An automatic biological sample tissue morphology measurement system

Abstract

The present invention relates to the field of topography measurement, and particularly to an automatic biological sample tissue topography measurement system. The system includes: a controller, a tissue tray, a reflective photoelectric sensor, and an image acquisition device. The controller performs the following steps: performing image segmentation processing on the image information of the sample to be measured; obtaining the amplitude threshold corresponding to the current scanning position according to the transparency coefficient corresponding to the independent image segmentation region to which the mapping position belongs, and determining the reception time of the signal amplitude of the first signal amplitude greater than or equal to the amplitude threshold as the signal reception time. According to the correspondence between the surface of different transparency regions and the reflection signal amplitude, the present invention obtains the amplitude threshold corresponding to the true reflection signal that better fits the surface of the current position. Also, based on the fact that the generation time of the surface signal is earlier than the generation time of the internal structure signal, thereby further improving the accuracy of the signal reception time and the accuracy of the thickness information of the biological tissue sample finally calculated.

Description

Technical Field

[0001] The present invention relates to the field of topography measurement, and particularly to an automatic biological sample tissue topography measurement system. Background Art

[0002] The topography, size, and weight of biological tissue samples are important research contents in medical, biological, and other research work. The size and weight data of biological samples are usually measured by marking with a ruler and an electronic scale. For smaller biological tissues, such as mouse uterine implantation sites and small-volume tumors, there are large subjectivity and measurement errors, which directly affect biological results, bringing great inconvenience to basic experiments and clinical sample collection. Moreover, manual measurement is time-consuming and laborious, with a heavy workload.

[0003] In the prior art, a reflective photoelectric sensor is used to emit a measurement signal to a biological tissue sample, and the thickness of the biological tissue sample is calculated by the time difference between the signal emission time and the signal reception time. Usually, the time corresponding to the signal with the largest amplitude in the received return signal waveform is used as the signal reception time. However, in some biological tissue samples, there may be a certain semi-transparent area or even the whole is in a semi-transparent state. Thus, in the semi-transparent area, not only its surface will reflect the signal, but also its interior and even the bottom tray will reflect the signal. And due to the different reflectivities and the superposition of light, the amplitude of the reflected signal of the internal structure of the semi-transparent area in the received return signal waveform may be greater than the amplitude of the reflected signal of the surface structure of the semi-transparent area, resulting in an error in the determined signal reception time and ultimately reducing the accuracy of the thickness information of the biological tissue sample. Summary of the Invention

[0004] In view of one of the above technical problems, the technical solution adopted by the present invention is as follows:

[0005] According to one aspect of the present invention, an automatic biological sample tissue topography measurement system is provided. The system includes: a controller, a tissue tray, a reflective photoelectric sensor, and an image acquisition device. The photoelectric sensor is arranged above the tissue tray. The tissue tray is a self-luminous tissue tray. Both the photoelectric sensor and the image acquisition device are communicatively connected to the controller.

[0006] The controller is configured to perform the following steps:

[0007] According to the transparency information of the image, perform image segmentation processing on the image information of the sample to be measured collected by the image acquisition device, and generate segmentation information corresponding to the image information. The segmentation information includes multiple independent image unit information. The independent image unit information includes the contour information and the transparency coefficient of the independent image segmentation region. The transparency coefficient is positively correlated with the transparency.

[0008] If the mapped position of the current scanning position of the photoelectric sensor in the image information is located in an independent image segmentation region, then according to the transparency coefficient corresponding to the independent image segmentation region to which the mapped position belongs, obtain the amplitude threshold corresponding to the current scanning position from the preset mapping table. The transparency coefficient and the amplitude threshold in the preset mapping table are negatively correlated.

[0009] Determine the signal reception time of the current scanning position as the reception time of the signal amplitude that is the first to be greater than or equal to the amplitude threshold in the reflected signal waveform collected by the photoelectric sensor at the current scanning position.

[0010] Further, the controller is further configured to perform the following steps:

[0011] Generate the thickness information d corresponding to the current scanning position of the sample to be measured according to the signal emission time t1, the signal reception time t2 corresponding to the current scanning position, and the distance L between the photoelectric sensor and the tissue tray.

[0012] d satisfies the following steps:

[0013] 。

[0014] Where C is the speed of light.

[0015] Further, the controller is further configured to perform the following steps:

[0016] Generate a three-dimensional contour model of the sample to be measured according to the thickness information corresponding to all scanning positions in the sample to be measured.

[0017] Further, the segmentation information includes multiple boundary image unit information. The boundary image unit information includes the contour information of the boundary image segmentation region, the first transparency coefficient, and the second transparency coefficient. The boundary image segmentation region includes an inward contraction region that contracts inward along the original segmentation boundary of the independent image segmentation region and an outward expansion region that expands outward along the original segmentation boundary of the independent image segmentation region. The first transparency coefficient is the transparency coefficient corresponding to the independent image segmentation region to which the inward contraction region belongs. The second transparency coefficient is the transparency coefficient corresponding to the independent image segmentation region to which the outward expansion region belongs.

[0018] The controller is further configured to perform the following steps:

[0019] If the mapped position of the current scanning position of the photoelectric sensor in the image information is located in the boundary image segmentation region, then generate the first predicted transparency coefficient Alpha corresponding to the current scanning position according to the image information of the independent image segmentation region to which the inward contraction region belongs 1 。Generate the second predicted transparency coefficient Alpha corresponding to the current scanning position according to the image information of the independent image segmentation region to which the outward expansion region belongs 2 。

[0020] According to Alpha 1 and Alpha 2 , generate the reference transparency coefficient Alpha corresponding to the current scanning position c . Alpha c satisfies the following conditions:

[0021] Alpha c = (Q2 × Alpha 1 + Q1 × Alpha 2 ) / h.

[0022] Where h is the width of the boundary image segmentation area, Q1 is the distance from the mapping position in the image information at the current scanning position to the boundary of the adduction area, and Q2 is the distance from the mapping position in the image information at the current scanning position to the boundary of the expansion area.

[0023] Take the transparency coefficient closest to Alpha among the first transparency coefficient and the second transparency coefficient as the transparency coefficient corresponding to the current scanning position. c

[0024] Further, after taking the transparency coefficient closest to Alpha among the first transparency coefficient and the second transparency coefficient as the transparency coefficient corresponding to the current scanning position, the controller is further configured to perform the following steps: c

[0025] According to the transparency coefficient corresponding to the current scanning position, obtain the amplitude threshold corresponding to the current scanning position from the preset mapping table.

[0026] Determine the reception time of the signal amplitude of the first signal greater than or equal to the amplitude threshold in the reflected signal waveform collected by the photoelectric sensor at the current scanning position as the signal reception time of the current scanning position.

[0027] Further, generate the first predicted transparency coefficient Alpha corresponding to the current scanning position according to the image information of the independent image segmentation area to which the adduction area belongs 1 , including:

[0028] Use the inpaint function to generate the first predicted transparency coefficient Alpha corresponding to the current scanning position according to the image information of the independent image segmentation area to which the adduction area belongs 1 .

[0029] Further, generate the first predicted transparency coefficient Alpha corresponding to the current scanning position according to the image information of the independent image segmentation area to which the adduction area belongs 1 , including:

[0030] ​​Using the interpolation method, according to the image information of the independent image segmentation region to which the adduction region belongs, generate the first predicted transparency coefficient Alpha corresponding to the current scanning position 1 。

[0031] Furthermore, the image acquisition device includes: an RGB camera. The image information includes an RGB image.

[0032] According to the transparency information of the image, perform image segmentation processing on the image information of the sample to be measured collected by the image acquisition device, and generate segmentation information corresponding to the image information, including:

[0033] Input the RGB image of the sample to be measured obtained by the RGB camera into the target U-Net model to generate segmentation information corresponding to the image information.

[0034] Furthermore, it further includes: a lateral movement component and a longitudinal movement component.

[0035] The longitudinal movement component is arranged above the lateral movement component, and the movement directions of the longitudinal movement component and the lateral movement component are perpendicular to each other.

[0036] The tissue tray is arranged on the lateral movement component, and the photoelectric sensor is arranged on the longitudinal movement component.

[0037] The lateral movement component includes: a first slider group, a second slider group, a lateral lead screw, a lead screw motor, and a tissue tray base.

[0038] The tissue tray base is arranged on the sliding parts of the first slider group and the second slider group. The driving part of the lateral lead screw is fixedly connected to the tissue tray base. The lead screw motor is fixedly connected to the lateral lead screw. The tissue tray base is used to place the tissue tray.

[0039] Furthermore, the longitudinal movement component includes: a longitudinal lead screw, a lead screw motor, and a bracket for the optical scanning measurement mechanism;

[0040] The lead screw motor is connected to the longitudinal lead screw, and the bracket for the optical scanning measurement mechanism is connected to the sliding part of the lead screw. The photoelectric sensor is arranged on the bracket for the optical scanning measurement mechanism.

[0041] The present invention has at least one of the following beneficial effects:

[0042] In the present invention, first, through image segmentation technology, the image of the sample to be measured (such as a biological sample tissue) collected is segmented according to different transparencies, so that regions with different transparency degrees in the tissue can be divided. Then, according to the corresponding relationship between the surface of the regions with different transparency degrees and the amplitude of the reflection signal, a preset mapping table is established. Then, according to the segmentation region to which the mapping position of the current scanning position in the image belongs, the amplitude threshold corresponding to the true reflection signal that better fits the surface of the current position is obtained from the mapping table. Furthermore, it can provide a favorable reference for screening the received signals, and the waveforms in the reflection signal waveform that may be the target received signals are screened through this amplitude threshold to ensure that the true reflection signals on the surface are screened out.

[0043] Meanwhile, since the distance between the surface of the sample to be measured and the optoelectronic sensor is less than the distance between the internal structure of the sample to be measured and the optoelectronic sensor, in the reflection signal waveform, the generation time of the signal corresponding to the surface of the sample to be measured is earlier than the generation time of the signal corresponding to the internal structure of the sample to be measured. Therefore, among the screened signals, the reception time of the amplitude of the first signal greater than or equal to the amplitude threshold is determined as the signal reception time of the current scanning position, so as to further improve the accuracy of the signal reception time and the accuracy of the thickness information of the biological tissue sample finally calculated. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0045] Figure 1 It is a schematic diagram of the overall structure of the automatic biological sample tissue topography measurement system provided by the embodiment of the present invention;

[0046] Figure 2 It is a schematic diagram of the decomposed structure of the automatic biological sample tissue topography measurement system provided by the embodiment of the present invention;

[0047] Figure 3 It is a schematic diagram of the decomposed structure of the base structure provided by the embodiment of the present invention;

[0048] Figure 4 It is a schematic diagram of the decomposed structure of the lateral movement component provided by the embodiment of the present invention;

[0049] Figure 5 It is a schematic diagram of the decomposed structure of the longitudinal movement component provided by the embodiment of the present invention;

[0050] Figure 6Schematic flow chart of the automatic biological sample tissue morphology measurement method provided by the embodiment of the present invention;

[0051] Figure 7 Schematic diagram of the biological sample tissue thickness calculation principle provided by the embodiment of the present invention;

[0052] Figure 8 Schematic diagram of the region after RGB image segmentation provided by the embodiment of the present invention.

[0053] Reference numerals

[0054] 1. Base; 101. Scanning measurement mechanism protective cover; 102. First support rod; 103. Second support rod; 104. Second support protective cover; 105. Installation base; 106. First support protective cover; 107. Third support rod; 108. Fourth support rod; 2. Lateral movement assembly; 201. First slider group; 202. Lateral lead screw mounting base; 203. Lateral lead screw; 204. First mounting claw; 205. Second mounting claw; 206. Tissue tray base; 207. Tissue tray; 208. Third mounting claw; 209. Lateral lead screw nut; 210. Fourth mounting claw; 211. Lateral lead screw motor; 212. Lateral lead screw motor mounting base; 213. Second slider group; 3. Longitudinal movement assembly; 301. Longitudinal lead screw mounting base; 302. Longitudinal lead screw; 303. Longitudinal movement assembly bracket; 304. Photoelectric sensor; 305. Sensor mounting plate; 306. Longitudinal lead screw motor; 307. Longitudinal lead screw motor mounting base; 308. Longitudinal nut guide rail group; 309. Longitudinal nut fitting; 310. Longitudinal nut; 40. Independent image segmentation region; 41. Boundary image segmentation region; 42. Original segmentation boundary. Detailed implementation manners

[0055] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without making creative efforts belong to the protection scope of the present invention.

[0056] As a possible embodiment of the present invention, as Figures 1 to 5 shown, an automatic biological sample tissue morphology measurement system is provided. The system includes: a controller, a tissue tray 207, a reflective photoelectric sensor, and an image acquisition device. The photoelectric sensor 304 is arranged above the tissue tray 207. The tissue tray 207 is a self-luminous tissue tray. Both the photoelectric sensor 304 and the image acquisition device are communicatively connected to the controller.

[0057] The measurement system also includes an installation and driving component for the tissue tray 207 and the photoelectric sensor 304, which is used to drive the photoelectric sensor 304 along a preset scanning path to scan and measure the sample to be measured (i.e., the biological sample tissue) placed on the tissue tray 207. Specifically, the installation and driving component includes the following specific structures:

[0058] A base 1, a lateral movement component 2, and a longitudinal movement component 3. The lateral movement component 2 is installed on the base 1 and can move horizontally left and right. The longitudinal movement component 3 is suspended on the base 1 and can move longitudinally back and forth, thereby driving the photoelectric sensor to measure the longitudinal and lateral dimensions and shapes of the sample tissue and generating a three-dimensional model.

[0059] As shown in FIG. 3, the base 1 includes a scanning and measuring mechanism protective cover 101, a first support rod 102, a second support rod 103, a second support protective cover 104, an installation base 105, a first support protective cover 106, a third support rod 107, and a fourth support rod 108.

[0060] The scanning and measuring mechanism protective cover 101 is a square protective cover installed outside the longitudinal movement component 3 and is fixedly connected to the first support rod 102, the second support rod 103, the third support rod 107, and the fourth support rod 108 by nuts.

[0061] The first support rod 102, the second support rod 103, the third support rod 107, and the fourth support rod 108 are threaded rods with smooth rods in the middle at both ends. The lower ends are threadedly installed on the installation base 105, and the upper ends are threaded for fixedly supporting the longitudinal movement component 3 and the scanning and measuring mechanism protective cover 101.

[0062] The second support protective cover 104 and the first support protective cover 106 are installed on the installation base 105 to assist the first support rod 102, the second support rod 103, the third support rod 107, and the fourth support rod 108 in fixedly supporting the longitudinal movement component 3 and the scanning and measuring mechanism protective cover 101.

[0063] The installation base 105 has threaded holes, which are installed in a matching manner with the first support rod 102, the second support rod 103, the third support rod 107, and the fourth support rod 108 and are used to fix the lateral movement component 2.

[0064] As Figure 4 shown, the lateral movement component 2 includes a first slider group 201, a lateral lead screw mounting base 202, a lateral lead screw 203, a first mounting claw 204, a second mounting claw 205, a tissue tray base 206, a tissue tray 207, a third mounting claw 208, a lateral lead screw nut 209, a fourth mounting claw 210, a lateral lead screw motor 211, a lateral lead screw motor mounting base 212, and a second slider group 213.

[0065] The first slider group 201 and the second slider group 213 are slider guide rail groups, guiding the movement of the transverse lead screw 203 and the transverse lead screw nut 209, installed on the mounting base 105, and installed in cooperation with the threaded holes on the mounting base 105.

[0066] The transverse lead screw mounting seat 202 is installed on the mounting base 105, serving as a bearing support seat, and is installed in cooperation with one end of the transverse lead screw 203.

[0067] The transverse lead screw 203 is a ball screw, with one end installed on the transverse lead screw mounting seat 202 and the other end connected to the transverse lead screw motor 211. The transverse lead screw nut 209 is a ball screw nut, installed in cooperation with the transverse lead screw 203. When the transverse lead screw 203 rotates, it drives the transverse lead screw nut 209 to move left and right, and its upper surface is fixedly connected to the tissue tray base 206 through screws, driving the tissue tray base 206 to move.

[0068] The first mounting claw 204, the second mounting claw 205, the third mounting claw 208, and the fourth mounting claw 210 are sample fixing claws on the tissue tray 207, made of colorless transparent material.

[0069] The tissue tray 207 is square. Its lower surface can be fixed to the weight sensor for weighing. A positioning groove is opened 5 mm away from the edge on its upper surface, serving as the starting position for the cyclic movement of the longitudinal movement component 3. The tissue sample is placed or fixed on the tissue tray 207.

[0070] The transverse lead screw motor 211 is a permanent magnet synchronous motor with high precision, which is the power source for horizontally moving the tissue tray 207 and is fixed on the transverse lead screw motor mounting seat 212.

[0071] The transverse lead screw motor mounting seat 212 is used to install the transverse lead screw motor 211 and is installed in cooperation with the threaded holes on the mounting base 105.

[0072] As Figure 5 shown, the longitudinal movement component 3 includes a longitudinal lead screw mounting seat 301, a longitudinal lead screw 302, a longitudinal movement component bracket 303, a photoelectric sensor 304, a sensor mounting plate 305, a longitudinal lead screw motor 306, a longitudinal lead screw motor mounting seat 307, a longitudinal nut guide rail group 308, a longitudinal nut fitting 309, and a longitudinal nut 310.

[0073] The longitudinal lead screw mounting seat 301 is installed on the longitudinal movement component bracket 303, serving as a bearing support seat, and is installed in cooperation with one end of the longitudinal lead screw 302.

[0074] The longitudinal lead screw 302 is a ball screw, with one end installed on the longitudinal lead screw mounting seat 301 and the other end connected to the longitudinal lead screw motor 306.

[0075] The longitudinal movement component bracket 303 is the mounting base for the longitudinal movement component 3. It is fixedly connected to the first support rod 102, the second support rod 103, the third support rod 107, the fourth support rod 108, and the scanning and measuring mechanism protective cover 101 by nuts. There are threaded holes on it, which are convenient for fitting and installing with the longitudinal lead screw motor mounting seat 307 and the longitudinal nut guide group 308.

[0076] The photoelectric sensor 304 is installed on the sensor mounting plate 305 and is a tissue sample profile recognition sensor for collecting the morphological information of the sample to be measured. The photoelectric sensor 304 in this embodiment is a reflective photoelectric sensor. It determines the presence, distance, or surface characteristics of an object by emitting a light beam and detecting the reflected light. Specifically, it determines the presence, distance, or surface characteristics of an object according to the intensity of the reflected light.

[0077] The sensor mounting plate 305 is provided with holes and is fitted and installed with the longitudinal nut guide group 308 and the photoelectric sensor 304. It moves longitudinally back and forth with the slider of the longitudinal nut guide group 308, driving the photoelectric sensor 304 to collect profile information.

[0078] The longitudinal lead screw motor 306 is a permanent magnet synchronous motor with high precision and is the power source for the longitudinal movement component 3. It is fixed on the longitudinal lead screw motor mounting seat 307.

[0079] The longitudinal lead screw motor mounting seat 307 is used to install the longitudinal lead screw motor 306 and is fitted and installed with the threaded holes on the longitudinal movement component bracket 303.

[0080] The longitudinal nut guide group 308 guides the movement of the longitudinal nut 310 and the longitudinal lead screw 302. It is installed on the longitudinal movement component bracket 303 and is fitted and installed with the threaded holes on the longitudinal movement component bracket 303.

[0081] The longitudinal nut fitting 309 is fixed to the longitudinal nut 310 and is also fixed to the sensor mounting plate 305 by screws, driving the photoelectric sensor 304 and the sensor mounting plate 305 to move.

[0082] The longitudinal nut 310 is a ball screw nut and is fitted and installed with the longitudinal lead screw 302. It is fixed to the longitudinal nut fitting 309. The rotation of the longitudinal lead screw 302 drives the longitudinal nut 310 to move longitudinally back and forth, and further drives the longitudinal nut fitting 309, the sensor mounting plate 305, and the photoelectric sensor 304 to move longitudinally back and forth.

[0083] Specifically, the scanning method of the morphology measurement system in this embodiment includes the following steps:

[0084] S1: Place the tissue sample to be measured in the tissue tray 207 of the transverse movement component 2, flatten and unfold it, and then go to step S2;

[0085] S2: Starting device, the photoelectric sensor 304 determines the initial position according to the right-edge positioning groove and the front positioning groove of the tissue tray 207, and proceeds to step S3; the sampling frequency of the photoelectric sensor 304 can be 10 Hz, and can be specifically adjusted according to the modeling accuracy.

[0086] S3: Wait for the photoelectric sensor 304 to reach the initial position, start scanning the tissue sample to be measured, and proceed to step S4;

[0087] S4: The transverse movement component 2 moves horizontally from left to right at a speed of 1 cm / s, and the photoelectric sensor 304 remains stationary, and proceeds to step S5;

[0088] S5: Judge the output signal of the photoelectric sensor 304. If it reaches the right-edge positioning groove, proceed to step S6; otherwise, continue step S4;

[0089] S6: The transverse movement component 2 stops moving and remains stationary, and the photoelectric sensor 304 steps longitudinally from front to back by 0.05 cm, and proceeds to step S7;

[0090] S7: Judge the output signal of the photoelectric sensor 304. If it reaches the rear-edge positioning groove, proceed to step S8; otherwise, continue step S4, but at this time the moving direction of the transverse movement component 2 is opposite to the previous moving direction, thus forming a serpentine scanning path.

[0091] S8: Scanning ends.

[0092] The automatic biological sample tissue morphology measurement system provided in this embodiment fully considers the morphological characteristics of biological samples, uses a high-precision photoelectric sensor, can complete the measurement of the tissue morphology of semi-transparent and opaque biological samples, and has a high recognition accuracy for smaller biological tissues, reducing the previous manual recognition error, improving the accuracy of data in biological experiments and clinical sample collection, and reducing the burden on experimental personnel. On the basis of morphological recognition, the output signal is recognized by the host computer to complete the tissue morphology modeling of biological samples, which is convenient for recording and retention. The usage method of the present invention is simple and can meet different sample requirements.

[0093] In addition, the following solution is also provided in this embodiment, as Figure 6 shown. This solution is used to obtain the thickness information of the biological sample tissue at each scanning position for subsequent three-dimensional modeling. Specifically, this method is executed by the controller and includes the following steps:

[0094] S100: Based on the transparency information of the image, perform image segmentation processing on the image information of the sample to be measured (i.e., biological sample tissue) collected by the image acquisition device, and generate segmentation information corresponding to the image information. The segmentation information includes multiple independent image unit information. The independent image unit information includes the contour information of the independent image segmentation region 40 and the transparency coefficient. The transparency coefficient is positively correlated with transparency.

[0095] Specifically, the image acquisition device includes: an RGB camera, and the image information includes an RGB image.

[0096] S100 includes:

[0097] S101: Input the RGB image of the sample to be measured obtained by the RGB camera into the target U-Net model to generate segmentation information corresponding to the image information.

[0098] In this embodiment, since the tissue tray 207 is a self-luminous tray, if there are different regions of the biological sample tissue with a semi-transparent tissue structure, due to the different degrees of transparency, the brightness of the light transmitted through different regions will inevitably be different. Furthermore, when taking pictures with the RGB camera, the boundaries of different semi-transparent regions can be represented more clearly. So as to facilitate the target U-Net model to more accurately divide different semi-transparent regions in the picture.

[0099] When training the U-Net model, a large number of pictures of different biological sample tissues can be obtained by taking pictures and then manually annotated to form a training set. In addition, according to needs, the input features can be changed to a combined image composed of an RGB image and the corresponding grayscale image and input into the target U-Net model together to further improve the segmentation accuracy of the model.

[0100] In addition, in this embodiment, the image acquisition device further includes a polarization camera. The characteristics of the transparent material (such as refractive index, surface roughness, internal scattering) will affect the polarization state of light. The polarization camera can capture the polarization information of light. Thus, according to the high and low degree of polarization (DoP) in the reflected light, different transparency regions in the tissue can be labeled. Then, combined with the RGB image obtained by the RGB camera, more accurate transparency information can be generated. Thus, the target U-Net model can perform more accurate segmentation processing according to the transparency information on the image. In addition, when obtaining training sample data, the transparency conditions of different parts can also be identified by existing optoelectronic sensors, and then the annotation of the transparency coefficient can be generated, so as to quickly obtain a large number of training samples.

[0101] Such as Figure 8As shown, after the segmentation process of this step, the image regions corresponding to different transparency regions in the image will be separated. However, since the transparency and texture of the biological sample tissue usually change slowly at the position of the segmentation boundary, the transparency near the segmentation boundary is generally between the two segmentation regions. Based on this, the present invention needs to perform special processing on the region of the segmentation boundary. Specifically, in the present invention, the region is further widened respectively inward and outward along the segmentation boundary to form a boundary image segmentation region 41, which includes the transparency transition region near the segmentation boundary. Thus, after the final segmentation in the present invention, there are two types of segmentation regions in total. One is an independent image segmentation region 40 with relatively uniform transparency, and the other is a boundary image segmentation region 41 with relatively gradual transparency change. The methods for obtaining the thickness of the sample tissue in the above two regions are different, and the specific description is as follows.

[0102] S200: If the mapped position of the current scanning position of the photoelectric sensor in the image information is located in the independent image segmentation region 40, then according to the transparency coefficient corresponding to the independent image segmentation region 40 to which the mapped position belongs, the amplitude threshold corresponding to the current scanning position is obtained from the preset mapping table. The transparency coefficient and the amplitude threshold in the preset mapping table are negatively correlated.

[0103] Specifically, in this embodiment, the transparency coefficient corresponding to the independent image segmentation region 40 can be the average transparency of all pixel points included in the independent image segmentation region 40.

[0104] The preset mapping table in this step is for the transparency coefficient and the amplitude threshold of the received signal. Specifically, the corresponding relationship can be set according to the actual usage situation. Different transparency levels correspond to different amplitude thresholds, which can provide a favorable reference for screening the received signals, so as to screen the waveforms in the reflected signal waveform that may be the target received signal through this amplitude threshold to ensure that the true reflected signal on the surface is screened out. Preferably, the amplitude threshold can be set slightly smaller than the amplitude of the true received signal. Thus, it can not only ensure that the true received signal is not missed as much as possible, but also filter out the influence of most environmental noises and reduce the amount of data for subsequent processing.

[0105] S300: In the reflected signal waveform collected by the photoelectric sensor at the current scanning position, the reception time of the signal amplitude that is the first to be greater than or equal to the amplitude threshold is determined as the signal reception time of the current scanning position.

[0106] Since the distance between the surface of the biological sample tissue and the optoelectronic sensor is less than the distance between the internal structure of the biological sample tissue and the optoelectronic sensor, in the reflected signal waveform, the generation time of the signal corresponding to the surface of the biological sample tissue is earlier than the generation time of the signal corresponding to the internal structure of the biological sample tissue. Therefore, among the selected signals, the reception time of the first signal amplitude greater than or equal to the amplitude threshold is determined as the signal reception time at the current scanning position, so as to further improve the accuracy of the signal reception time and the accuracy of the thickness information of the biological tissue sample finally calculated.

[0107] S400: Generate the thickness information d corresponding to the current scanning position of the sample to be measured according to the signal emission time t1, the signal reception time t2 corresponding to the current scanning position, and the distance L between the optoelectronic sensor and the tissue tray.

[0108] As Figure 7 shown, d satisfies the following steps:

[0109] .

[0110] Where C is the speed of light.

[0111] S500: Generate a three-dimensional contour model of the sample to be measured according to the thickness information corresponding to all scanning positions in the sample to be measured.

[0112] Through the above calculations, the thickness information corresponding to each scanning position point can be obtained. After the optoelectronic sensor 304 completes scanning, a large number of contour points on the surface of the biological sample tissue can be obtained, and a three-dimensional contour model of the biological sample tissue can be formed based on these points.

[0113] As another possible embodiment of the present invention, the segmentation information includes multiple boundary image unit information. The boundary image unit information includes the contour information of the boundary image segmentation region 41, the first transparency coefficient, and the second transparency coefficient. The boundary image segmentation region 41 includes an inward contraction region that contracts inward along the original segmentation boundary 42 of the independent image segmentation region 40 and an outward expansion region that expands outward along the original segmentation boundary 42 of the independent image segmentation region 40. The first transparency coefficient is the transparency coefficient corresponding to the independent image segmentation region 40 to which the inward contraction region belongs. The second transparency coefficient is the transparency coefficient corresponding to the independent image segmentation region 40 to which the outward expansion region belongs. Specifically, the outward expansion region and the inward contraction region can be annular regions formed by expanding or contracting several (such as 3) pixels outward or inward along the original segmentation boundary 42.

[0114] The controller is further configured to perform the following steps:

[0115] S600: If the mapped position of the current scanning position of the photoelectric sensor in the image information is located in the boundary image segmentation region 41, then, based on the image information of the independent image segmentation region 40 to which the inward region belongs, generate a first predicted transparency coefficient Alpha corresponding to the current scanning position 1 。Based on the image information of the independent image segmentation region 40 to which the outward expansion region belongs, generate a second predicted transparency coefficient Alpha corresponding to the current scanning position 2 。

[0116] In this step, since the transparency of the boundary image segmentation region 41 transitions from two adjacent independent image segmentation regions 40, thus, based on the image information in the two independent image segmentation regions 40, the transparency coefficient at the current scanning position can be inferred

[0117] In this embodiment, Alpha can be obtained through the following method 1 and Alpha 2 , because, Alpha 1 and Alpha 2 are obtained in the same way, thus only the obtaining method of Alpha 1 will be taken as an example for illustration

[0118] Firstly

[0119] Use the inpaint function to generate a first predicted transparency coefficient Alpha corresponding to the current scanning position based on the image information of the independent image segmentation region 40 to which the inward region belongs 1 。

[0120] The inpaint function is an image restoration algorithm in OpenCV. It can restore the unknown region based on the information of the known region. Thus, based on the known transparency information in the independent image segmentation region 40 image, the transparency coefficient at the current scanning position can be generated. Moreover, this function is more suitable for regions with similar textures and structures, and this feature exactly matches the imaging characteristics of biological sample tissues in the boundary image segmentation region 41 in this usage scenario

[0121] Secondly

[0122] Use the interpolation method to generate a first predicted transparency coefficient Alpha corresponding to the current scanning position based on the image information of the independent image segmentation region 40 to which the inward region belongs 1 。

[0123] Similarly, since the transparency of the boundary image segmentation region 41 transitions from two adjacent independent image segmentation regions 40, the transparency of the boundary image segmentation region 41 should smoothly change from the transparencies of the two independent image segmentation regions 40. Thus, using the transparency values of the known regions, the transparency values of the unknown regions are predicted through an interpolation algorithm (such as bilinear interpolation or bicubic interpolation).

[0124] S700: According to Alpha 1 and Alpha 2 , generate the reference transparency coefficient Alpha c corresponding to the current scanning position. Alpha c satisfies the following conditions:

[0125] Alpha c = (Q2 × Alpha 1 + Q1 × Alpha 2 ) / h.

[0126] Where h is the width of the boundary image segmentation region 41, Q1 is the distance from the mapped position in the current scanning position image information to the boundary of the adduction region, and Q2 is the distance from the mapped position in the current scanning position image information to the boundary of the expansion region.

[0127] Usually, for the current scanning position in the boundary image segmentation region 41, there will also be a situation where it is closer to the independent image segmentation region 40 on the adduction region side or closer to the independent image segmentation region 40 on the expansion region side. And usually, the closer the position is to the independent image segmentation region 40, the higher the accuracy of the predicted transparency information. Thus, in this embodiment, weights for Alpha 1 and Alpha 2 are formed by distance, and Q2 / h is used as the weight for Alpha 1 , and Q1 / h is used as the weight for Alpha 2 . Thus, if the mapped position in the current scanning position image information is closer to the independent image segmentation region 40 on the adduction region side, then Q2 / h > Q1 / h, and this will cause Alpha c to occupy a larger weight proportion when generating Alpha 1 , so that Alpha c will be closer to Alpha 1 generated according to the image information of the independent image segmentation region 40 to which the adduction region belongs.

[0128] S800: Use the transparency coefficient closest to Alpha c among the first transparency coefficient and the second transparency coefficient as the transparency coefficient corresponding to the current scanning position.

[0129] In this embodiment, by comparing the first transparency coefficient and the second transparency coefficient with the predicted Alpha c more accurate acquisition of the transparency coefficient corresponding to the current transition position can be achieved, and then the corresponding amplitude threshold can be more accurately obtained from the mapping table, and finally the signal reception time at the current scanning position can be more accurately determined.

[0130] In addition, in this embodiment, a more fine-grained correspondence between the transparency coefficient and the amplitude threshold can be adaptively set in the preset mapping table. Then, after S700 obtains Alpha c the transparency coefficient corresponding to the current scanning position can be directly obtained from the preset mapping table according to Alpha c In addition, although the steps of the methods in the present disclosure are described in a specific order in the drawings, this does not require or imply that these steps must be performed in that specific order, or that all of the shown steps must be performed to achieve the desired result. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step for execution, and / or one step may be decomposed into multiple steps for execution, etc.

[0131] Through the description of the above embodiments, those skilled in the art can easily understand that the example embodiments described herein can be implemented by software or by a combination of software and necessary hardware. Therefore, the technical solutions according to the embodiments of the present disclosure can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, a USB flash drive, a mobile hard disk, etc.) or on a network, including several instructions to enable a computing device (such as a personal computer, a server, a mobile terminal, or a network device, etc.) to execute the methods according to the embodiments of the present disclosure.

[0132] The above are only the specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present invention should be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

[0133] The above are only the specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any changes or substitutions that can be easily thought of by those skilled in the art within the technical scope disclosed by the present invention should be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention should be subject to the protection scope of the claims.

Claims

1. An automatic biological sample tissue morphology measurement system, characterized in that The system includes: a controller, a tissue tray, a reflective photoelectric sensor, and an image acquisition device; the photoelectric sensor is arranged above the tissue tray; the tissue tray is a self-luminous tissue tray; both the photoelectric sensor and the image acquisition device are communicatively connected to the controller; The controller is configured to perform the following steps: According to the transparency information of the image, perform image segmentation processing on the image information of the sample to be measured collected by the image acquisition device, and generate segmentation information corresponding to the image information; the segmentation information includes multiple independent image unit information; the independent image unit information includes the contour information and transparency coefficient of the independent image segmentation region; the transparency coefficient is positively correlated with transparency; If the mapped position of the current scanning position of the photoelectric sensor in the image information is located in the independent image segmentation region, then according to the transparency coefficient corresponding to the independent image segmentation region to which the mapped position belongs, obtain the amplitude threshold corresponding to the current scanning position from the preset mapping table; the transparency coefficient and the amplitude threshold in the preset mapping table are negatively correlated; Determine the signal reception time of the current scanning position as the reception time of the first signal amplitude greater than or equal to the amplitude threshold in the reflected signal waveform collected by the photoelectric sensor at the current scanning position.

2. The automatic biological sample tissue topography measurement system according to claim 1, characterized in that, The controller is further configured to perform the following steps: Generate the thickness information d corresponding to the current scanning position of the sample to be measured according to the signal emission time t1, signal reception time t2 corresponding to the current scanning position, and the distance L between the photoelectric sensor and the tissue tray; d satisfies the following steps: ； Where C is the speed of light.

3. An automatic biological sample tissue morphology measurement system according to claim 1, characterized in that, The controller is further configured to perform the following steps: Generate a three-dimensional contour model of the sample to be measured according to the thickness information corresponding to all scanning positions in the sample to be measured.

4. An automatic biological sample tissue morphology measurement system according to claim 1, wherein The segmentation information includes multiple boundary image unit information; the boundary image unit information includes the contour information, first transparency coefficient, and second transparency coefficient of the boundary image segmentation region; the boundary image segmentation region includes an inward contraction region that contracts inward along the original segmentation boundary of the independent image segmentation region and an outward expansion region that expands outward along the original segmentation boundary of the independent image segmentation region; the first transparency coefficient is the transparency coefficient corresponding to the independent image segmentation region to which the inward contraction region belongs; the second transparency coefficient is the transparency coefficient corresponding to the independent image segmentation region to which the outward expansion region belongs; The controller is further configured to perform the following steps: If the mapped position of the current scanning position of the photoelectric sensor in the image information is located in the boundary image segmentation area, then according to the image information of the independent image segmentation area to which the adduction area belongs, a first predicted transparency coefficient Alpha corresponding to the current scanning position is generated 1 ; Generate a second predicted transparency coefficient Alpha corresponding to the current scanning position based on the image information of the independent image segmentation region to which the extended region belongs 2 ; According to Alpha 1 and Alpha 2 , generate the reference transparency coefficient Alpha corresponding to the current scanning position c ; Alpha c satisfies the following conditions: Alpha c = (Q2 × Alpha 1 + Q1 × Alpha 2 ) / h; Where h is the width of the boundary image segmentation region, Q1 is the distance from the mapped position of the current scanning position in the image information to the boundary of the inward contraction region, and Q2 is the distance from the mapped position of the current scanning position in the image information to the boundary of the outward expansion region; Take the transparency coefficient among the first transparency coefficient and the second transparency coefficient that is closest to Alpha c as the transparency coefficient corresponding to the current scanning position.

5. An automatic biological sample tissue morphology measurement system according to claim 4, characterized in that After taking the transparency coefficient closest to Alpha among the first transparency coefficient and the second transparency coefficient as the transparency coefficient corresponding to the current scanning position, the controller is further configured to perform the following steps: c ​ According to the transparency coefficient corresponding to the current scanning position, obtain the amplitude threshold corresponding to the current scanning position from the preset mapping table; Determine the signal reception time of the current scanning position as the reception time of the first signal amplitude greater than or equal to the amplitude threshold in the reflected signal waveform collected by the photoelectric sensor at the current scanning position.

6. The automatic biological sample tissue morphology measurement system according to claim 4, wherein Generate a first predicted transparency coefficient Alpha corresponding to the current scanning position based on the image information of the independent image segmentation region to which the adduction region belongs 1 , including: Use the inpaint function to generate the first predicted transparency coefficient Alpha corresponding to the current scanning position according to the image information of the independent image segmentation area to which the adduction area belongs 1 .

7. An automatic biological sample tissue morphology measurement system according to claim 4, wherein Generate a first predicted transparency coefficient Alpha corresponding to the current scanning position based on the image information of the independent image segmentation region to which the adduction region belongs 1 , including: Using the interpolation method, generate the first predicted transparency coefficient Alpha corresponding to the current scanning position according to the image information of the independent image segmentation region to which the adduction region belongs 1 .

8. An automatic biological sample tissue morphology measurement system according to claim 1, characterized in that, The image acquisition device includes: an RGB camera; the image information includes an RGB image; According to the transparency information of the image, perform image segmentation processing on the image information of the sample to be measured collected by the image acquisition device, and generate segmentation information corresponding to the image information, including: Input the RGB image of the sample to be measured obtained by the RGB camera into the target U-Net model to generate segmentation information corresponding to the image information.

9. The automatic biological sample tissue morphology measurement system according to claim 1, characterized in that It further includes: a lateral movement component and a longitudinal movement component; The longitudinal movement component is arranged above the lateral movement component, and the movement directions of the longitudinal movement component and the lateral movement component are perpendicular to each other; The tissue tray is arranged on the lateral movement component, and the photoelectric sensor is arranged on the longitudinal movement component; The lateral movement component includes: a first slider group, a second slider group, a lateral lead screw, a lead screw motor, and a tissue tray base; The tissue tray base is arranged on the sliding parts of the first slider group and the second slider group; the driving part of the lateral lead screw is fixedly connected to the tissue tray base; the lead screw motor is fixedly connected to the lateral lead screw; the tissue tray base is used for placing the tissue tray.

10. An automatic biological sample tissue topography measurement system according to claim 9, characterized in that, The longitudinal movement component includes: a longitudinal lead screw, a lead screw motor, and an optical scanning measurement mechanism bracket; The lead screw motor is connected to the longitudinal lead screw, and the optical scanning measurement mechanism bracket is connected to the sliding part of the lead screw; the photoelectric sensor is arranged on the optical scanning measurement mechanism bracket.

Images