A spectrum adaptive extraction method for digital holographic image by eliminating stray spectrum

Abstract

The present invention discloses a method for adaptively extracting digital holographic object image spectrum that eliminates stray spectrum. The holographic interference pattern of the object to be measured is collected and converted into a spectrum diagram, and the center coordinates of the object image spectrum are determined by phase information; iterative threshold segmentation processing is performed to obtain a spectrum binary segmentation diagram; the object image spectrum binary segmentation diagram is extracted and generated in the foreground area of the spectrum binary segmentation diagram according to the center coordinates of the object image spectrum; the object image spectrum binary segmentation diagram is subtracted from the spectrum binary segmentation diagram to obtain a stray spectrum area binary segmentation diagram, and a spectrum diagram after eliminating stray spectrum is obtained on the spectrum diagram according to the stray spectrum area binary segmentation diagram; the object image spectrum is filtered by an adaptively generated Butterworth filter, and a three-dimensional morphology diagram is reconstructed by unwrapping and distortion compensation. The present invention solves the contradiction between the amount of object detail information extracted and the amount of stray spectrum introduced, minimizes the amount of stray spectrum introduced while increasing the amount of object detail information extracted, thereby improving the quality of object three-dimensional morphology reconstruction.

Description

Technical Field

[0001] The present invention relates to a method for measuring the three-dimensional shape of an object in the field of digital holography technology, and in particular to a method for adaptively extracting a digital holographic object image spectrum by eliminating stray spectrum. Background Art

[0002] In off-axis digital holographic measurement, the frequency domain of the holographic interferogram consists of four regions: the object image, conjugate terms, zero-order terms, and spurious spectra. Because the conjugate, zero-order, and spurious spectra represent interference information that hinders high-quality reconstruction of the object's three-dimensional topography, filtering the holographic spectrum to adaptively extract the object image spectrum is a key step in digital holography. Spurious spectra, representing the spectral manifestation of interference information such as coherent noise, ring noise, parasitic fringes, and scattered laser interference signals, are randomly distributed across the entire spectrum. In recent years, researchers both domestically and internationally have proposed numerous automatic filtering methods. However, the filter window generated by threshold segmentation often becomes too small due to excessively large thresholds or excessive iterations, resulting in partial loss of object spectral information and distorted object contour distribution during numerical reconstruction. While an overly large filter window can capture all object details, it also introduces a significant amount of spurious spectra, reducing the quality of the reconstructed image and the accuracy of phase measurements. In summary, the research field of adaptive spectrum extraction of digital holographic images faces the urgent need to resolve the contradiction between the amount of object detail information extracted and the amount of spurious spectrum introduced due to the selection of the filter range size. Summary of the Invention

[0003] To solve the above technical problems, the present invention provides a method for adaptively extracting spectrum from digital holographic images by eliminating stray spectra. This method can reduce the amount of stray spectra introduced while increasing the amount of object detail information extracted, thereby achieving high-quality reconstruction of the object's three-dimensional appearance.

[0004] The present invention is achieved through the following technical solutions:

[0005] Step 1: Collect the holographic interferogram of the object to be measured, then convert the holographic interferogram into a spectrum diagram P0. Determine the center coordinates (x1, y1) of the object image spectrum through the phase information of the spectrum diagram P0.

[0006] Step 2: Perform iterative threshold segmentation on the spectrum map P0 to obtain the spectrum binary segmentation map P1;

[0007] Step 3: According to the center coordinates (x1, y1) of the object image spectrum, search and extract the object image spectrum region binary segmentation mask in the foreground area of the spectrum binary segmentation map P1, that is, the white area with a pixel intensity value of 1, to generate the object image spectrum binary segmentation map P2;

[0008] Step 4: Subtract the object spectrum binary segmentation map P2 from the spectrum binary segmentation map P1 to obtain the spurious spectrum region binary segmentation map P3. Process the spectrum map P0 according to the spurious spectrum region binary segmentation map P3 to obtain the spectrum map P4 after eliminating the spurious spectrum, thereby eliminating the main interference information on the spectrum map in advance.

[0009] Step 5: After eliminating the spurious spectrum, the object image spectrum is automatically filtered using an adaptive Butterworth filter on the spectrum graph P4. Finally, the three-dimensional topography of the object to be measured is reconstructed through phase unwrapping and distortion compensation operations.

[0010] The object to be tested is a micro-nano structure object, such as ultra-precision parts, MEMS chips, and biological cells.

[0011] The present invention adopts a resolution test target as the object to be tested in the embodiment, and collects a holographic interference pattern of the surface of the object to be tested.

[0012] The step 1 is specifically as follows:

[0013] 1.1) Using a CCD (charge-coupled device) photosensitive electronic imaging device to record the holographic interference pattern produced by the interference between the object light wave and the reference light wave:

[0014] I=|O| 2 +|R| 2 +OR * +O * R

[0015] Where I represents the holographic interference signal, O is the object light signal formed by the diffraction of the laser after passing through the object to be measured, and R is the reference light signal of the laser without passing through the object to be measured;

[0016] 1.2) Convert the holographic interference pattern into a spectrum graph P0 through two-dimensional Fourier transform:

[0017] P0=FFT{I}=FFT{|O| 2}+FFT{|R| 2}+FFT{OR *}+FFT{O * R}

[0018] Wherein, FFT{} represents two-dimensional Fourier transform; O and R are the object light signal and reference light signal of the object to be measured, respectively; O* and R* are the conjugate of the object light signal and the conjugate of the reference light signal, respectively; FFT{|O| 2}+FFT{|R| 2} together constitute the zero-order spectrum in the spectrum diagram P0, FFT{O(x,y)R * (x,y)} and FFT{O *(x,y)R(x,y)} are the object-image spectrum and the conjugate term spectrum respectively;

[0019] 1.3) Spectrum P0 is a complex amplitude signal, containing both intensity and phase information. Extract the wrapped phase signal of spectrum P0:

[0020] φ FFT =arctan{Im(P0) / Re(P0)}

[0021] Among them, φ FFT Represents the wrapped phase signal of the spectrogram P0, Re() represents the extraction of the real part information of the spectrogram P0, and Im() represents the extraction of the imaginary part information of the spectrogram P0;

[0022] 1.4) The wrapped phase signal is subjected to a further phase unwrapping operation to obtain the unfolded phase. By searching for the position of the maximum unfolded phase of the holographic interferogram as the center coordinate (x1, y1) of the object image spectrum, the unfolded phase distribution shows an extreme value corresponding to the carrier frequency.

[0023] The step 2 is specifically as follows:

[0024] 2.1) Pre-set the global threshold T0 of the spectrum graph P0 and the critical area to be deleted S, where T0 is greater than 0 and less than 1;

[0025] 2.2) After mean filtering the spectrum map P0, perform an iterative threshold segmentation operation based on the global threshold T0 and the critical area to be deleted S to obtain a spectrum segmentation map;

[0026] 2.3) The computer automatically identifies the number of foreground areas in the current spectrum segmentation diagram and makes a judgment:

[0027] If the number of foreground regions in the spectrum segmentation image obtained by the first threshold segmentation operation is less than 3, then increase the global threshold T0 by 0.01 and return to step 2.2) for processing. Continue iterating until the number of foreground regions is not less than 3. The global threshold T0 of the last iteration is recorded as the segmentation threshold T.

[0028] If the number of foreground regions in the spectrum segmentation image obtained by the first threshold segmentation operation is greater than 3, the global threshold T0 at this time is directly recorded as the segmentation threshold T;

[0029] If the number of foreground regions in the spectrum segmentation image obtained by the first threshold segmentation operation is equal to 3, then reduce the global threshold T0 by 0.01 and return to step 2.2) for processing. Continue iterating until the number of foreground regions is not equal to 3. The global threshold T0 of the second-to-last iteration is recorded as the segmentation threshold T;

[0030] The segmentation threshold T obtained by the final iteration is used to perform threshold segmentation processing on the spectrum image P0 to obtain the spectrum binary segmentation image P1.

[0031] In the above 2.2), the threshold segmentation operation is specifically as follows: performing iterative threshold segmentation on the spectrum map P0, setting the intensity values of the pixels less than the global threshold T0 to 0, and changing the intensity values of the remaining pixels to 1, where 1 represents the foreground area and 0 represents the background area; then deleting all foreground areas whose area is less than the critical area to be deleted S.

[0032] Expressed as:

[0033]

[0034] Among them, P1(x,y) represents each pixel point of the spectrum binary segmentation map, P0(x,y) represents the corresponding pixel point of the spectrum map, and T0 represents the initial segmentation threshold.

[0035] The step three is specifically as follows: sorting all foreground regions of the spectrum binary segmentation map P1 by area, leaving only the three foreground regions with the largest areas, namely the zero-order term region, the conjugate term region, and the object-image spectrum region; obtaining the centroid coordinates (x i ,y i ), calculate the centroid coordinates (x i ,y i ) and the distance L between the center coordinates (x1, y1) of the object image spectrum i , take the distance L i The smallest foreground area is used as the object image spectrum area, and an object image spectrum area binary segmentation mask P2 is established for the object image spectrum area.

[0036] The distance L i It is calculated according to the following formula:

[0037]

[0038] Among them, L i Represents the centroid coordinates of the foreground area (x i ,y i ) to the center coordinate of the object image spectrum area, (x i ,y i ) is the centroid coordinate of the foreground area, and (x1, y1) is the center coordinate of the object image spectrum.

[0039] The step 4 is specifically as follows:

[0040] The spurious spectrum region binary segmentation map P3 is obtained by subtracting the object spectrum binary segmentation map P2 from the spectrum binary segmentation map P1. The intensity values of all pixels in the spurious spectrum region binary segmentation map P3 are either 0 or 1. The spectral information contained in the foreground area of the spurious spectrum region binary segmentation map P3 is noise information that will have a significant adverse effect on the quality of the reconstructed image, including zero-order terms, conjugate terms, and spurious spectrum noise.

[0041] On the spectrum graph P0, the intensity values of the pixels in the foreground area of the stray spectrum region binary segmentation map P3 are set to 0, that is, on the spectrum graph P0, the intensity values of the pixels in the stray spectrum region are all set to 0 according to the stray spectrum region binary segmentation map P3, and the spectrum graph P4 after eliminating the stray spectrum is obtained, thereby eliminating the main interference information on the spectrum graph in advance.

[0042] The spectrum diagram P4 after eliminating the spurious spectrum is obtained according to the following formula:

[0043]

[0044] Among them, P4(x,y) represents each pixel point of the spectrum graph after eliminating the spurious spectrum, P0(x,y) represents the corresponding pixel point of the spectrum graph, and P3(x,y) represents the corresponding pixel point of the binary segmentation map of the spurious spectrum area.

[0045] The step five is specifically as follows:

[0046] On the spectrum graph P4 after eliminating spurious spectra, a circle is automatically generated, with the object image spectrum center coordinates (x1, y1) as the center and a radius r, one-third of the distance from the circle center (x1, y1) to the zero-order term center coordinates (x0, y0). This Butterworth filter is then used to perform spatial filtering of the object image spectrum on the spectrum graph P4 to obtain the object image spectrum signal. The zero-order term center coordinates are the center of the spectrum graph P4. Finally, the object image spectrum information is used to reconstruct a 3D topography of the object through phase unwrapping and distortion compensation.

[0047] The Butterworth filter is constructed as follows:

[0048]

[0049] Where H represents the adaptive Butterworth filter, (x,y) represents each pixel point on the spectrum, (x1,y1) represents the center coordinate of the object image spectrum, r is the cutoff frequency, which is one-third of the distance from (x1,y1) to the center coordinate of the zero-order term (x0,y0); n is the order of the Butterworth filter, which is set to 4 here;

[0050] According to the following formula, the Butterworth filter is used to perform spatial filtering on the spectrum graph P4 to obtain the object image spectrum signal:

[0051] P +1 =H×P4

[0052] Among them, P +1 P represents the spectrum signal of the object image extracted by filtering, and P4 represents the spectrum diagram after eliminating the spurious spectrum.

[0053] Compared with the existing technology, the beneficial effects of the present invention are:

[0054] The present invention uses spectrum image features to identify spurious spectrum regions that significantly negatively impact reconstructed image quality and preemptively eliminates these regions from the spectrum. A wide-range adaptive Butterworth filter is then used to automatically extract and filter the off-axis holographic image spectrum. This resolves the conflict between the amount of object detail information extracted and the amount of spurious spectrum introduced, minimizing the amount of spurious spectrum introduced while increasing the amount of object detail extracted, thereby improving the quality of the object's three-dimensional reconstruction. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] Figure 1 Flowchart of the method for adaptively extracting spectrum of digital holographic image to eliminate spurious spectrum;

[0056] Figure 2 This is the holographic spectrum diagram P0 of the embodiment;

[0057] Figure 3 This is the binary spectrum segmentation map P1 of the embodiment;

[0058] Figure 4 The binary segmentation map P2 of the object image spectrum in the embodiment;

[0059] Figure 5 The spurious spectrum region binary segmentation map P3 of the embodiment;

[0060] Figure 6 The spectrum diagram P4 is after eliminating the spurious spectrum in the embodiment;

[0061] Figure 7 Graph of the adaptive Butterworth filter of the embodiment;

[0062] Figure 8 3D shape reconstruction result of the object in the embodiment; DETAILED DESCRIPTION

[0063] The present invention will be further described below with reference to the accompanying drawings and examples.

[0064] The present invention is implemented as follows Figure 1 As shown in the flowchart, the specific steps are as follows:

[0065] Step 1: Collect the holographic interference pattern of the object to be tested and then convert it into a spectrum graph P0. Determine the center coordinates (x1, y1) of the object image spectrum through the phase information of the spectrum graph P0. Specifically:

[0066] 1.1) Using a CCD (charge-coupled device) photosensitive electronic imaging device to record the holographic interference pattern produced by the interference between the object light wave and the reference light wave:

[0067] I=|O| 2 +|R| 2 +OR * +O * R

[0068] 1.2) Convert the holographic interference pattern into a spectrum graph P0 through two-dimensional Fourier transform:

[0069] P0=FFT{I}=FFT{|O| 2}+FFT{|R| 2}+FFT{OR *}+FFT{O * R}

[0070] 1.3) Spectrum P0 is a complex amplitude signal, containing both intensity and phase information. Extract the wrapped phase signal of spectrum P0:

[0071] φ FFT =arctan{Im(P0) / Re(P0)}

[0072] 1.4) The wrapped phase signal is subjected to a further phase unwrapping operation to obtain the unfolded phase. By searching for the position of the maximum unfolded phase of the holographic interferogram as the center coordinate (x1, y1) of the object image spectrum, the unfolded phase distribution shows an extreme value corresponding to the carrier frequency.

[0073] The specific process adopted in this embodiment is: first, a CCD industrial camera device is used to shoot a holographic interference pattern, and then the holographic interference pattern is Fourier transformed to obtain a spectrum diagram P0, such as Figure 2 As shown. The wrapped phase signal of the spectrum graph P0 is then extracted. After the phase unwrapping operation, the phase graph of the holographic spectrum is obtained. This phase distribution always has an extreme value corresponding to the carrier frequency. The position of the maximum phase value in the spectrum phase graph is recorded as the center coordinate (x1, y1) of the object image spectrum.

[0074] Step 2: Perform iterative threshold segmentation on the spectrum graph P0 to obtain the spectrum binary segmentation graph P1. Specifically:

[0075] 2.1) Preset the global threshold T0 and the critical area to be deleted S of the spectrum graph P0, where the global threshold T0 is greater than 0 and less than 1;

[0076] 2.2) After mean filtering the spectrum map P0, perform an iterative threshold segmentation operation based on the global threshold T0 and the critical area to be deleted S to obtain a spectrum segmentation map;

[0077] The global threshold T0 is greater than 0 and less than 1. Specific implementation The global threshold T0 of the spectrum graph P0 is obtained by the OTSU threshold segmentation method as an initial value, and the critical area to be deleted S is set to 100.

[0078] In 2.2), the threshold segmentation operation is as follows: perform iterative threshold segmentation on the spectrum graph P0, set the intensity values of pixels smaller than the global threshold T0 to 0, and change the intensity values of the remaining pixels to 1, where 1 represents the foreground area and 0 represents the background area; then delete all foreground areas whose area is smaller than the critical area to be deleted S, which can prevent some special small areas from affecting the iterative condition judgment.

[0079] 2.3) The computer automatically identifies the number of foreground areas in the current spectrum segmentation diagram and makes a judgment:

[0080] (a) If the number of foreground regions in the first iteration is less than 3, this indicates that the initial threshold T0 is too small, causing the zero-order term, conjugate term, and object-image spectrum to be combined into a single foreground region, making it impossible to subsequently extract the binary mask of the object-image spectrum region. Therefore, the threshold should be increased by 0.01 each iteration until the number of identified foreground regions is at least 3. The final segmentation threshold T obtained at this point should be recorded.

[0081] (b) If the number of foreground regions in the first iteration is greater than 3, the threshold T is directly recorded as the segmentation threshold. At this point, the spectrum segmentation map contains the zero-order term, the conjugate term, the object-image spectrum, and some spurious spectrum regions. Subsequent operations can be performed to extract the binary mask of the object-image spectrum region.

[0082] (c) If the number of foreground regions in the first iteration is equal to 3, reduce the global threshold T0 by 0.01 and return to step 2.2) for processing. Continue iterating until the number of foreground regions is not equal to 3. The global threshold T0 of the second-to-last iteration is recorded as the segmentation threshold T.

[0083] In this case, although the segmentation of the initial threshold T0 has successfully separated the zero-order term, conjugate term, and object-image spectrum, and can meet the conditions for subsequent extraction of the binary mask of the object-image spectrum region, this is not the maximum spectrum segmentation result. By iteratively searching with a continuously decreasing initial threshold, the first case is to find the critical threshold that connects the zero-order term, conjugate term, and object-image spectrum to form the same foreground region, which is manifested as the number of identified foreground regions being less than 3; the second case is that due to the iterative reduction of the segmentation threshold T, the area of some spurious spectrum regions continues to grow, exceeding the spurious spectrum critical area S, which is manifested as the number of identified foreground regions being greater than 3.

[0084] Therefore, in (c), the iteration termination condition is that the number of identified foreground regions is not equal to 3, and after the iteration is completed, the threshold of the previous iteration needs to be recorded as the final segmentation threshold T, which can ensure that the zero-order term, conjugate term and object image spectrum are not connected together to synthesize the same foreground region.

[0085] The segmentation threshold T obtained by the final iteration is used to perform threshold segmentation on the spectrum image P0 to obtain the spectrum binary segmentation image P1, as shown in Figure 3 shown.

[0086] Thus, the optimal threshold T is searched through the threshold iteration, and the maximum spectrum segmentation result can be obtained.

[0087] Step 3: Sort all foreground regions of the spectrum binary segmentation map P1 by area, leaving only the three foreground regions with the largest areas, namely the zero-order term region, the conjugate term region, and the object-image spectrum region; obtain the centroid coordinates (x i ,y i ), calculate the centroid coordinates (x i ,y i ) and the distance L between the center coordinates (x1, y1) of the object image spectrum i , take the distance L i The smallest foreground area is used as the object image spectrum area, and an object image spectrum area binary segmentation mask P2 is established for the object image spectrum area.

[0088] The specific process adopted in this embodiment is as follows: obtain the area information of all foreground regions in the spectrum binary segmentation map P1, sort them in descending order, take the area ranked third as the critical area S1, delete all foreground regions with an area smaller than the critical area S1, and only keep the three regions with the largest area: the zero-order term, the conjugate term, and the object-image spectrum. Then obtain the centroid coordinates (x i ,y i ), and calculate (x i ,y i) and the object spectrum center coordinates (x1, y1), and the foreground area with the smallest distance is the object spectrum area. Finally, the binary mask of the object spectrum area is extracted separately to generate the object spectrum binary segmentation map P2, as shown in Figure 4 shown.

[0089] Step 4: Subtract the object spectrum binary segmentation map P2 from the spectrum binary segmentation map P1 to obtain the stray spectrum region binary segmentation map P3. The intensity values of all pixels in the stray spectrum region binary segmentation map P3 are either 0 or 1. The spectrum information contained in the foreground area of the stray spectrum region binary segmentation map P3 is noise information that will have a significant adverse effect on the quality of the reconstructed image, including zero-order terms, conjugate terms, and stray spectrum noise.

[0090] On the spectrum graph P0, the intensity values of all pixels in the foreground area of the stray spectrum region binary segmentation map P3 are set to 0, that is, on the spectrum graph P0, the intensity values of all pixels in the stray spectrum region are set to 0 according to the stray spectrum region binary segmentation map P3, and the spectrum graph P4 after eliminating the stray spectrum is obtained, thereby eliminating the main interference information on the spectrum graph in advance.

[0091] The specific process adopted in this embodiment is: the spurious spectrum region binary segmentation map P3 is obtained by subtracting the object spectrum binary segmentation map P2 from the spectrum binary segmentation map P1, as shown in FIG. Figure 5 Then, each pixel of the stray spectrum region binary segmentation map P3 is traversed in turn. When the intensity value of the pixel point is identified as 1, the intensity of the pixel at the same coordinate of the spectrum map P0 is set to 0, and the spectrum map P4 after eliminating the stray spectrum is obtained, as shown in FIG. Figure 6 This operation eliminates the interference information in the spurious spectrum area on the spectrum diagram P0 in advance.

[0092] Step 5: On the spectrum graph P4 after eliminating the spurious spectrum, a circle is constructed with the center coordinates (x1, y1) of the object-image spectrum as the center and a radius r that is one-third of the distance from the center (x1, y1) to the center coordinates of the zero-order term (x0, y0) as the radius. This automatically generates a Butterworth filter. This Butterworth filter is used to spatially filter the spectrum graph P4 to obtain the object-image spectrum signal. The center coordinates of the zero-order term are the center of the spectrum graph P4. Finally, the object's 3D topography is reconstructed from the object-image spectrum signal through phase unwrapping and distortion compensation.

[0093] The specific process adopted in this embodiment is: an adaptive Butterworth filter with a center of (x1, y1) and a radius of r is performed on the spectrum graph P4 after eliminating the spurious spectrum, such as Figure 7As shown in Figure 4, (x1, y1) represents the center coordinates of the object-image spectrum, and the radius r is one-third of the absolute distance from the center coordinates of the object-image spectrum to the center coordinates of the zero-order term, i.e., the center coordinates (x0, y0) in Figure P4. Finally, the 3D topography of the object is reconstructed through phase unwrapping and distortion compensation.

[0094] The phase unwrapping results of this embodiment are as follows Figure 8 As shown, it can be seen that the unfolded phase obtained by the present invention clearly shows the surface morphology of the object and has good uniformity, which proves the effectiveness of the present invention.

[0095] The present invention addresses the difficult-to-solve problem of the contradiction between the amount of object detail information extracted and the amount of stray spectrum introduced due to the selection of the filtering range size in the adaptive filtering of off-axis holographic spectrum graphs. First, the position information of the stray spectrum noise is obtained through image processing technology. Then, by performing adaptive Butterworth filtering on the spectrum graph from which the stray spectrum noise has been eliminated in advance, the amount of stray spectrum introduced is reduced to the greatest extent while the amount of object detail information extracted is increased, thereby achieving high-quality reconstruction of the object's three-dimensional morphology.

Claims

1. A method for adaptively extracting spectrum of a digital holographic image by eliminating spurious spectrum, characterized by: Step 1: Collect the holographic interferogram of the object to be measured, then convert the holographic interferogram into a spectrum diagram P0. Determine the center coordinates (x1, y1) of the object image spectrum through the phase information of the spectrum diagram P0. Step 2: Perform iterative threshold segmentation on the spectrum map P0 to obtain the spectrum binary segmentation map P1; Step 3: According to the center coordinates (x1, y1) of the object image spectrum, search and extract the object image spectrum region binary segmentation mask in the foreground area of the spectrum binary segmentation map P1 to generate the object image spectrum binary segmentation map P2; Step 4: Subtract the object spectrum binary segmentation map P2 from the spectrum binary segmentation map P1 to obtain the spurious spectrum region binary segmentation map P3. Process the spectrum map P0 according to the spurious spectrum region binary segmentation map P3 to obtain the spectrum map P4 after eliminating the spurious spectrum. Step 5: After eliminating the spurious spectrum, the object image spectrum is automatically filtered using an adaptive Butterworth filter on the spectrum graph P4. Finally, the three-dimensional topography of the object to be measured is reconstructed through phase unwrapping and distortion compensation operations.

2. The method for adaptively extracting spectrum of a digital holographic image by eliminating spurious spectrum according to claim 1, characterized in that: The step 1 is specifically as follows: 1.1) Using CCD photosensitive electronic imaging devices to record the holographic interference pattern produced by the interference between the object light wave and the reference light wave: I=|O| 2 +|R| 2 +OR * +O * R Where I represents the holographic interference signal, O is the object light signal formed by the diffraction of the laser after passing through the object to be measured, and R is the reference light signal of the laser without passing through the object to be measured; 1.2) Convert the holographic interference pattern into a spectrum graph P0 through two-dimensional Fourier transform: P0=FFT{I}=FFT{|O| 2 }+FFT{|R| 2 }+FFT{OR * }+FFT{O * R} Wherein, FFT{} represents two-dimensional Fourier transform; O and R are the object light signal and reference light signal of the object to be measured, respectively; O* and R* are the conjugate of the object light signal and the conjugate of the reference light signal, respectively; FFT{|O| 2 }+FFT{|R| 2 } together constitute the zero-order spectrum in the spectrum diagram P0, FFT{O(x,y)R * (x,y)} and FFT{O * (x,y)R(x,y)} are the object-image spectrum and the conjugate term spectrum respectively; 1.3) Spectrum P0 is a complex amplitude signal, containing both intensity and phase information. Extract the wrapped phase signal of spectrum P0: φ FFT =arctan{Im(P0) / Re(P0)} Among them, φ FFT Represents the wrapped phase signal of the spectrogram P0, Re() represents the extraction of the real part information of the spectrogram P0, and Im() represents the extraction of the imaginary part information of the spectrogram P0; 1.4) The wrapped phase signal is subjected to a further phase unwrapping operation to obtain the unwrapped phase, and the position of the maximum value of the unwrapped phase is searched as the center coordinate (x1, y1) of the object image spectrum.

3. The method for adaptively extracting spectrum of a digital holographic image by eliminating spurious spectrum according to claim 1, characterized in that: The step 2 is specifically as follows: 2.1) Pre-set the global threshold T0 and the critical area to be deleted S of the spectrum graph P0; 2.2) After mean filtering the spectrum map P0, perform an iterative threshold segmentation operation based on the global threshold T0 and the critical area to be deleted S to obtain a spectrum segmentation map; 2.3) The computer automatically identifies the number of foreground areas in the current spectrum segmentation diagram and makes a judgment: If the number of foreground regions in the spectrum segmentation image obtained by the first threshold segmentation operation is less than 3, then increase the global threshold T0 by 0.01 and return to step 2.2) for processing. Continue iterating until the number of foreground regions is not less than 3. The global threshold T0 of the last iteration is recorded as the segmentation threshold T. If the number of foreground regions in the spectrum segmentation image obtained by the first threshold segmentation operation is greater than 3, the global threshold T0 at this time is directly recorded as the segmentation threshold T; If the number of foreground regions in the spectrum segmentation image obtained by the first threshold segmentation operation is equal to 3, then reduce the global threshold T0 by 0.01 and return to step 2.2) for processing. Continue iterating until the number of foreground regions is not equal to 3. The global threshold T0 of the second-to-last iteration is recorded as the segmentation threshold T; The spectrum image P0 is segmented using the segmentation threshold T to obtain the spectrum binary segmentation image P1.

4. The method for adaptively extracting spectrum of a digital holographic image by eliminating spurious spectrum according to claim 3, characterized in that: In the above 2.2), the threshold segmentation operation is specifically as follows: performing iterative threshold segmentation on the spectrum map P0, setting the intensity values of the pixels less than the global threshold T0 to 0, and changing the intensity values of the remaining pixels to 1, where 1 represents the foreground area and 0 represents the background area; then deleting all foreground areas whose area is less than the critical area to be deleted S.

5. The method for adaptively extracting spectrum of a digital holographic image by eliminating spurious spectrum according to claim 1, characterized in that: The step three is specifically as follows: sorting all foreground regions of the spectrum binary segmentation map P1 by area, leaving only the three foreground regions with the largest areas, namely the zero-order term region, the conjugate term region, and the object-image spectrum region; Get the centroid coordinates (x i ,y i ), calculate the centroid coordinates (x i ,y i ) and the distance L between the center coordinates (x1, y1) of the object image spectrum i , take the distance L i The smallest foreground area is used as the object image spectrum area, and an object image spectrum area binary segmentation mask P2 is established for the object image spectrum area.

6. The method for adaptively extracting spectrum of a digital holographic image by eliminating spurious spectrum according to claim 5, characterized in that: The distance L i It is calculated according to the following formula: Among them, L i Represents the centroid coordinates of the foreground area (x i ,y i ) to the center coordinate of the object image spectrum area, (x i ,y i ) is the centroid coordinate of the foreground area, and (x1, y1) is the center coordinate of the object image spectrum.

7. The method for adaptively extracting spectrum of a digital holographic image by eliminating spurious spectrum according to claim 1, characterized in that: The fourth step is specifically as follows: subtracting the object-image spectrum binary segmentation map P2 from the spectrum binary segmentation map P1 to obtain the stray spectrum region binary segmentation map P3; on the spectrum map P0, setting the intensity values of the pixels in the foreground region of the stray spectrum region binary segmentation map P3 to 0, to obtain the spectrum map P4 after eliminating the stray spectrum.

8. The method for adaptively extracting spectrum of a digital holographic image by eliminating spurious spectrum according to claim 7, characterized in that: The spectrum diagram P4 after eliminating the spurious spectrum is obtained according to the following formula: Among them, P4(x,y) represents each pixel point of the spectrum graph after eliminating the spurious spectrum, P0(x,y) represents the corresponding pixel point of the spectrum graph, and P3(x,y) represents the corresponding pixel point of the binary segmentation map of the spurious spectrum area.

9. The method for adaptively extracting spectrum of a digital holographic image by eliminating spurious spectrum according to claim 1, characterized in that: The step five is specifically as follows: on the spectrum graph P4 after eliminating the stray spectrum, a circle is established with the center coordinates (x1, y1) of the object image spectrum as the center and a radius r that is one-third of the distance from the center (x1, y1) to the center coordinates (x0, y0) of the zero-order term to generate a Butterworth filter, and the Butterworth filter is used to perform object image spectrum spatial filtering on the spectrum graph P4 to obtain an object image spectrum signal; finally, a three-dimensional morphology image of the object is reconstructed from the object image spectrum information through phase unwrapping and distortion compensation operations.

10. The method for adaptively extracting spectrum of a digital holographic image by eliminating spurious spectrum according to claim 9, characterized in that: The Butterworth filter is constructed as follows: Where H represents the adaptive Butterworth filter, (x,y) represents each pixel point on the spectrum, (x1,y1) represents the center coordinate of the object spectrum, r is the cutoff frequency, which is one-third of the distance from (x1,y1) to the center coordinate of the zero-order term (x0,y0); n is the order of the Butterworth filter; According to the following formula, the Butterworth filter is used to perform spatial filtering on the spectrum graph P4 to obtain the object image spectrum signal: P +1 =H×P4 Among them, P +1 P represents the spectrum signal of the object image extracted by filtering, and P4 represents the spectrum diagram after eliminating the spurious spectrum.

Images