A fringe projection dynamic measurement method suitable for detecting foreign matter in coal mine belt conveyors

Abstract

The present invention discloses a fringe projection dynamic measurement method suitable for detecting foreign objects in coal mine belt conveyor transmission. The method includes the following steps: 1. Designing three-step phase-shifted fringes and encoding them into three RGB channels, synthesizing a color phase-shifted fringe, projecting it onto the surface of the foreign object, modulating it with the foreign object, and collecting the color-deformed fringes with a camera, and separating the three-channel information from it; 2. Using the phase shift method to calculate the wrapped phase of the separated RGB three-channel three-step phase-shifted fringes; 3. Selecting the phase-shifted fringes of one of the channels and using Fourier transform profilometry to calculate the preliminary unwrapped phase; 4. Using the preliminary unwrapped phase in step 3 as the basis to calculate the fringe order of the wrapped phase in step 2, and correctly unwrapping the wrapped phase according to the fringe order to obtain the foreign object modulation phase, and further measuring three-dimensional information. The method of the present invention can realize object phase calculation using only a single color fringe, and is suitable for the dynamic measurement requirements of foreign object detection on the surface of belt conveyors.

Description

Technical Field

[0001] The present invention belongs to the technical field of fringe projection three-dimensional measurement, and in particular relates to a fringe projection dynamic measurement method suitable for detecting foreign matter in transmission of a coal mine belt conveyor. Background Art

[0002] Coal mines are complex transportation environments. During transportation, conveyor belts are susceptible to damage from foreign objects such as anchors, barbed wire, and large pieces of gangue. This can lead to economic losses and even safety accidents. Therefore, foreign object detection during coal mine conveyor belt transport is extremely important. Traditional detection methods primarily include manual inspection, radar detection, and metal detection. Manual inspection is inefficient and poses safety risks, radar detection is expensive and difficult to maintain, and metal detection is limited in its detection capabilities. Furthermore, image recognition technology is currently a hot research area and has demonstrated promising detection results. However, image recognition technology relies on two-dimensional images. Using three-dimensional information to increase the amount of information about foreign object features would lead to even better detection results.

[0003] Fringe projection 3D measurement technology offers advantages over traditional measurement methods, such as speed, low cost, and non-contact capabilities. It is widely used in industrial inspection, cultural relic preservation, medical diagnosis, and reverse engineering. Standard sine and cosine fringes are projected onto the surface of the object being measured using a digital projector. After being modulated by the object, a camera captures the deformed fringe information and transmits it to a computer. Using algorithms such as phase calculation and height information conversion, the 3D surface profile of the object is obtained.

[0004] Phase calculation includes two parts: wrapped phase calculation and phase unwrapping. Since the wrapped phase is obtained using the inverse tangent function, its value folds around (-π,π], so the folded phase needs to be unfolded into a continuous phase. Phase calculation methods mainly include phase-shift profilometry and Fourier transform profilometry. Phase-shift profilometry has high measurement accuracy and is widely used for three-dimensional measurement of static objects. However, it requires a large number of fringe patterns and is not suitable for dynamic measurement. Fourier transform profilometry can calculate the unwrapped phase using only a single fringe pattern, so it can be used for dynamic measurement. However, the selection of the spectrum will lose detailed information about the object, resulting in poor phase accuracy. To address these issues, in Fourier transform profilometry, researchers have mainly studied filter design and phase unwrapping algorithms in FTP to improve measurement accuracy. In phase-shift profilometry, researchers have improved the dynamic performance of PSP by reducing the number of images and using high-speed equipment to accelerate image acquisition. Summary of the Invention

[0005] Purpose of the Invention: To address the shortcomings of traditional coal mine belt conveyor foreign body detection in terms of detection efficiency, cost, and detection effectiveness, and to compensate for the shortcoming of image processing methods with limited information content, the present invention proposes the use of fringe projection to measure the three-dimensional information of objects on coal mine belt conveyors, thereby increasing the amount of foreign body feature information and improving detection effectiveness. In response to the dynamic target measurement needs of belt conveyors, and to address the problem that traditional phase shift profilometry requires a large number of fringe patterns that are unsuitable for dynamic measurement, as well as the problem that Fourier transform profilometry loses details in phase calculation and has poor accuracy, the present invention provides a fringe projection dynamic measurement method suitable for detecting foreign bodies in coal mine belt conveyor transmission.

[0006] Technical solution: To achieve the above-mentioned purpose, the present invention adopts the following technical solution: a fringe projection dynamic measurement method suitable for detecting foreign matter in coal mine belt conveyor transmission, specifically comprising the following steps:

[0007] Step 1: Design an N-step phase-shifted fringe pattern. The projector projects N fringe patterns onto the surface of the foreign object in sequence. The fringe pattern is modulated by the foreign object to produce deformation. The foreign object information is contained in the deformed fringe pattern. The camera captures each deformed fringe pattern. The grayscale value of the deformed fringe pattern is expressed as:

[0008]

[0009] Where (x, y) is the pixel coordinate in the fringe pattern, a(x, y) is the background light intensity, b(x, y) is the modulation intensity, f0 is the fringe frequency, is the foreign body modulation phase, i represents the phase shift of the i-th step, N is the total number of phase shift steps, g i (x,y) is the grayscale value of pixel (x,y) in the i-th phase-shift fringe pattern;

[0010] Step 2: Use the phase shift method to calculate the foreign body wrapping phase ψ(x,y) from the deformed fringe pattern. The calculation formula is:

[0011]

[0012] According to the above inverse tangent operation, the foreign body encapsulation phase is cut off in the range of (-π,π];

[0013] Step 3: Select one of the deformed fringe patterns and calculate the initial unwrapped phase from the selected deformed fringe pattern using Fourier transform profilometry The calculation steps of Fourier transform profilometry are as follows:

[0014] According to Euler's formula cosθ=[exp(iθ)+exp(-iθ)] / 2, formula (1) is expressed as follows:

[0015] g(x,y)=a(x,y)+c(x,y)+c *(x,y) (3)

[0016] Where:

[0017]

[0018] Performing Fourier transform on formula (3) yields:

[0019] G(f)=A(f)+C(f-f0)+C * (f+f0) (4)

[0020] The spectrum after Fourier transform contains A(f), C(f-f0), C * (f+f0), A(f) represents the background signal at zero frequency, C(f-f0) represents the spectral component at frequency f0, and C * (f+f0) represents the spectral component at frequency -f0, where C(f-f0), C * (f+f0) contains the phase information of the foreign object. Extract the C(f-f0) term and perform inverse Fourier transform to obtain:

[0021]

[0022] The Fourier transform profilometry wrapped phase is calculated based on the real and imaginary parts:

[0023]

[0024] Wrapping the Fourier transform profilometry around the phase ψ FTP (x,y) is expanded using the spatial phase unwrapping algorithm to obtain the initial unwrapped phase

[0025]

[0026] Step 4: Based on the preliminary phase The fringe order K is calculated by the wrapping phase ψ(x,y) using the following formula:

[0027]

[0028] The foreign body modulation phase is obtained by expanding the wrapped phase ψ(x,y) according to the fringe order K. The calculation formula is:

[0029]

[0030] Furthermore, the number of phase-shifting fringes in step 1 is 3, and the three-step phase-shifting fringes can be encoded into the RGB channels of the color fringe image respectively. The object phase calculation can be realized using only a single color fringe image, which is suitable for three-dimensional measurement of dynamic objects.

[0031] Furthermore, the three-dimensional information of the foreign matter is calculated based on the phase information according to the method described in step 1, and the foreign matter on the surface of the belt conveyor is detected based on the three-dimensional information.

[0032] Beneficial effects: Compared with the prior art, the technical solution of the present invention has the following beneficial technical effects:

[0033] The present invention proposes a fringe projection dynamic measurement method suitable for detecting foreign objects in the transmission of coal mine belt conveyors. It only needs to project a color fringe image to achieve three-dimensional measurement of the object. The three-step phase-shifted fringes are respectively encoded into the RGB channels of the color fringe image, projected onto the surface of the foreign object and the deformed fringe image is collected. The phase shift method is first used to calculate the wrapped phase of the three-step phase-shifted fringes. Then, the Fourier transform profilometry is used to calculate the preliminary unfolding phase of the phase-shifted fringes of one of the channels. The fringe order of the wrapped phase calculated by the phase shift method is thus calculated, and the wrapped phase is correctly unfolded according to the fringe order. Compared with Fourier transform profilometry, the method proposed by the present invention has higher calculation accuracy, is insensitive to the filter window used in Fourier transform, and has better robustness. Compared with traditional phase shift profilometry, the method proposed by the present invention only needs to collect a single color fringe image, which is suitable for the dynamic measurement needs of foreign object detection on the belt conveyor surface. Foreign objects on the belt conveyor surface are detected based on three-dimensional information, and the characteristic information of foreign objects on the belt conveyor surface is enhanced, which is conducive to improving the detection effect. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 This is a flow chart of a fringe projection dynamic measurement method for detecting foreign matter in coal mine belt conveyor transmission, proposed by the present invention;

[0035] Figure 2 The RGB channel of the color phase shift fringe pattern generated by simulation of the present invention

[0036] (a) is the color phase-shift fringe pattern R channel generated by simulation of the present invention;

[0037] (b) is the color phase-shift fringe pattern G channel generated by simulation of the present invention;

[0038] (c) is the color phase-shift fringe pattern B channel generated by simulation of the present invention;

[0039] Figure 3 It is a simulation object of the present invention;

[0040] Figure 4 The RGB channels of the color deformation fringe image simulated and generated by the present invention;

[0041] (a) is the color deformation fringe image R channel generated by simulation of the present invention;

[0042] (b) is the color deformation fringe image G channel generated by simulation of the present invention;

[0043] (c) is the color deformation fringe image B channel generated by simulation of the present invention;

[0044] Figure 5 is the parcel phase diagram calculated by the method proposed in the present invention;

[0045] Figure 6 is the preliminary unwrapped phase map calculated using Fourier transform profilometry;

[0046] Figure 7 is the fringe order map calculated by the method proposed in the present invention;

[0047] Figure 8 is the modulation phase diagram of the simulated object calculated by the method proposed in the present invention;

[0048] Figure 9 is the modulation phase diagram of the simulated object after subtracting the carrier information calculated by the method proposed in the present invention;

[0049] Figure 10 Schematic diagram of foreign body detection during belt conveyor transmission according to the method proposed by the present invention;

[0050] In the figure: 1. Coal flow; 2. Foreign matter; 3. DLP projector; 4. Industrial camera; 5. Fixed bracket; 6. Belt conveyor. DETAILED DESCRIPTION

[0051] The specific embodiments of the present invention will be further described below with reference to the accompanying drawings and examples:

[0052] The flow chart of the fringe projection dynamic measurement method for detecting foreign matter in coal mine belt conveyor transmission proposed by the present invention and its relationship with traditional phase shift profilometry and Fourier transform profilometry methods are shown in the figure. Figure 1 As shown, the specific steps include:

[0053] Step 1: Build a fringe projection measurement system, such as Figure 10 As shown in the figure, the industrial camera and DLP projector are fixed on the top of the belt conveyor through a fixed bracket to perform three-dimensional measurement of the coal flow below and detect foreign objects in it. The effect of the proposed method is verified by simulation in a computer. The simulation generates a three-step phase shift fringe pattern and encodes it into the three RGB channels to generate a color fringe pattern. The schematic diagram is shown in the figure. Figure 2 As shown, the simulation object uses a 20-fold peaks function as shown in Figure 3 As shown, the color fringe pattern is projected onto the surface of the simulated object. After being modulated by the simulated object, the camera collects the color deformed fringe pattern and separates the RGB three-channel information from it. The simulated RGB three-channel deformed fringe pattern is shown in Figure 4 shown.

[0054] Step 2: Calculate the wrapped phase of the phase-shifted fringe pattern using a three-step phase-shift algorithm. In formula (2), N = 3, and the formula is:

[0055]

[0056] The calculated wrapping phase is as follows Figure 5 shown.

[0057] Step 3: Select the first phase shift fringe pattern and use Fourier transform profilometry to calculate the initial unwrapped phase. In formula (1), N = 3. According to the Euler formula cosθ = [exp(iθ) + exp(-iθ)] / 2, it can be expressed as

[0058] g(x,y)=a(x,y)+c(x,y)+c * (x,y)

[0059] In the formula

[0060]

[0061] Perform Fourier transform on the fringe pattern to obtain

[0062] G(f)=A(f)+C(f-f0)+C * (f+f0)

[0063] The spectrum after Fourier transform contains A(f), C(f-f0), C * (f+f0), A(f) represents the background signal at zero frequency, C(f-f0) represents the spectral component at frequency f0, and C * (f+f0) represents the spectral component at frequency -f0, where C(f-f0), C * (f+f0) contains the phase information of the simulated object. Extract C(f-f0) and perform inverse Fourier transform to get:

[0064]

[0065] The Fourier transform profilometry wrapped phase is calculated based on the real and imaginary parts:

[0066]

[0067] Wrapping the Fourier transform profilometry around the phase ψ FTP (x,y) is expanded using the spatial phase unwrapping algorithm to obtain the initial unwrapped phase Initial phase expansion Figure 6 shown.

[0068] Step 4: Based on the preliminary phase The fringe order K is calculated by the wrapping phase ψ(x,y) using the following formula:

[0069]

[0070] Calculate the stripe level as Figure 7 shown.

[0071] The modulation phase of the simulated object is obtained by expanding the wrapped phase ψ(x,y) according to the fringe order K. The calculation formula is:

[0072]

[0073] The calculated modulation phase of the simulated object is as follows Figure 8 As shown, since the calculated simulation object modulation phase value is Contains fringe carrier information 2πf0x, which is inconvenient to observe the shape of the object. Therefore, the fringe pattern is projected without a simulated object to calculate the carrier information. The modulation phase of the simulated object obtained after subtracting the carrier information is as follows: Figure 9 shown.

Claims

1. A fringe projection dynamic measurement method suitable for detecting foreign matter in coal mine belt conveyor transmission, characterized in that: The method comprises the following steps: Step 1: Design an N-step phase-shifted fringe pattern. The projector projects the N-spoke fringe pattern onto the surface of the foreign object in sequence. The fringe pattern is modulated by the foreign object to produce deformation. The foreign object information is contained in the deformed fringe pattern. The camera captures each deformed fringe pattern. Step 2: Calculate the foreign body wrapping phase ψ(x,y) from the deformed fringe pattern using the phase shift method; Step 3: Select one of the deformed fringe patterns and calculate the initial unwrapped phase from the selected deformed fringe pattern using Fourier transform profilometry Step 4: Based on the preliminary phase The fringe order K is calculated by combining the wrapping phase ψ(x, y), and the foreign body modulation phase is obtained by expanding the wrapping phase ψ(x, y) according to the fringe order K. The grayscale value of the deformed fringe image in step 1 is expressed as: Where (x, y) is the pixel coordinate in the fringe pattern, a(x, y) is the background light intensity, b(x, y) is the modulation intensity, f0 is the fringe frequency, is the foreign body modulation phase, i represents the phase shift of the i-th step, N is the total number of phase shift steps, g i (x, y) is the grayscale value of pixel (x, y) in the i-th phase-shift fringe pattern; The Fourier transform profilometry calculation steps in step 3 are as follows: According to Euler's formula cosθ=[exp(iθ)+exp(-iθ)] / 2, formula (1) is expressed as follows: g(x,y)=a(x,y)+c(x,y)+c * (x,y) (3) Where: Performing Fourier transform on formula (3) yields: G(f)=A(f)+C(f-f0)+C * (f+f0) (4) Among them, the spectrum after Fourier transform includes A(f), C(f-f0), C * (f+f0), A(f) represents the background signal at zero frequency, C(f-f0) represents the spectral component at frequency f0, and C * (f+f0) represents the spectral component at frequency -f0, where C(f-f0), C * (f+f0) contains the phase information of the foreign object. Extract the C(f-f0) term and perform inverse Fourier transform to obtain: The Fourier transform profilometry wrapped phase is calculated based on the real and imaginary parts: Wrapping the Fourier transform profilometry around the phase ψ FTP (x,y) is expanded using the spatial phase unwrapping algorithm to obtain the initial unwrapped phase 2. The fringe projection dynamic measurement method for detecting foreign matter in coal mine belt conveyor transmission according to claim 1 is characterized in that: The calculation formula for ψ(x,y) in step 2 is: According to the above inverse tangent operation, the foreign body encapsulation phase is cut off within the range of (-π, π].

3. The fringe projection dynamic measurement method for detecting foreign matter in coal mine belt conveyor transmission according to claim 1 is characterized in that: In step 4, the calculation method of the fringe level K is as follows:

4. The fringe projection dynamic measurement method for detecting foreign matter in coal mine belt conveyor transmission according to claim 3 is characterized in that: In step 4, the wrapped phase ψ(x,y) is expanded according to the fringe order K to obtain the foreign body modulation phase, which is calculated as follows:

5. The fringe projection dynamic measurement method for detecting foreign matter in coal mine belt conveyor transmission according to claim 1 is characterized in that: The number of phase shift fringes steps described in step 1 is 3. The three-step phase shift fringes are respectively encoded into the RGB channels of the color fringe image. The foreign body phase calculation can be realized using only a single color fringe image.

6. The fringe projection dynamic measurement method for detecting foreign matter in coal mine belt conveyor transmission according to claim 1 is characterized in that: The three-dimensional information of the foreign object is calculated based on the phase information in step 1, and the foreign object on the surface of the belt conveyor is detected based on the three-dimensional information.

Images