CN106846383A - High dynamic range images imaging method based on 3D digital micro-analysis imaging systems - Google Patents

Abstract

The invention relates to a high dynamic range image imaging method based on a 3D digital microscopic imaging system. By generating a high dynamic range image of an object to be observed and obtaining an original high dynamic multi-focus sequence image of a sample to be observed, a phase matching method and a Fourier Lie transform performs image registration and superpixel level movement, and then segments the target object through the foreground and background segmentation method; performs quadtree decomposition processing on the segmented image, marks clear image blocks in the image sequence, and records The height information corresponding to each image; finally, the marked clear image blocks are fused into the three-dimensional shape of the logistics to be observed, and the median filter is used to filter the generated three-dimensional shape to eliminate the three-dimensional shape due to the sampling frequency The sawtooth effect caused by the deficiency makes the three-dimensional formation of the object to be observed smoother.

Description

High Dynamic Range Image Imaging Method Based on 3D Digital Microscopic Imaging System

technical field

The invention relates to the technical field of high-definition and high-precision microscopic imaging detection, in particular to a high dynamic range image imaging method based on a 3D digital microscopic imaging system.

Background technique

Multi-focal length 3D technology (Shape from Focus, referred to as SFF) is currently a commonly used 3D technology in the field of digital microscopic image processing. Since the multi-focal length 3D technology can obtain the three-dimensional shape of the observed sample only by using a traditional monocular microscope, it has attracted widespread attention from experts and scholars. Different from stereo vision technology that uses binocular lens to obtain depth information, multi-focal length 3D technology only detects clear areas in the image by moving and observing the distance from the object to the lens, so as to restore and reconstruct the depth information of the object.

However, the main defect of the multi-focal length 3D technology is that when the observed sample has high reflection, the dynamic range of the collected image is insufficient, resulting in insufficient image details in some areas, or even no image details in some areas, so It greatly affects the accuracy of the three-dimensional shape of the reconstructed object. However, many current scientific researches still mainly focus on the influence of the focusing factor on the accuracy of 3D shape reconstruction of objects, but ignore the influence of the quality of the original image in terms of dynamic range on the results of 3D shape reconstruction of objects.

In order to overcome the influence of insufficient dynamic range in the obtained images, high dynamic range imaging technology is proposed. A high dynamic range image (High-Dynamic Range, HDR for short) can be obtained by using a high dynamic range imaging technology. Through calibration, the images of different exposure times of the same scene are fused to obtain a 32-bit high dynamic range light spectrum of the scene. These 32-bit light spectrum images can accurately and truly reflect the dynamic range in the scene, and then map these 32-bit light spectrum images to 8-bit ordinary images through local tone mapping, so that it is convenient for traditional display devices to display and save these 8-bit images. normal image. However, current microscopic 3D reconstruction methods on the market still have limitations in implementing this technique due to the high computational complexity of high dynamic range imaging techniques.

Contents of the invention

The technical problem to be solved by the present invention is to provide a high dynamic range image imaging method based on a 3D digital microscopic imaging system in view of the above prior art. The high dynamic range image imaging method can solve the defect that the existing image imaging methods cannot capture high dynamic scenes clearly, and can simultaneously and accurately generate the three-dimensional shape of the object to be observed, thereby providing observers with a full range of 3D stereoscopic visual enjoyment.

The technical solution adopted by the present invention to solve the above-mentioned technical problems is: a high dynamic range image imaging method based on a 3D digital microscopic imaging system, which is characterized in that it includes the following steps:

Step 1. For the object to be observed on the microscope stage, adjust the height of the stage and use the camera to obtain high dynamic multi-focus images of each layer from the bottom of the object to be observed to the top of the object to be observed to obtain a three-dimensional Original high dynamic multi-focus sequence images required for stereo imaging;

Step 2, using the phase matching method to register the obtained original high dynamic multi-focus sequence images, so that the spatial positions, zoom scales and image sizes of the consecutive image pairs in the original high dynamic multi-focus sequence images are correspondingly consistent, so that Get the registered high dynamic multi-focus sequence images;

Step 3, for the registered high dynamic multi-focus sequence images, use the foreground and background segmentation method of background accumulation to extract the observation sample area that needs to generate three-dimensional stereo;

Step 4: Segment the observed sample area using a quadtree segmentation method, detect clear parts in each image of the high dynamic multi-focus sequence images, and record the height information corresponding to each image;

In step 5, the clear parts in the detected images are fused to generate a three-dimensional shape of the object to be observed.

Further, the process of using the camera to obtain the high dynamic range image of each layer in the step 1 includes:

(a) calibrate the corresponding curve of the camera; (b) acquire images for different exposure values in the same scene; (c) utilize the corresponding curve of the calibrated camera to generate a 32-bit light spectrogram of the scene; (d ) using local tone mapping to map the 32-bit light spectrogram to an 8-bit normal image, and save the normal image in a format that can be displayed and stored by a computer.

Further, in step 1, the acquisition process of the original high dynamic multi-focus sequence image includes: firstly, by moving the height of the stage, changing the distance between the object to be observed and the objective lens of the microscope, so as to achieve different A sequence of focal plane images; secondly, the requirements for recording the height information of each focal plane image; again, performing focus detection for each focal plane image, and recording the pixel with the maximum focus definition in each focal plane image points for subsequent 3D shape reconstruction.

Specifically, in step 2, the process of registering the obtained original high dynamic multi-focus sequence images by the phase matching method includes:

Firstly, in the original high dynamic multi-sequence image, for every two image pairs connected back and forth, each image in the image pair is converted into a grayscale image, thereby obtaining a grayscale image pair;

Secondly, the phase information of each frequency band is extracted from the converted gray-scale image pair by using a complex band-pass filter;

Again, using the extracted phase information, realize the movement of the grayscale image pair at the superpixel level through Fourier transform, so as to ensure the consistency of the positions of the two consecutive images;

Finally, for each set of image pairs in the original high-dynamic multi-focus image sequence, the process is repeated until the scale and displacement of all images in the high-dynamic multi-focus image sequence are consistent.

Specifically, the process of using the quadtree segmentation method in the step 4 to segment the observation sample area includes:

First, the original high dynamic multi-focus sequence image is input into the quadtree as a layer of the root of the quadtree;

Secondly, set the image decomposition conditions, and process according to whether the images of each layer in the quadtree meet the decomposition conditions:

If the image decomposition condition is satisfied for a layer of image, the image of this layer is quadrupled and input to the next layer of the quadtree; and so on until the smallest image block obtained by decomposing the image sequence does not satisfy Described image decomposition condition, then end quadtree decomposition process; Wherein, the image decomposition condition of setting is:

For the decomposed image blocks of each layer in the image sequence in the quadtree, the maximum difference value of the focus factor MDFM and the value of the gradient difference SMDG are respectively calculated; the calculation formulas of the maximum difference value of the focus factor MDFM and the value of the gradient difference SMDG are as follows:

MDFM = FM _max - FM _min ;

Among them, FM _max represents the maximum value of focal length measurement, FM _min represents the minimum value of focal length measurement; grad _max (x, y) represents the maximum gradient value, and grad _min (x, y) represents the minimum gradient value;

For a layer of image blocks in the quadtree, if MDFM≥0.98×SMDG is satisfied, and there is a fully focused image block in the image sequence of this layer, the image block of this layer will not continue to decompose downward; otherwise, the image of this layer Blocks will continue to be decomposed until all images in the quadtree have been decomposed into sub-image blocks that cannot be decomposed.

Specifically, the acquisition process of the maximum value FM _max of the focal length measurement and the minimum value FM _min of the focal length measurement is:

First, calculate the gradient matrix of each pixel in the layer image of the root of the quadtree, the calculation formula is:

GM _i = gradient(I _i ), i=1,2,...,n;

Wherein, I _i is the i-th original high dynamic multi-focus image, GM _i is the gradient matrix corresponding to I _i ; n is the total number of images in the original high dynamic multi-focus sequence images;

Secondly, find the largest gradient matrix and the smallest gradient matrix among all the gradient matrices of each point of this layer of image, the formula is as follows:

GM _max = max(GM _i (x,y)), i=1,2,...,n;

GM _min =min(GM _i (x,y)),i=1,2,...,n;

Again, calculate the sum of the gradient matrix of all points in this layer of image, the calculation formula is as follows:

FM _i = Σ _x Σ _y grad _i (x, y), i = 1, 2,..., n;

Finally, find the maximum and minimum values of the sum of the above gradient matrices respectively, and the calculation formula is as follows:

FM _max = max{FM _i }, i=1,2,...,n;

FM _min = min{FM _i }, i=1, 2, . . . , n.

Specifically, the process of fusing the clear parts of each image in step 5 includes: taking all the obtained clear parts as clear sub-image blocks, recording their height information respectively, and fusing all the clear sub-image blocks into a complete three-dimensional image of the observed sample.

Improvement, the step 5 also includes: filtering the generated three-dimensional shape by using a median filter method to eliminate the jagged effect caused by insufficient sampling frequency of the three-dimensional shape, so as to make the generated three-dimensional shape smoother.

Compared with the prior art, the present invention has the advantages of:

First of all, the high dynamic range image imaging method provided by the present invention adopts high dynamic range imaging, three-dimensional imaging and multi-depth image fusion technology, simultaneously acquires image sequences of different exposure times of the same scene, and generates a 32-bit light spectrum of the scene Then map the 32-bit light spectrum map to an 8-bit common image by using local tone, and save it into a high dynamic range video format that can be displayed and stored by the computer, and use the tone mapping technology to display and transmit the high dynamic range display in real time Micro video, which is convenient for the observer to watch the object to be observed dynamically in real time;

Secondly, due to the high computational complexity involved in the application of high dynamic range video technology, the present invention uses methods such as phase matching and quadtree segmentation to process video signals in real time and generate real-time microscopic video displays, thereby reducing computational complexity. Spend;

Again, the high dynamic range image imaging method in the present invention can observe the object to be observed in real time and in high definition, which overcomes the defect that the current image imaging technology cannot see the reflective and non-reflective areas clearly at the same time for high-contrast samples;

Finally, the high dynamic range image imaging method of the present invention can obtain a fully focused image synthesized from images with different focal points; in the process of processing images with different focal points, the depth of the midpoint of the image can be automatically obtained to restore the image on the surface. The three-dimensional coordinates of the points provide a strong auxiliary guarantee for the quality inspection of emerging materials.

Description of drawings

1 is a schematic flow diagram of a high dynamic range image imaging method based on a 3D digital microscopic imaging system in Embodiment 1 of the present invention;

2 is a schematic diagram of a 3D digital microscopic imaging system in Embodiment 1 of the present invention;

Fig. 3 is the original high dynamic multi-focus sequence image corresponding to the metal screw in the first embodiment;

Fig. 4 is a comparison diagram between the high dynamic range image of the metal screw obtained in Example 1 and the ordinary automatic exposure image; wherein, the left column is the corresponding high dynamic range image, and the right column is the corresponding ordinary automatic exposure image;

5 is a schematic diagram of extracting a foreground image in Embodiment 1;

Figure 6a is a 3D stereoscopic image without image texture mapping generated using a high dynamic range image in Embodiment 1;

Figure 6b is a 3D stereoscopic image with image texture mapping generated using a high dynamic range image in Embodiment 1;

Figure 6c is a 3D stereoscopic image without image texture mapping generated using the original auto-exposure image in Embodiment 1;

Figure 6d is a 3D stereoscopic image with image texture mapping generated using the original auto-exposure image in Embodiment 1;

FIG. 6e is a truth map of a 3D stereo image without image texture mapping in Embodiment 1;

Fig. 6f is a truth map of a 3D stereo image with image texture mapping in Embodiment 1;

Figure 7a is a 3D stereoscopic image without image texture mapping generated using a high dynamic range image in Embodiment 2;

Figure 7b is a 3D stereoscopic image with image texture mapping generated using a high dynamic range image in Embodiment 2;

Figure 7c is a 3D stereoscopic image without image texture mapping generated using the original auto-exposure image in Embodiment 2;

Figure 7d is a 3D stereoscopic image with image texture mapping generated using the original auto-exposure image in Embodiment 2;

Fig. 7e is the truth map of the 3D stereoscopic image without image texture mapping in the second embodiment;

Fig. 7f is the truth map of the 3D stereoscopic image with image texture mapping in the second embodiment;

Fig. 8 is a comparison diagram of the square root error between the method of generating a 3D stereoscopic shape using a high dynamic range image and the method of not using a high dynamic range image in Embodiment 2;

Figure 9a is a 3D stereoscopic image without image texture mapping generated using a high dynamic range image in Embodiment 3;

Figure 9b is a 3D stereoscopic image with image texture mapping generated using a high dynamic range image in Embodiment 3;

Fig. 9c is a 3D stereoscopic image without image texture mapping generated using the original auto-exposure image in Embodiment 3;

Fig. 9d is a 3D stereoscopic image with image texture mapping generated using the original auto-exposure image in Embodiment 3;

FIG. 9e is a truth map of a 3D stereoscopic image without image texture mapping in Embodiment 3;

Fig. 9f is the truth map of the 3D stereoscopic image with image texture mapping in the third embodiment;

FIG. 10 is a comparison diagram of the square root error corresponding to the 3D stereoscopic image generated by the method of generating a 3D stereoscopic shape from a high dynamic range image and the method of generating a 3D stereoscopic shape from an original automatic exposure image.

detailed description

The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments.

Embodiment one

As shown in Figure 2, the 3D digital microscopic imaging system adopted in the first embodiment includes a traditional optical microscope, an automatic stage capable of moving in any direction of the X-axis, Y-axis and Z-axis, a CMOS camera and computer. Wherein, the object to be observed in the first embodiment is a metal screw, and the metal screw is placed on the automatic stage. Referring to Fig. 1, the high dynamic range image imaging method based on the 3D digital microscopic imaging system in the first embodiment includes the following steps:

Step 1. For the object to be observed on the microscope stage, that is, the metal screw, by adjusting the height of the stage, the CMOS camera is focused on each layer of the object to be observed, that is, each layer of the metal screw, And use the camera to obtain high dynamic multi-focus images of each layer from the bottom of the metal screw to the top of the metal screw to obtain the original high dynamic multi-focus sequence images required for three-dimensional imaging; the original high dynamic multi-focus sequence images for metal screws See Figure 3; wherein, the process of obtaining a high dynamic multi-focus image of the object to be observed includes two processes of high dynamic range image acquisition and multi-focus image acquisition; specifically, obtaining a high dynamic range image of each layer of the object to be observed The process includes:

(a) Calibrate the corresponding curve of the camera; (b) Acquire images with different exposure values in the same scene; (c) Use the corresponding curve of the calibrated camera to generate a 32-bit light spectrum of the scene; (d) Use the local tone Mapping, mapping the 32-bit light spectrum to an 8-bit common image, and saving the normal image into a format that can be displayed and stored by a computer. The high dynamic range image of each layer corresponding to the metal screw is shown in the left column in Figure 4;

Step 2: Use the phase matching method to register the obtained original high dynamic multi-focus sequence images, so that the spatial positions, zoom scales and image sizes of the consecutive image pairs in the original high dynamic multi-focus sequence images are correspondingly consistent, so as to obtain the registration The aligned high dynamic multi-focus sequence images; wherein, the process of registering the obtained original high dynamic multi-focus sequence images by the phase matching method includes:

Firstly, in the original high dynamic multi-sequence image, for every two image pairs that are connected back and forth, each image in the image pair is converted into a grayscale image, thereby obtaining a grayscale image pair;

Secondly, the phase information of each frequency band is extracted from the converted gray-scale image pair by using a complex band-pass filter;

Thirdly, using the extracted phase information, the gray-scale image pair is moved at the superpixel level through Fourier transform to ensure the consistency of the positions of the two consecutive images;

Finally, for each set of image pairs in the original high-dynamic multi-focus image sequence, the process is repeated until the scale and displacement of all images in the high-dynamic multi-focus image sequence are consistent.

Step 3, for the registered high dynamic multi-focus sequence images, use the foreground and background segmentation method of background accumulation to extract the observation sample area that needs to be generated in three dimensions; see Figure 5, that is, use the inter-frame difference to divide the corresponding metal screw The background is extracted and then thresholded to obtain the foreground image;

Step 4: Segment the observation sample area using a quadtree segmentation method, and detect the clear part in each image of the high dynamic multi-focus sequence image, and record the height information corresponding to each image; wherein,

The description of the quadtree segmentation method in the first embodiment is as follows:

First, the original high dynamic multi-focus sequence image is input into the quadtree as a layer of the root of the quadtree;

Secondly, set the image decomposition conditions, and process according to whether the images of each layer in the quadtree meet the decomposition conditions:

If the image decomposition condition is satisfied for a layer of image, the image of this layer is quadrupled and input to the next layer of the quadtree; and so on, until the smallest image block obtained by decomposing the image sequence does not satisfy the image decomposition condition, then end the quadtree decomposition process; wherein, the description of the image decomposition conditions is as follows:

For the decomposed image blocks of each layer in the image sequence in the quadtree, the maximum difference value of the focus factor MDFM and the value of the gradient difference SMDG are respectively calculated; the calculation formulas of the maximum difference value of the focus factor MDFM and the value of the gradient difference SMDG are as follows:

MDFM = FM _max - FM _min ;

Among them, FM _max represents the maximum value of focal length measurement, FM _min represents the minimum value of focal length measurement; grad _max (x, y) represents the maximum gradient value, grad _min (x, y) represents the minimum gradient value; the maximum value for focal length measurement The calculation of FM _max and the minimum value FM _min of focal length measurement is:

First, calculate the gradient matrix of each pixel in the layer image of the root of the quadtree, the calculation formula is:

GM _i = gradient(I _i ), i=1,2,...,n;

Wherein, I _i is the i-th original high dynamic multi-focus image, GM _i is the gradient matrix corresponding to I _i ; n is the total number of images in the original high dynamic multi-focus sequence images;

Secondly, find the largest gradient matrix and the smallest gradient matrix among all the gradient matrices of each point of this layer of image, the formula is as follows:

GM _max = max(GM _i (x,y)), i=1,2,...,n;

GM _min =min(GM _i (x,y)),i=1,2,...,n;

Again, calculate the sum of the gradient matrix of all points in this layer of image, the calculation formula is as follows:

FM _i = Σ _x ∑ _y grad _i (x, y), i = 1, 2,..., n;

Finally, find the maximum and minimum values of the sum of the above gradient matrices respectively, and the calculation formula is as follows:

FM _max =max{FM _i }, i=1,2,...,n; FM _min =min{FM _i },i=1,2,...,n.

The process of detecting the clear part in each image of the high dynamic multi-focus sequence image is described as follows:

For each image block sequence in the quadtree, find an image block with the largest gradient matrix in the image block sequence, and record the position and height information of the image block with the largest gradient matrix in the image sequence;

i=1,2,...,n; where, fm _i (x,y) represents the gradient matrix of the i-th image in the image sequence.

Step 5, fuse the clear parts of the detected images to generate the three-dimensional shape of the object to be observed, that is, the three-dimensional image of the metal screw; where the three-dimensional image corresponding to the metal screw is marked as Z :

Z(x, y)=z _i (x, y), z _i (x, y) represents the clear image block of the i-th image in the image sequence.

In order to compare the traditional method of using the original auto-exposure image to generate a 3D stereo shape with the method of the present invention using a high dynamic range image to generate a 3D stereo shape, this embodiment 1 provides the metal screws generated by using the above two 3D stereo shape methods respectively. For the comparison diagram corresponding to the stereoscopic image, refer to FIG. 6 for details. In the present invention, the use of the original automatic exposure image technology is recorded as Normal SFF, and the use of high dynamic range image technology is recorded as HDR-SFF. in:

In order to compare the accuracy of the above two 3D three-dimensional shape generation methods, Embodiment 1 of the present invention introduces the square root error to measure the gap between the two 3D three-dimensional shape generation methods and the true value under the same conditions:

Wherein, G _T (i, j) represents a true value, and Z(i, j) represents a Normal SFF or HDR-SFF value.

Table 1 shows the square root errors corresponding to the two 3D stereo shape generation methods using 22 different focusing factors. By comparing the results in Table 1, it can be seen that for the same focus factor, the square root error value of the focus factor corresponding to the 3D stereo shape generated by the high dynamic range image is smaller than the focus factor corresponding to the 3D stereo shape generated by the high dynamic range image. Factor square root error value. The results in Table 1 show that the 3D stereoscopic shape generated using the high dynamic range image in the present invention is more accurate than the 3D stereoscopic shape generated without using the high dynamic range image.

Table 1

Embodiment two

In the second embodiment, a plastic bank card is used as the object to be observed, and the bank card has a lowercase English letter "d". Wherein, the steps for generating the three-dimensional image of the bank card are the same as the steps for generating the three-dimensional image of the metal screw in Embodiment 1, and will not be repeated here.

In the second embodiment, in order to verify the accuracy and robustness of the high dynamic range image imaging method of the present invention, the second embodiment provides the corresponding high dynamic range image generated by the bank card, see Fig. 7a-Fig. 7f shown. Fig. 8 is a comparison diagram of the square root error between the method of generating a 3D stereoscopic shape using a high dynamic range image and the method without using a high dynamic range image for the bank card in the second embodiment.

It can be seen from Figure 8 that for the same focus factor, the square root error value of the focus factor corresponding to the 3D stereoscopic shape generated by using the high dynamic range image is smaller than the square root error value of the focus factor corresponding to the 3D stereoscopic shape generated by the high dynamic range image . It can be seen that the 3D stereoscopic shape generated using the high dynamic range image in the present invention is more accurate than the 3D stereoscopic shape generated without using the high dynamic range image.

Embodiment three

In the third embodiment, a metal chip is used as the object to be observed. Wherein, the steps for generating the three-dimensional image of the metal chip are the same as the steps for generating the three-dimensional image of the metal screw in Embodiment 1, and will not be repeated here.

FIG. 10 is a comparison diagram of the square root error corresponding to the 3D stereoscopic image generated by the method of generating a 3D stereoscopic shape from a high dynamic range image and the method of generating a 3D stereoscopic shape from an original automatic exposure image.

It can be seen from Figure 10 that for the same focus factor, the square root error value of the focus factor corresponding to the 3D stereoscopic shape generated by the high dynamic range image is smaller than the square root error value of the focus factor corresponding to the 3D stereoscopic shape generated by the high dynamic range image . It can be seen that the 3D stereoscopic shape generated using the high dynamic range image in the present invention is more accurate than the 3D stereoscopic shape generated without using the high dynamic range image.

Claims (8)

1. The high dynamic range image imaging method based on 3D digital microscopic imaging system, is characterized in that, comprises the steps: Step 1. For the object to be observed on the microscope stage, adjust the height of the stage and use the camera to obtain high dynamic multi-focus images of each layer from the bottom of the object to be observed to the top of the object to be observed to obtain a three-dimensional Original high dynamic multi-focus sequence images required for stereo imaging; Step 2, using the phase matching method to register the obtained original high dynamic multi-focus sequence images, so that the spatial positions, zoom scales and image sizes of the consecutive image pairs in the original high dynamic multi-focus sequence images are correspondingly consistent, so that Get the registered high dynamic multi-focus sequence images; Step 3, for the registered high dynamic multi-focus sequence images, use the foreground and background segmentation method of background accumulation to extract the observation sample area that needs to generate three-dimensional stereo; Step 4: Segment the observed sample area using a quadtree segmentation method, detect clear parts in each image of the high dynamic multi-focus sequence images, and record the height information corresponding to each image; In step 5, the clear parts in the detected images are fused to generate a three-dimensional shape of the object to be observed. 2. The high dynamic range image imaging method according to claim 1, wherein the process of utilizing a camera to obtain the high dynamic range image of each level in the step 1 comprises: (a) calibrate the corresponding curve of the camera; (b) acquire images for different exposure values in the same scene; (c) utilize the corresponding curve of the calibrated camera to generate a 32-bit light spectrogram of the scene; (d ) using local tone mapping to map the 32-bit light spectrogram to an 8-bit normal image, and save the normal image in a format that can be displayed and stored by a computer. 3. The high dynamic range image imaging method according to claim 1, wherein in step 1, the obtaining process of the original high dynamic multi-focus sequence image comprises: In step 1, firstly, by moving the height of the stage, changing the distance between the object to be observed and the objective lens of the microscope to realize the image sequence of different focal planes of the monocular microscope; secondly, recording the height information of each focal plane image Requirements; Again, focus detection is performed on each focal plane image, and the pixel point with the maximum focus definition in each focal plane image is recorded for subsequent three-dimensional shape reconstruction. 4. The high dynamic range image imaging method according to claim 1, wherein, in step 2, the process of registering the obtained original high dynamic multi-focus sequence images by the phase matching method comprises: Firstly, in the original high dynamic multi-sequence image, for every two image pairs connected back and forth, each image in the image pair is converted into a grayscale image, thereby obtaining a grayscale image pair; Secondly, the phase information of each frequency band is extracted from the converted gray-scale image pair by using a complex band-pass filter; Again, using the extracted phase information, realize the movement of the grayscale image pair at the superpixel level through Fourier transform, so as to ensure the consistency of the positions of the two consecutive images; Finally, for each set of image pairs in the original high-dynamic multi-focus image sequence, the process is repeated until the scale and displacement of all images in the high-dynamic multi-focus image sequence are consistent. 5. The high dynamic range image imaging method according to claim 1, wherein the process of adopting the quadtree segmentation method to segment the observation sample area in the step 4 comprises: First, the original high dynamic multi-focus sequence image is input into the quadtree as a layer of the root of the quadtree; Secondly, set the image decomposition conditions, and process according to whether the images of each layer in the quadtree meet the decomposition conditions: If the image decomposition condition is satisfied for a layer of image, the image of this layer is quadrupled and input to the next layer of the quadtree; and so on until the smallest image block obtained by decomposing the image sequence does not satisfy Described image decomposition condition, then end quadtree decomposition process; Wherein, the image decomposition condition of setting is: For the decomposed image blocks of each layer in the image sequence in the quadtree, the maximum difference value of the focus factor MDFM and the value of the gradient difference SMDG are respectively calculated; the calculation formulas of the maximum difference value of the focus factor MDFM and the value of the gradient difference SMDG are as follows: MDFM = FM _max - FM _min ; $\begin{array}{c}\mathrm{<span\; class="notranslate">\; S</span>}\mathrm{<span\; class="notranslate">\; m</span>}\mathrm{<span\; class="notranslate">\; D.</span>}\mathrm{<span\; class="notranslate">\; G</span>}<span\; class="notranslate">\; =</span>\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}\mathrm{<span\; class="notranslate">\; \&Sigma;</span>}<span\; class="notranslate">\; \&lsqb;</span>{\mathrm{<span\; class="notranslate">\; grad</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; the\; y</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; grad</span>}}_{\mathrm{<span\; class="notranslate">\; min</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; the\; y</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; \&rsqb;</span>\\ <span\; class="notranslate">\; =</span>{\mathrm{<span\; class="notranslate">\; \&Sigma;\&Sigma;grad</span>}}_{\mathrm{<span\; class="notranslate">\; max</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; the\; y</span>}<span\; class="notranslate">\; )</span><span\; class="notranslate">\; -</span>{\mathrm{<span\; class="notranslate">\; \&Sigma;\&Sigma;grad</span>}}_{\mathrm{<span\; class="notranslate">\; min</span>}}<span\; class="notranslate">\; (</span>\mathrm{<span\; class="notranslate">\; x</span>}<span\; class="notranslate">\; ,</span>\mathrm{<span\; class="notranslate">\; the\; y</span>}<span\; class="notranslate">\; )</span>\end{array}<span\; class="notranslate">\; ;</span>$ Among them, FM _max represents the maximum value of focal length measurement, FM _min represents the minimum value of focal length measurement; grad _max (x, y) represents the maximum gradient value, and grad _min (x, y) represents the minimum gradient value; For a layer of image blocks in the quadtree, if MDFM≥0.98×SMDG is satisfied, and there is a fully focused image block in the image sequence of this layer, the image block of this layer will not continue to decompose downward; otherwise, the image of this layer Blocks will continue to be decomposed until all images in the quadtree have been decomposed into sub-image blocks that cannot be decomposed. 6. The high dynamic range image imaging method according to claim 5, wherein the acquisition process of the maximum value FM _max of the focal length measurement and the minimum value FM _min of the focal length measurement is: First, calculate the gradient matrix of each pixel in the layer image of the root of the quadtree, the calculation formula is: GM _i = gradient(I _i ), i=1,2,...,n; Wherein, I _i is the i-th original high dynamic multi-focus image, GM _i is the gradient matrix corresponding to I _i ; n is the total number of images in the original high dynamic multi-focus sequence images; Secondly, find the largest gradient matrix and the smallest gradient matrix among all the gradient matrices of each point of this layer of image, the formula is as follows: GM _max = max(GM _i (x,y)), i=1,2,...,n; GM _min =min(GM _i (x,y)),i=1,2,...,n; Again, calculate the sum of the gradient matrix of all points in this layer of image, the calculation formula is as follows: FM _i = Σ _x Σ _y grad _i (x, y), i = 1, 2,..., n; Finally, find the maximum and minimum values of the sum of the above gradient matrices respectively, and the calculation formula is as follows: FM _max = max{FM _i }, i=1,2,...,n; FM _min = min{FM _i }, i=1, 2, . . . , n. 7. The high dynamic range image imaging method according to claim 5, wherein the process of merging the clear parts of each image in the step 5 comprises: using all the clear parts obtained as clear sub-image blocks , respectively record their height information, and fuse all the clear sub-image blocks into a complete three-dimensional image of the observed sample. 8. The high dynamic range image imaging method according to claim 1, characterized in that, said step 5 also includes: using a median filter method to filter the generated three-dimensional shape, to eliminate the three-dimensional shape due to sampling frequency The aliasing effect caused by insufficient, so that the generated three-dimensional stereoscopic formation is smoother.