frame of the video image;

[0045] Since the embodiment of the present application has high real-time requirements for the tracked person, a deep convolutional neural network model with fast calculation speed is used for image judgment;

[0046] Among them, the image frame of the tracked person is searched from each frame of the video image, specifically: constructing a deep convolutional neural network model, starting from the input layer, and sequentially going through the convolution C 1 layer (output image size [256,256,8]), depthwise convolutional layer D 1 (output image size [128,128,16]), convolutional layer C 2 (output image size [64,64,32]), depth convolutional layer D 2 (output image size [32,32,64]), convolutional layer C 3 (output image size [16,16,1...

Embodiment 2

[0083] Embodiment 2 of the present application provides an image fusion quality detection device, such as Figure 5 shown, including:

[0084] The tracked person's image search module 510 is used to search for the tracked person's image frame from each frame of the video image;

[0085] The image fusion module 520 is used to perform image fusion on multiple tracked person images according to the wavelet transform image fusion method, the contour wavelet fusion method and the scale-invariant feature transform image fusion method;

[0086] The fusion image quality detection module 530 based on the standard deviation is used to calculate the standard deviation respectively according to the fusion results, and determine the image fusion quality according to the standard deviation.

[0087] Wherein, the tracked person image search module 510 is specifically used to construct a deep convolutional neural network model; starting from the input layer, sequentially passing through the ...

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 discloses an image fusion quality detection method and device. The method comprises the following steps: searching a tracked person image frame from each frame of video images; respectively carrying out image fusion on the plurality of tracked person images according to a wavelet transform image fusion method, a contour wavelet fusion method and a scale invariant feature transform image fusion method; respectively calculating standard deviations according to the fusion result, and judging the image fusion quality according to the standard deviations. By adopting the image fusionquality detection method provided by the invention, the quality of image fusion is detected by calculating the standard deviations of three different image fusion modes, so the tracked person is better positioned.

Description

technical field [0001] The present application relates to the technical field of image processing, in particular to an image fusion quality detection method and device. Background technique [0002] In video security monitoring, the tracking and positioning of people is the most important issue. However, when identifying a person, it is easy to appear that the person's location cannot be easily recognized due to being blocked by an object. [0003] In the prior art, it is generally used to find one or more pictures that can best reflect the face of the person from the video to manually infer the facial features of the person being photographed, so as to realize the person tracking. However, artificial inference cannot accurately understand the situation of the subject to a large extent, which leads to the failure of character tracking. More and more existing technologies have abandoned manual recognition methods and introduced various automatic image recognition methods, b...

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): G06T7/00G06T7/11G06T7/136G06T5/50G06N3/04
CPCG06T7/0002G06T7/11G06T7/136G06T5/50G06T2207/10016G06T2207/20064G06T2207/20221G06T2207/20076G06T2207/30168G06N3/047G06N3/045
Inventor 徐小君李慧
Owner 深圳市车宝信息科技有限公司
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
Eureka Blog
Learn More
PatSnap group products