CN102023231A - Video image wind speed measurement method - Google Patents

Abstract

The invention discloses a method, which relates to a new method for automatically measuring the wind speed in a scene through continuous videos returned by a camera about mark wind wheels in the scene and is applicable to wind speed measurement under a natural condition of good illumination, such as a meadow condition. As computer hardware and software develop, scene state measurement by computer vision has developed into an important subject in computer vision, drawing interest and concern of more and more scholars. The invention makes an attempt to provide the new method for measuring wind speed by computer vision. The method is to measure the wind speed in the scene automatically through the videos returned by the camera about mark wind wheels in the scene. The method expands the application range of the measurement by computer vision. The invention provides the new method for automatically measuring the wind speed in the scene through continuous videos returned by the camera about mark wind wheels in the scene.

Description

Video image measuring wind method

Technical field

The inventive method relates to a kind of continuous videos of passing back by camera about sign wind runner in the scene, measures the new method of wind speed in the scene automatically, applicable to measuring wind speed under the good natural conditions of illumination, as: grassland etc.

Background technology

Traditional measuring wind speed has a variety of, and major part all is based on the method for sensor: hot type anemoscope, three cyphelia eddy-current transducers, three glasss of optocoupler inductor formula sensors.

Development along with calculating software and hardware utilizes computer vision methods that scene state is measured an important topic that develops in the computer vision, and computer vision measurement has caused more and more scholars interest and concern.This paper just utilizes the trial of a kind of new method of computer vision methods measuring wind.

Summary of the invention

The objective of the invention is continuous videos, measure wind speed in the scene automatically about sign wind runner in the scene in order to pass back by camera.Expanded the range of application of computer vision measurement.

The invention discloses the continuous videos that a kind of utilization is passed back by camera, measure the new method of wind speed in the scene automatically about sign wind runner in the scene.

The present invention relates to two parts:

One: the making part of sign wind speed wheel: sketch as shown in Figure 1:

Middle redness circle rotating disk is preferably just round, and color can be a picture, but requires the most of color difference in the figure scene, and is preferably red, can not consider the influence of the shade that illumination brings when utilizing image processing method so again;

Black part on the red disc is not necessarily leaveed no choice but black, but requires with the color of circle rotating disk tangible picture is arranged, and selects black herein for use.

Remaining part does not have special requirement, just does just passable according to appearance.

Two: Digital Image Processing and calculating section: in the process of measuring wind, at first, obtain continuous video image by sign wind runner in the camera photographed scene.Read each frame in the video image then, to each frame carry out binaryzation, gradient distribution processing obtain outline line, obtain the largest contours line and to the largest contours line carry out fitted ellipse, obtain oval center myCenter (midX, midY), major and minor axis lengthh and width, obtain black identification division on the ellipse again, calculate its center coordinate points center (nextX, nextY), the central point center (fristX, fristY) that calculated by previous frame calculates the angle of two frames rotation, calculate angular velocity thus, and then calculate wind speed.

The condition and the step of the inventive method are as follows:

1. equipment needed thereby: imaging system to be corrected, ordinary PC, the wind runner of sign; Wherein imaging system can be camera or video camera etc. (this paper adopt be camera);

2. performing step:

The inventive method (010) part, the pre-service of image obtains outline line.Concrete steps are as follows:

Step 011: get and work as former frame, the pixel that satisfies " redness " in the image is taken as " 255 " otherwise is " 0 ", obtain the image of binaryzation as shown in Figure 2 like this.

Step 012: remove the image after the binaryzation, it is carried out the gradient computing, obtain gradient map and propose all outline lines, result as shown in Figure 3;

The inventive method (020) part obtains the largest contours line and it is carried out fitted ellipse, obtains oval parameter simultaneously.Obtain the density bullet part of elliptic region then.Concrete steps are as follows:

Step 021: obtain all coordinate points on the largest contours line, bring oval general equation ax into ²+ by ²+ cxy+dx+ey=1; Utilize least square, The Fitting Calculation goes out each coefficient.

Step 022: in elliptic region, the pixel that satisfies " black " in the image is taken as " 255 " otherwise is " 0 ",

The inventive method (030) part calculates the rotation angle of this frame and preceding frame, and concrete steps are as follows:

Step C031: the centre coordinate point center (nextX, nextY) that calculates elliptic region " black " part, the central point center (fristX, fristY) that calculates by previous frame, oval central point myCenter (midX, midY), calculate the angle of two frames rotation, calculate angular velocity thus.

Beneficial effect:

The purpose of this invention is to provide a kind of new method that can pass through the computer vision calculation of wind speed.Its advantage is to adopt digital image processing techniques, and method is flexible, is easy to realize that cost is lower; Use this method simple general-purpose simultaneously, for any camera or video camera, the image of arbitrary resolution all can use.

Description of drawings

Fig. 1 identifies the synoptic diagram of wind runner

The binaryzation result of Fig. 2 circle rotating disk part

The extraction synoptic diagram of Fig. 3 profile (blue markings part)

Fig. 4 is to the fitted ellipse of largest contours line

Fig. 5 is to the extraction result of the mark on the circle rotating disk

Fig. 6 routine processes process flow diagram

Claims (1)

1. the objective of the invention is continuous videos, measure wind speed in the scene automatically about sign wind runner in the scene in order to pass back by camera.The invention discloses the continuous videos that a kind of utilization is passed back by camera, measure the new method of wind speed in the scene automatically about sign wind runner in the scene.Wherein system is mainly taken turns by the sign wind speed, the pc machine, and imaging system to be corrected is formed.(the sign wind speed wheel that we adopt is made by oneself, as Fig. 1)

The inventive method step is as follows:

1. performing step:

The inventive method (010) part, the pre-service of image obtains outline line.Concrete steps are as follows:

Step 011: get and work as former frame, the pixel that satisfies " redness " in the image is taken as " 255 " otherwise is " 0 ", obtain the image of binaryzation as shown in Figure 2 like this.

Step 012: remove the image after the binaryzation, it is carried out the gradient computing, obtain gradient map and propose all outline lines, result as shown in Figure 3;

The inventive method (020) part obtains the largest contours line and it is carried out fitted ellipse, obtains oval parameter simultaneously.Obtain the density bullet part of elliptic region then.Concrete steps are as follows:

Step 021: obtain all coordinate points on the largest contours line, bring oval general equation ax^2+by^2+cxy+dx+ey+f=0 into; Utilize least square, The Fitting Calculation goes out each coefficient.

Step 022: in elliptic region, the pixel that satisfies " black " in the image is taken as " 255 " otherwise is " 0 ",

The inventive method (030) part calculates the rotation angle of this frame and preceding frame, and concrete steps are as follows:

Step C031: the centre coordinate point center (nextX, nextY) that calculates elliptic region " black " part, the central point center (fristX, fristY) that calculates by previous frame, oval central point myCenter (midX, midY), calculate the angle of two frames rotation, calculate angular velocity thus.

Images