Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Smoking detection method based on two-dimensional curved surface fitting

A detection method and technology of two-dimensional curved surfaces, which are applied in the fields of instruments, character and pattern recognition, computer parts, etc., can solve the problems of inconspicuous smoke and inability to meet the real-time performance of the system, so as to reduce product costs, be feasible and simple, and reduce errors. The effect of reporting

Pending Publication Date: 2019-05-24
TIANJIN UNIV +1
View PDF4 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This method is more effective for images with obvious smoke features, but considering that the smoke imaging of the image of the driver smoking is often affected by the light, making the smoke not obvious, there are many interferences in this method
[0005] At present, there are very few published papers and patents on smoking detection methods based on visual features at home and abroad, and they often cannot meet the real-time requirements of the system.

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
  • Smoking detection method based on two-dimensional curved surface fitting
  • Smoking detection method based on two-dimensional curved surface fitting
  • Smoking detection method based on two-dimensional curved surface fitting

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0033] An embodiment of the present invention provides a method for smoking detection based on two-dimensional surface fitting, see figure 1 , the method consists of the following steps:

[0034] 101: Obtain 68 key points and select 7 specific key points from them, draw the facial features frame based on the specific key points, and determine the area of ​​interest for smoking;

[0035] 102: Perform image grayscale processing on the smoking area of ​​interest to obtain a binarized image, then extract connected regions on this basis, obtain a rectangular region, and establish a two-dimensional Gaussian function on the rectangular region;

[0036] 103: Bring the points and gray values ​​of the connected area into the two-dimensional Gaussian function, use the least square method to fit, obtain the variance of the center point and the bright spot, and determine the center point and the bright spot according to the calculated center point and variance The size of the radius, and ...

Embodiment 2

[0046] The following is combined with specific calculation formulas, examples, and Figure 2-Figure 5 The scheme in Example 1 is further introduced, see the following description for details:

[0047] 201: Collect images through the infrared camera in the car;

[0048] Install the infrared camera in the car (the embodiment of the present invention does not limit the model of the infrared camera) in the car and above the console, and ensure that the collected images can make the driver appear in the middle of the image.

[0049] Among them, the camera can be tilted upward (the specific tilt angle can be determined according to the resolution of the camera and the height of the driving console, as long as the complete normal frontal face can be captured).

[0050] 202: Perform face detection on the images collected by the camera, using the Viola Jones detector combined with HAAR (Hal) features and adaboost classification [1] Confirm the location of the driver on the image in t...

Embodiment 3

[0069] Combine below Figure 5 The scheme in embodiment 1 and 2 is carried out feasibility verification, see the following description for details:

[0070] Use face detection and face key point detection technology to locate the position of the facial features, and then determine the smoking area according to the position of the facial features, and perform two-dimensional Gaussian surface fitting in the smoking area. If the fitted surface is not If the feature of the bright spot is met, it is detected that the driver has not smoked, such as Figure 5 As shown in (a), if the fitted Gaussian surface conforms to the characteristics of the bright spot, it is judged that the driver is smoking, such as Figure 5 As shown in (b), at this time give a voice warning not to smoke, Figure 5 The final rendering for smoking detection.

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 smoking detection method based on two-dimensional curved surface fitting, and the method comprises the following steps: obtaining 68 key points, selecting 7 specific key points from the 68 key points, drawing a five-sense-organ frame based on the specific key points, and determining a smoking region of interest; carrying out image graying processing on the region of interest of smoking to obtain a binary image, carrying out connected region extraction on the basis of the binary image to obtain a rectangular region, and establishing a two-dimensional Gaussian functionon the rectangular region; substituting the point and the gray value of the connected region into a two-dimensional Gaussian function, fitting by using a least square method to obtain the variance ofthe central point and the bright spot, determining the central point and the radius of the bright spot according to the calculated central point and the variance, and further judging according to a threshold value to realize smoking detection. According to the invention, the smoking behavior of a driver in the driving process is detected from the perspective of vision, a warning prompt is given out, and various requirements in practical application are met.

Description

technical field [0001] The invention relates to the field of behavior detection of visual features, in particular to a smoking detection method based on two-dimensional surface fitting. Background technique [0002] In recent years, with the development of the economy, the number of vehicles per capita in the world has increased year by year, and the incidence of traffic accidents has also increased. Traffic accidents have become a common concern of the society. Among the causes of many traffic accidents, in addition to overloading, speeding and drunk driving, driver fatigue, smoking and distracted behaviors such as playing with mobile phones are also very common safety hazards. However, if the abnormal behavior of the driver during driving can be detected in advance and corresponding warnings are given in advance, then the driver can be aware of possible dangers, thereby reducing the risk of driving, thereby reducing the occurrence of traffic accidents . [0003] Most of ...

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): G06K9/00G06K9/32G06K9/62
Inventor 褚晶辉张姗王鹏李敏吕卫
Owner TIANJIN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

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