Object Classification Method and Device Based on Differential Chain Code Histogram

A target classification and differential chain technology, applied in the field of pattern recognition, can solve problems such as complex calculations and affecting real-time performance of video analysis, and achieve the effect of less calculation

Active Publication Date: 2019-03-29
SHENZHEN ZTE NETVIEW TECH
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The inventors of the present invention found when studying the prior art that in the object classification method characterized by the aspect ratio of the object shape, the contour perimeter ratio, etc., usually when the video capture device has a visual change, a certain In addition, in the target classification method characterized by moments such as Hu invariant moments, although it has strong translation, scale and rotation invariance, its operation is complicated, and a certain To a certain extent, it affects the real-time performance of video analysis

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
  • Object Classification Method and Device Based on Differential Chain Code Histogram
  • Object Classification Method and Device Based on Differential Chain Code Histogram
  • Object Classification Method and Device Based on Differential Chain Code Histogram

Examples

Experimental program
Comparison scheme
Effect test

Embodiment approach

[0055] In a preferred embodiment, in the step S101, the step of performing offline classifier training on the training image set to obtain the corresponding target category discriminant function and its parameters includes:

[0056] S1011. Perform target feature extraction on each foreground target in the training image set;

[0057] S1012. Perform target classifier calibration on each foreground target in the training image set;

[0058] S1013. Taking the calibrated target classifier and its corresponding target features as input parameters, perform offline training, so as to obtain a corresponding target class discriminant function and its parameters.

[0059] Under a preferred embodiment, in the step S1011 or S103, the processing steps of performing feature extraction on the detected foreground target include:

[0060] Step 1. Perform boundary chain code expression on the foreground target;

[0061] Step 2, performing a first-order difference on the chain code;

[0062] ...

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 a target classification method based on a differential chain code histogram and a device thereof. The method comprises the steps that off-line classifier training is performed on a training image set so that a corresponding target class discrimination function and parameters thereof are obtained; prospect target detection is performed on the inputted video sequence images; target characteristic extraction is performed on the detected prospect target; and the extracted target characteristics act as input, and on-line class discrimination is performed on the prospect target in the video sequence images by utilizing the target class discrimination function and the parameters thereof which are obtained via training. First-order differential chain code histogram characteristics of the prospect target are extracted, calculation amount is low and the characteristics are invariant in target translation, scale and rotation.

Description

technical field [0001] The invention belongs to the field of pattern recognition, relates to the technical fields of image processing and computer vision, and in particular relates to a method and device for classifying objects based on differential chain code histograms in video monitoring. Background technique [0002] Target classification is an important content in intelligent video analysis technology. In the practical application of intelligent video analysis, different foreground areas in video sequence images may correspond to different moving targets. In order to better understand the behavior of different targets, we use It is necessary to correctly associate class signs with target areas, for example, in order to understand the behavior of people in parking lots, or to calculate the flow speed of different types of vehicles on highways, etc. [0003] At present, object classification has a wide range of applications in intelligent security monitoring, traffic road...

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
Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/62
Inventor 徐庆华林彬吴贻刚
Owner SHENZHEN ZTE NETVIEW 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