CN100470215C - Video Wave Measurement Method and Measurement System - Google Patents

Abstract

The video wave measurement method and measurement system involved in the present invention can perform more convenient video digital image processing on the collected sea surface images to obtain more wave measurement parameters. The measurement method includes: collecting sea surface images (S1), preprocessing image enhancement sea surface textures (S2), isotropically clipping sea surface images (S30), performing Radon transformation on clipped images (S31), extracting wavelength parameters (S32 ), stack the time stack images along the wave propagation orientation (S40), cut the time stack images isotropically (S41), perform Radon transformation (S42) on the cut time stack images, and extract wave direction, wave velocity, period parameters ( S43), outputting wave direction, wave velocity, wavelength, period parameters (S5). Its measurement system includes a sea surface image acquisition device, an image processing and wave parameter extraction device, and a measurement result output device.

Description

Video Wave Measurement Method and Measurement System

technical field

The invention relates to ocean wave measurement technology, in particular to a method and system for performing wave measurement by using continuous frame images.

Background technique

The generation and movement of ocean waves is one of the most common physical phenomena in the ocean. Wave measurement is very important for ocean engineering design, marine transportation and fishing operations, marine environment forecasting, and marine scientific research.

The traditional ocean wave monitoring equipment is to put the wave measuring sensor directly into the seawater, and measure the ocean wave parameters by detecting the movement of seawater particles and the water level change of the sea surface. Traditional ocean wave monitoring equipment adopts a single-point measurement method with a small measurement range. Even if the measuring instruments are deployed at multiple points, the measuring range is limited, and the integrity of the measuring data is poor.

With the development of remote sensing and telemetry technology, the method of measuring ocean wave parameters using video images has been further applied. Using video images to measure ocean parameters, compared with traditional ocean monitoring equipment, has the advantages of large measurement range and strong integrity. At present, images obtained by various methods include satellite remote sensing images, aerial images taken by aircraft, images taken by ships sailing, images taken by shore or platform fixed camera devices, etc., which can be used for ocean wave parameter measurement.

The application of video images can not only record the static information of sea surface waves, but also record the dynamic information of sea surface waves, and can measure the motion parameters of waves. In the image of the sea surface, the different brightness of the crest and trough constitutes the texture structure of the image. In particular, the breaking of waves generally occurs at the crest, and the broken wave and white waves are high-brightness pixels in the image, forming a wave crest in the vertical direction of the wave propagation direction. The linear texture features of valleys can be used to obtain wave motion parameters by studying wave video images.

The research work of video image wave measurement that has been reported so far includes: by detecting the breaking waves and white waves generated in wave motion, and tracking the wave peak band with higher brightness value in image processing to obtain the motion parameters; through Fourier transform A method for detecting frequency spectrum and correlation functions in wave motion to obtain motion parameters; a method for obtaining direction parameters by separating spectrum anisotropy in wave motion through wavelet (Wavelet) transform.

However, the video image wave measurement method in the prior art has problems such as complicated calculation and few measurement parameters.

Contents of the invention

Aiming at the problems existing in the video image wave measurement method of the above-mentioned prior art, the present invention introduces a new video wave measurement method for more convenient processing of digital video wave images, and its purpose is to perform Radon transformation on isotropic cut images (Radon transform) and perform Radon transform on the time stack image to obtain the wave parameters of wave direction, wave velocity, wavelength and wave period. Simultaneously, the present invention also introduces a video wave measurement system implementing the above video wave measurement method.

The video wave measurement method involved in the present invention is to process the sea surface wave image taken by a CCD camera by a computer and extract wave parameters such as wave direction, wave velocity, wavelength, and wave period. The video wave measuring method comprises the following steps:

1. Collect sea surface images

Use a CCD camera to take images of the sea surface, and then transfer the image data to a computer.

2. Preprocessing images to enhance sea surface texture

The sea surface image is preprocessed, and for the darker image, the method of enhancing the contrast is adopted to enhance the texture of the sea surface and improve the image quality.

3. Isotropically cropped images

Cut the texture-enhanced sea surface image to the largest available circular image within the original image range to achieve isotropic clipping of the sea surface image.

4. Perform a Radon transform on the image

A Radon transform is performed on the isotropically cropped image to transform the digital image into a projected image in the Radon transform domain.

Specifically: on the two-dimensional image plane, define a coordinate system, take the x-axis as the coordinate axis in the horizontal direction, take the y-axis as the coordinate axis in the vertical direction, and f(x, y) is the image brightness at (x, y) . Radon transform is a transformation method that calculates the projection of the image function f(x, y) on the same two-dimensional image plane along the direction of the specified angle ray. The projection of the image function f(x, y) is its line integral in a certain direction, and the result of integrating the pixel brightness in the image is the projection intensity in the Radon transform domain. For example, the two-dimensional line integral of f(x, y) in the vertical direction is the projection of f(x, y) on the x-axis; the two-dimensional line integral of f(x, y) in the horizontal direction is f(x , y) projection on the y-axis. The projection of the function f(x, y) can be calculated along any angle θ from 0° to 360°, that is to say, there is a Radon transformation of the function f(x, y) at any angle. Assuming that the coordinate system composed of x-axis and y-axis is centered on the origin on the same plane and rotated at an angle θ, the x-axis rotated at an angle θ is the x' axis, and the y-axis is rotated at an angle θ to be the y' axis, so x The 'axis and the y' axis form a new coordinate system. Here, the Radon transformation of f(x, y) at an angle of θ is defined as a projection line of f(x, y) parallel to the y' axis on the x' axis integral:

${\mathrm{<span\; class="notranslate">\; R</span>}}_{\mathrm{<span\; class="notranslate">\; \&theta;</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; \&prime;</span><span\; class="notranslate">\; )</span><span\; class="notranslate">\; =</span>{<span\; class="notranslate">\; \&Integral;</span>}_{<span\; class="notranslate">\; -</span><span\; class="notranslate">\; \&infin;</span>}^{<span\; class="notranslate">\; \&infin;</span>}\mathrm{<span\; class="notranslate">\; f</span>}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; \&prime;</span>\mathrm{<span\; class="notranslate">\; cos</span>}\mathrm{<span\; class="notranslate">\; \&theta;</span>}<span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; the\; y</span>}<span\; class="notranslate">\; \&prime;</span>\mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&theta;</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; \&prime;</span>\mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&theta;</span>}<span\; class="notranslate">\; +</span>\mathrm{<span\; class="notranslate">\; the\; y</span>}<span\; class="notranslate">\; \&prime;</span>\mathrm{<span\; class="notranslate">\; cos</span>}\mathrm{<span\; class="notranslate">\; \&theta;</span>}<span\; class="notranslate">\; )</span>\mathrm{<span\; class="notranslate">\; dy</span>}<span\; class="notranslate">\; \&prime;</span>$

In the formula:

$\left[\begin{array}{c}\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; \&prime;</span>\\ \mathrm{<span\; class="notranslate">\; the\; y</span>}<span\; class="notranslate">\; \&prime;</span>\end{array}\right]<span\; class="notranslate">\; =</span>\left[\begin{array}{cc}\mathrm{<span\; class="notranslate">\; cos</span>}\mathrm{<span\; class="notranslate">\; \&theta;</span>}& \mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&theta;</span>}\\ <span\; class="notranslate">\; -</span>\mathrm{<span\; class="notranslate">\; sin</span>}\mathrm{<span\; class="notranslate">\; \&theta;</span>}& \mathrm{<span\; class="notranslate">\; cos</span>}\mathrm{<span\; class="notranslate">\; \&theta;</span>}\end{array}\right]\left[\begin{array}{c}\mathrm{<span\; class="notranslate">\; x</span>}\\ \mathrm{<span\; class="notranslate">\; the\; y</span>}\end{array}\right]$

The Radon transform with the angle θ as a variable makes a projection image in the Radon transform domain. The horizontal axis is the angle θ, and the vertical axis is x′. The intensity R _θ (x′) of the projection line integral of the Radon transform is the Radon transform domain. brightness of the projected image.

5. Extract wavelength parameters

Extract the wavelength parameter on the wave propagation orientation (Orientation).

In the projection image in the Radon transform domain, detect the maximum intensity R _θ (x′), and extract the angle θ _M corresponding to the maximum intensity R _θ (x′) in the Radon transform domain, where θ _M is the wave propagation orientation. There is an ambiguity of ±180° in the θ _M angle, which may be along the direction of wave propagation or against the direction of wave propagation.

To extract the wavelength parameter on the wave propagation orientation, the angle θ _M along the wave propagation orientation is selected, that is, the intensity peak or valley is detected within the intensity corresponding to the angle θ _M in the direction parallel to the wave forward path, and the peak distance or valley value is calculated spacing to get the wavelength.

6. Stack images when stacking

From the sea surface image with enhanced texture, according to the corresponding angle θ _M in the parallel direction of the wave forward path, intercept line-shaped image slices in the direction of wave propagation, and then intercept line-shaped image slices in the same way for consecutive adjacent frames at the same position, And for the linear image slices of consecutive adjacent frames, according to the time sequence of the original image frames, similar to the stacking operation process of a computer, the linear image slices are stacked one by one, and integrated into a time-stack (time-stack) image.

7. Perform isotropic clipping on timestack images

Clips the largest available circular image within the image extent of the 2D time-stack image plane.

8. Perform Radon transformation on the time stack image

Radon transform is performed on the clipped time stack image, and the digital image is transformed into a projection image in the Radon transform domain.

9. Extract wave direction, wave velocity, period parameters

In the projected image within the Radon transform domain, detect the maximum intensity;

Extract the angle corresponding to the maximum intensity in the Radon transform domain, and determine the wave propagation direction according to the angle value and the wave propagation orientation, which is the wave direction;

The angle corresponding to the maximum intensity extracted in the Radon transform domain is calculated according to the angle value corresponding to the slope of the peak propagation direction in the time stack image to obtain the wave velocity; then calculate the quotient of the wavelength and the wave velocity to obtain the period.

10. Output wave parameters

According to the geodetic scale and direction angle corresponding to the actual imaging, the extracted wave parameters are converted to actual geodetic coordinates, and displayed and printed on the output device. The output wave parameters are wave direction, wave speed, wavelength and period.

The video wave measurement system involved in the present invention includes a sea surface image acquisition device, an image processing and wave parameter extraction device, and a measurement result output device.

The sea surface image acquisition device is a CCD camera, with USB, IEEE1394 (Firewire) standard data transmission mode or embedded network transmission mode, which can directly and quickly transmit image data to the computer. The sea surface image acquisition device is installed on lighthouses, ocean platforms, near-shore buildings or supports above the sea surface, and the shooting lens is aimed at the sea surface for measurement, and the sea surface image sequence of the measurement area is taken.

The image processing and wave parameter extraction device is composed of a microcomputer, which has the functions of implementing the above-mentioned video wave measurement method: preprocessing the image to enhance the sea surface texture, isotropically cutting the image, stacking the time stack image, and executing Radon on the sea surface image and the time stack image. Transform, and extract wave parameters such as wavelength, wave direction, wave velocity, and period in the wave propagation orientation.

The measurement result output device is a display and a printer connected to the microcomputer of the image processing and wave parameter extraction device, which can convert wave parameters such as wave direction, wave speed, wavelength, period, etc. geodetic coordinates.

The video wave measurement method and measurement system involved in the present invention can perform more convenient video digital image processing on the collected sea surface images to obtain more wave measurement parameters.

Description of drawings

Fig. 1 is a flow chart of the video wave measuring method involved in the present invention.

Fig. 2 is a block diagram of the video wave measurement system involved in the present invention.

Figure 3 is a wave simulation image diagram.

Figure 4 is an image obtained by isotropically cutting the image in Figure 3 .

Fig. 5 is an intensity map obtained by performing Radon transformation on the image in Fig. 4 .

FIG. 6 is a graph of intensity curves obtained from angles corresponding to peak intensity detected in FIG. 5 .

Fig. 7 is a time stack image obtained by intercepting line-like image slices along the wave propagation orientation and stacking them.

Fig. 8 is an image obtained by isotropically cutting the image in Fig. 7 .

FIG. 9 is an intensity map obtained by performing Radon transformation on the image in FIG. 8 .

Explanation of marks in the attached drawings:

S _1. Acquisition of sea surface images S _2. Preprocessing images

S ₃₀ , cutting the image isotropically S ₃₁ , performing Radon transformation on the image

S ₃₂ , extract wavelength parameters S ₄₀ , stack image when stacking

S _41. Isotropically cut the time stack image S _42. Perform Radon transformation on the time stack image

S _43. Extract wave direction, wave speed, period S _5. Output wave parameters

θ, the angle of Radon transformation x′, the projected linear distance of Radon transformation

R', the intensity of the peak intensity corresponding to the angle

Detailed ways

The present invention is described in further detail now in conjunction with accompanying drawing.

Fig. 1 shows the flowchart of the video wave measurement method involved in the present invention. 3 to 9 show images processed during the application of the video wave measurement method of the present invention.

As shown in Figure 1, the video wave measurement method includes the following steps:

S ₁ — Acquisition of sea surface images

Fig. 3 is a wave simulation image, which is a frame of numerical simulation image according to the linear wave theory, the brightest pixel represents the peak, and the darkest pixel represents the trough.

S ₂ — preprocessing image enhancement sea surface texture

S ₃₀ — Isotropically cropped image

Since the collected images have isotropic or anisotropic features, in order to make the method of the present invention isotropic, the image is carried out based on circular isotropic clipping, and the largest available circular clipping is performed within the image range image. Figure 4 is an image obtained by isotropically cutting the image in Figure 3 .

S ₃₁ — Perform Radon transformation on the cropped image

The Radon transformation has the characteristic that the projection is symmetrical along 180° in a 360° circle, so the calculation is simplified within the range of 180°. FIG. 5 is a projection image in the Radon transform domain obtained by performing Radon transform on the image in FIG. 4 . The origin of the digital image coordinate axis is in the center of the image, and the digital image coordinate axis is rotated counterclockwise with an angle increment of 1° centered on the origin, and the digital image is transformed into a projection of the Radon transform domain within 180°.

S ₃₂ —Extract wavelength parameter

In the projected image within the Radon transform domain, the maximum intensity is detected. In FIG. 5 , a peak appears at an angle θ ₁ =123°, and the angle corresponding to the maximum intensity in the Radon transform domain is extracted. In Fig. 5, for the angle θ ₁ =123° extracted according to the texture direction, it cannot be determined whether the wave propagates from the upper left to the lower right, or from the lower right to the upper left, so there is an ambiguity of ±180° in this angle, which may be along The direction of wave propagation, it may also be the direction of reverse wave propagation, this time is to obtain the orientation along wave propagation. Figure 6 is the angle intensity curve obtained by detecting the angle corresponding to the peak intensity in Figure 5. The intensity peak or valley is detected within the intensity corresponding to the angle along the wave propagation orientation, and the peak or valley distance is calculated to obtain a wavelength of L = 32.29.

S ₄₀ —Stack image when stacked along wave propagation orientation

Line-shaped image slices are intercepted along the wave propagation direction, and then the line-shaped image slices are intercepted in the same way for consecutive adjacent frames at the same position, and the linear image slices of consecutive adjacent frames are sequenced according to the time sequence of the original image frames. Similar to the stacking operation process of a computer, the line image slices are stacked one by one and integrated into a time stack image. Fig. 7 is a time stack image obtained by intercepting line-like image slices along the wave propagation orientation θ ₁ =123° and stacking them, and the time sequence is from top to bottom.

S ₄₁ — Isotropic clipping of time stack images

Performs isotropic image clipping on timestacked images, clipping the largest available circular image within the image extent. Fig. 8 is an image obtained by isotropically cutting the image in Fig. 7 .

S ₄₂ — Perform Radon transform on the clipped timestack image

Radon transform is performed on the clipped time stack image. FIG. 9 is a projection image in the Radon transform domain obtained by performing Radon transform on the image in FIG. 8 .

S ₄₃ —Extract wave direction, wave speed, period

In the projected image within the Radon transform domain, the maximum intensity is detected. In Fig. 9, a peak appears at an angle θ ₂ =172°; the angle corresponding to the maximum intensity in the Radon transform domain is extracted. In Fig. 9, according to the angle value of angle θ ₂ =172° and the orientation θ ₁ =123° along wave propagation, the direction of wave propagation is determined, specifically, tanθ ₂ =-0.14 is calculated, and the sign is negative, that is, the direction of wave propagation is determined. The direction is from bottom right to top left, and the wave direction is determined to be 123°. The maximum intensity extracted in the Radon transform domain corresponds to the angle θ ₂ =172°, and the slope tanθ ₂ =-0.14 corresponding to the propagation direction of the wave crest in the time stack image is calculated according to the angle value, and the wave velocity = 0.14, period = 229.72 .

S ₅ — Output wave parameters: wave direction, wave speed, wavelength, period

Output wave parameters: wave direction = 123°, wave speed = 0.14, wavelength = 32.29, period = 229.72. According to the geodetic scale and direction angle corresponding to the actual imaging, the wave parameters can be converted to the actual geodetic coordinates.

Fig. 2 is a block diagram of the video wave measurement system involved in the present invention. As shown in Figure 2, the components of the video wave measurement system include a sea surface image acquisition device, an image processing and wave parameter extraction device, and a measurement result output device.

The sea surface image acquisition device can directly and quickly transmit the image data to the microcomputer that constitutes the image processing and wave parameter extraction device, and the measurement result output device is a display and a printer connected to the microcomputer of the image processing and wave parameter extraction device.

Claims (9)

1, a kind of video wave measuring method is characterized in that comprising: gather sea level chart picture (S ₁), pretreatment image strengthens sea texture (S ₂), isotropy is sheared sea level chart picture (S ₃₀), to cut image carry out Radon conversion (S ₃₁), extract wavelength parameter (S ₃₂), stack image (S when on the wave propagation orientation, piling up ₄₀), to the time stack image isotropy shear (S ₄₁), to cut time stack image carry out Radon conversion (S ₄₂), extract wave direction, velocity of wave, cycle parameter (S ₄₃), output wave is to, velocity of wave, wavelength, cycle parameter (S ₅).

2, video wave measuring method according to claim 1 is characterized in that, gathers sea level chart picture (S ₁) be to take the sea level chart picture with ccd video camera, then image data transmission is arrived computing machine.

3, video wave measuring method according to claim 1 is characterized in that, image is carried out Radon conversion (S ₃₁) be that the image that isotropy is sheared is carried out the Radon conversion, be projected image in the Radon transform domain with digital service unit.

4, video wave measuring method according to claim 1 is characterized in that, extracts wavelength parameter (S ₃₂) be on wave propagation orientation, to extract wavelength parameter, with the corresponding angle θ of wave progresses path parallel direction _MIntensity in detected intensity peak value or valley, calculate peak separation or valley spacing to draw wavelength.

5, video wave measuring method according to claim 1 is characterized in that, stack image (S when piling up ₄₀) be corresponding angle θ according to wave progresses path parallel direction _MIntercepting line strip image slices on the wave propagation orientation, again the continuous adjacent frame is used the same manner intercepting line strip image slices at the same position place, and to the linear image slices of continuous adjacent frame, time sequencing according to original picture frame is one by one piled up the linear image slices, stack image when being integrated into.

6, video wave measuring method according to claim 1 is characterized in that, to the time stack image carry out Radon conversion (S ₄₂) be to isotropy shear circular the time stack image carry out the Radon conversion, be projected image in the Radon transform domain with digital service unit.

7, video wave measuring method according to claim 1 is characterized in that, extracts wave direction, velocity of wave, cycle parameter (S ₄₃) comprising: in the projected image in the Radon transform domain, detect maximum intensity; Extract the angle of maximum intensity correspondence in the Radon transform domain, determine that according to this angle value with on the wave propagation orientation direction of wave propagation is wave direction; According to the angle value of the maximum intensity correspondence of extracting in the Radon transform domain, calculate corresponding the time the slope of the crest direction of propagation in the stack image, obtain velocity of wave; Calculate the merchant of wavelength and velocity of wave again, obtain the cycle.

8, video wave measuring method according to claim 1 is characterized in that, output wave parameter (S ₅) be according to pairing the earth yardstick of actual imaging and orientation angle, the wave parameter of extracting is converted to actual terrestrial coordinate, and shows on output unit and printing that the wave parameter of output is wave direction, velocity of wave, wavelength, cycle.

9, a kind of video wave measuring system of implementing video wave measuring method according to claim 1 is characterized in that, comprises sea image collecting device, Flame Image Process and wave parameter extraction element and measurement result output unit; The sea image collecting device is a ccd video camera, has the data transfer mode or the embedded network transmission mode of USB, IEEE1394 (Firewire) standard, view data can be transferred to computing machine directly, fast; Flame Image Process and wave parameter extraction element are made of microcomputer, have the function of implementing above-mentioned video wave measuring method; The measurement result output unit is display and the printer that is connected with the microcomputer of Flame Image Process and wave parameter extraction element, according to pairing the earth yardstick of actual imaging and orientation angle wave parameters such as wave direction, velocity of wave, wavelength, cycle is converted to actual terrestrial coordinate.

Images