A video analysis-based real-time people flow monitoring method suitable for a dense scene
A real-time monitoring and video analysis technology, applied in the field of information processing, can solve the problems that cannot be solved in the actual scene of the airport, and achieve the effect of improving accuracy
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
Method used
Image
Examples
Embodiment 1
[0093] A method for real-time monitoring of crowd flow in dense scenes based on video analysis, see figure 2 , including the following steps:
[0094] S1: Foreground extraction that eliminates lighting effects, including:
[0095] S11: Connect to the airport security platform according to the H264 standard protocol, and extract a frame of background image every second.
[0096] S12: Use the mixed Gaussian model extracted every frame to perform background learning on the background image, for example, perform learning and extraction every 10 seconds, and learn 100 frames of images in total.
[0097] S13: Input the background image (set as B), the original image of the foreground (set as O), grayscale the background image B and the original image O of the foreground (set as H respectively B 、H O ); Calculate the normalized histogram of the background image B and the foreground original image O and calculate the average brightness (set as m B , m O ).
[0098] S14: Determi...
Embodiment 2
[0115] Embodiment 2 On the basis of Embodiment 1, combined with the collected images of the airport, the application of the present invention will be described in detail.
[0116] (1) Extract the foreground image.
[0117]Use the mixed Gaussian model to extract the background image B, and extract and learn every 10 frames of images;
[0118] Set the pixel value I of each pixel in the background image at time t t (x, y) is described by K (the value of K is 5) Gaussian models; then the probability P(I t (x, y)) is:
[0119]
[0120] Among them, x, y are the abscissa and ordinate of the pixel; i∈[1,2,...,K]; and are the weight, mean and variance of the i-th Gaussian model of the pixel point (x, y) at time t, respectively, and N represents the vector composed of the pixel value at time t, the mean and variance of the i-th Gaussian model.
[0121] After the mixed Gaussian background model of each pixel is established, it can be used to judge whether each pixel of the curr...
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