Microfluidic chip channel identification method, device and equipment and readable storage medium

Abstract

The application belongs to the technical field of microfluidic chip quality detection, and discloses a microfluidic chip channel identification method, device, equipment and readable storage medium, comprising: acquiring a microfluidic chip image collected by an imaging sensor; preprocessing the microfluidic chip image to obtain a first image; sliding a sliding window of a preset size on the first image, and saving the center coordinates of the sliding window meeting a preset condition as first target points to a target coordinate list; judging whether all the first target points in the target coordinate list are in the microfluidic chip channel; performing a flood fill processing on the first target points in the chip channel as seed points to obtain an image mask of a channel image in the microfluidic chip image; and cutting the microfluidic chip image by using the image mask to obtain a target image containing only the microfluidic chip channel. The application can realize efficient and high-precision division of the channel in the microfluidic chip image.

Description

Technical Field

[0001] This application relates to the technical field of microfluidic chip quality inspection, and in particular to a method, apparatus, device and readable storage medium for microfluidic chip channel identification. Background Technology

[0002] Biochip technology originated in the 1980s and is also known as "microfluidic technology" or "lab-on-a-chip." It's a technology that integrates basic operations such as sample preparation, reaction, separation, and detection in biological, chemical, and medical analysis processes onto a single micrometer-scale chip, automating the entire analysis process. It is characterized by low cost, small sample size, short processing time, and simple operation. However, compared to the increasing maturity of the microfluidic chip industry, the development of microfluidic chip quality inspection technology has lagged behind. Due to the small size of microfluidic chips, their channels are invisible to the naked eye and require manual observation using an integrated platform. Prolonged manual visual inspection can easily cause eye fatigue for the inspectors, thus affecting the test results. Currently, manual visual inspection remains the primary method for microfluidic chip testing.

[0003] For microfluidic chip images, algorithms can be designed to locate the chip channels within the image, thus enabling channel segmentation. However, digital PCR instrument microfluidic chip channel images present two main technical challenges: incomplete channel imaging and overall chip transparency, making channel identification difficult. Firstly, the chip is generally transparent, its color determined by the background color under illumination, lacking color characteristics. Furthermore, the chip is thicker at the droplet outlet, and shadows under illumination partially obscure the channel observation. Secondly, the incomplete imaging prevents observation of the entire channel. Unlike conventional detection methods where the target is fully visible in the image, observing the entire microfluidic chip channel requires sequential observation of individual chip structures to determine channel cleanliness. This means that traditional template matching methods used in image processing cannot be used for detection. Summary of the Invention

[0004] This application provides a method, apparatus, device, and readable storage medium for channel identification of microfluidic chips, which can achieve efficient and high-precision segmentation of channels in microfluidic chip images.

[0005] In a first aspect, embodiments of this application provide a method for identifying channels in a microfluidic chip, the method comprising:

[0006] Acquire images of the microfluidic chip captured by the imaging sensor;

[0007] The microfluidic chip image is preprocessed to obtain a first image, wherein the preprocessing includes image enhancement, image binarization, and small target removal;

[0008] Slide a sliding window of a preset size on the first image, and save the center coordinates of the sliding window that meets the preset conditions as the first target point to the target coordinate list. The preset conditions are the coordinates of the target pixel value appearing in the sliding window.

[0009] Determine whether all first target points in the target coordinate list are within the microfluidic chip channel;

[0010] The first target point within the chip channel is used as the seed point for flood filling processing to obtain the image mask of the channel image in the microfluidic chip image. The gray value of the seed point after flood filling processing is the target pixel value.

[0011] The microfluidic chip image is cropped using the image mask to obtain a target image containing only the microfluidic chip channels.

[0012] Furthermore, the step of preprocessing the microfluidic chip image to obtain the first image includes:

[0013] The MSRCP algorithm, developed based on the Retinex algorithm, is used to enhance images of microfluidic chips to obtain a second image.

[0014] The second image is binarized based on the adaptive equalization algorithm to obtain the third image. During the binarization process, the gray value of the microfluidic chip channel contour part in the microfluidic chip image is set to the target pixel value.

[0015] The first image is obtained by removing background impurities from the third image based on the small target removal algorithm. In the background impurity removal process, connected regions with an overall pixel value not greater than a preset threshold are removed.

[0016] Furthermore, the step of determining whether all first target points in the target coordinate list are within the microfluidic chip channel includes:

[0017] Check all first target points in the target coordinate list. The check includes emitting line segments with a length not greater than the minimum length and width of the sliding window in the four directions of up, down, left and right, centered on the first target point, and determining the intersection points of the line segments with the microfluidic chip channel contour in the first image.

[0018] If line segments in the up-down or left-right directions intersect simultaneously, and the absolute value of the difference between the number of intersections on both sides of the line segments in the up-down or left-right directions is less than a preset difference, then the first target point is determined to be within the microfluidic chip channel.

[0019] Furthermore, the preset size is set to 1 / 16th of the size of the first image.

[0020] Secondly, the present invention also provides a microfluidic chip channel identification device, the device comprising:

[0021] The acquisition module is used to acquire images of the microfluidic chip collected by the imaging sensor;

[0022] An image preprocessing module is used to preprocess the microfluidic chip image to obtain a first image, wherein the preprocessing includes image enhancement, image binarization, and small target removal;

[0023] The sliding window processing module is used to slide a sliding window of a preset size on the first image and save the center coordinates of the sliding window that meets the preset conditions as the first target point to the target coordinate list. The preset conditions are the coordinates of the target pixel value appearing in the sliding window.

[0024] The judgment module is used to determine whether all the first target points in the target coordinate list are within the microfluidic chip channel;

[0025] The flood filling processing module is used to perform flood filling processing on the first target point in the chip channel as the seed point to obtain the image mask of the channel image in the microfluidic chip image. The gray value of the seed point after flood filling processing is the target pixel value.

[0026] The cropping module is used to crop the microfluidic chip image using the image mask to obtain a target image that contains only the microfluidic chip channels.

[0027] Furthermore, the image preprocessing module is specifically used for:

[0028] The MSRCP algorithm, developed based on the Retine microfluidic chip channel recognition algorithm, is used to enhance the microfluidic chip image to obtain a second image.

[0029] The second image is binarized based on the adaptive equalization algorithm to obtain the third image. During the binarization process, the gray value of the microfluidic chip channel contour part in the microfluidic chip image is set to the target pixel value.

[0030] The first image is obtained by removing background impurities from the third image based on the small target removal algorithm. In the background impurity removal process, connected regions with an overall pixel value not greater than a preset threshold are removed.

[0031] Furthermore, the judgment module is specifically used for:

[0032] Check all first target points in the target coordinate list. The check includes emitting line segments with a length not greater than the minimum length and width of the sliding window in the four directions of up, down, left and right, centered on the first target point, and determining the intersection points of the line segments with the microfluidic chip channel contour in the first image.

[0033] If line segments in the up-down or left-right directions intersect simultaneously, and the absolute value of the difference between the number of intersections on both sides of the line segments in the up-down or left-right directions is less than a preset difference, then the first target point is determined to be within the microfluidic chip channel.

[0034] Furthermore, the preset size is set to 1 / 16th of the size of the first image.

[0035] Thirdly, this application also provides a microfluidic chip channel identification device, which includes a processor, a memory, and a microfluidic chip channel identification program stored in the memory and executable by the processor. When the microfluidic chip channel identification program is executed by the processor, it implements the steps of the microfluidic chip channel identification method as described above.

[0036] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a microfluidic chip channel identification program, wherein when the microfluidic chip channel identification program is executed by a processor, it implements the steps of the microfluidic chip channel identification method as described above.

[0037] In summary, compared with the prior art, the beneficial effects of the technical solution provided in this application include at least the following:

[0038] This application provides a method, apparatus, device, and readable storage medium for identifying microfluidic chip channels. The method involves preprocessing, sliding window processing, flooding processing, and cropping of the acquired microfluidic chip image to obtain a target image containing only the microfluidic chip channels. This enables efficient and high-precision segmentation of channels within the microfluidic chip image. Furthermore, the obtained target image corresponds to only the visible chip channels within the microfluidic chip image, thus providing timely assistance to manual visual inspection and replacing the existing step of determining channel locations during manual inspection. Subsequent manual inspection can further check the determined target image for foreign objects, quality defects, etc., effectively alleviating visual fatigue for inspectors during manual inspection. Attached Figure Description

[0039] Figure 1 A flowchart illustrating a microfluidic chip channel identification method provided in one embodiment of this application;

[0040] Figure 2 A schematic diagram of sliding window processing for a microfluidic chip channel identification method provided in one embodiment of this application;

[0041] Figure 3 A schematic diagram of the functional modules of a microfluidic chip channel identification device provided in one embodiment of this application;

[0042] Figure 4 This is a schematic diagram of the hardware structure of a microfluidic chip channel identification device involved in one embodiment of this application. Detailed Implementation

[0043] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0044] Please see Figure 1 This application provides a method for identifying channels in a microfluidic chip, which specifically includes:

[0045] Step S10: Acquire the microfluidic chip image captured by the imaging sensor;

[0046] Step S20: Preprocess the microfluidic chip image to obtain a first image, wherein the preprocessing includes image enhancement, image binarization, and small target removal;

[0047] Step S30: Slide a sliding window of a preset size on the first image, and save the center coordinates of the sliding window that meets the preset conditions as the first target point to the target coordinate list. The preset conditions are the coordinates of the target pixel value appearing in the sliding window.

[0048] Step S40: Determine whether all the first target points in the target coordinate list are within the microfluidic chip channel;

[0049] Step S50: The first target point in the chip channel is used as the seed point for flood filling processing to obtain the image mask of the channel image in the microfluidic chip image, wherein the gray value of the seed point after flood filling processing is the target pixel value.

[0050] Step S60: Use the image mask to crop the microfluidic chip image to obtain a target image that contains only the microfluidic chip channels.

[0051] In this embodiment, the microfluidic chip image is obtained by directly photographing the chip under natural light using an imaging sensor such as a digital microscope. Considering the difficulties in channel identification in existing solutions due to incomplete imaging channels and the overall transparency of the chip, conventional template matching strategies, when searching for chip channels in microfluidic chip images, only obtain partial channels because the entire chip cannot be observed. Furthermore, due to the transparency, only the outline of the chip channels is obtained, not the channels themselves, making channel segmentation in the microfluidic chip image impossible. Therefore, to achieve efficient and high-precision channel segmentation in microfluidic chip images, this embodiment preprocesses, performs sliding window processing, flooding processing, and cropping processing on the acquired microfluidic chip image after acquisition by the imaging sensor, resulting in a target image containing only the microfluidic chip channels. Since the microfluidic chip image only contains a portion of the chip channels, the resulting target image also corresponds to containing only the visible chip channel portions of the microfluidic chip image. Although the target image only contains the chip channel portion visible in the microfluidic chip image, it can promptly assist manual visual inspection by replacing the existing step of determining the channel location during manual visual inspection. Subsequent manual visual inspection can then further check the channel for foreign objects, quality defects, or other issues on the determined target image, thus effectively alleviating visual fatigue for inspectors during manual visual inspection.

[0052] Specifically, because the chip is entirely transparent, the channel outline of the microfluidic chip is not clearly distinguishable from other parts of the chip. Therefore, after acquiring the microfluidic chip image from the imaging sensor, the microfluidic chip image undergoes preprocessing including image enhancement, image binarization, and small target removal to obtain the first image. Image enhancement increases the brightness of the channel outline (corresponding to the shadow area) in the microfluidic chip image, making the contrast between the chip outline and other chip parts more obvious, laying the foundation for subsequent image processing algorithms. Since the main goal is to segment the channels in the microfluidic chip, the grayscale values ​​of the chip channel outline are binarized to distinguish them from the grayscale values ​​of other parts. Furthermore, since small targets similar to the chip channel outline may appear in other parts, these small targets could interfere with channel segmentation in the chip image. Therefore, small target removal is performed during the binarization process of the microfluidic chip image to ensure that the preserved channel outline is clearer and more accurate.

[0053] Based on the premise that the preserved channel contours are clearer and more accurate, in order to more accurately divide the actual intra-channel regions based on the channel contours, this embodiment performs sliding window processing and flooding processing on the preprocessed first image to obtain the image mask of the intra-channel region. Specifically, when performing sliding window processing on the first image, a sliding window of a preset size slides across the first image. When the coordinates of the target pixel value (corresponding to the grayscale value of the binarized chip channel contour portion) appear in the sliding window, it is confirmed that the sliding window has contacted the chip channel contour point. (Refer to...) Figure 2 At this point, the center coordinates of the sliding window can be saved as the first target point to the target coordinate list. After sliding across the entire first image, by determining whether the first target point in the target coordinate list is located within the microfluidic chip channel, it can be determined whether the first target point needs to be used as a seed point for flooding treatment.

[0054] The first target point within the chip channel can be used as a seed point for flood filling. Using the chip channel's outline as a boundary, coordinate points are set in the middle of the channel, and flooding is performed to obtain an image mask of the chip channel in the microfluidic chip image. When using the first target point within the chip channel as a seed point for flood filling to obtain the image mask of the channel image in the microfluidic chip image, the process continuously checks whether the seed point has undergone flood filling. After flood filling, the channel outline in the first image, including the channel outline itself, maintains the same pixel value, which is the target pixel value of the channel outline portion set during binarization. When the target pixel value is set to 255, the grayscale value of the seed point after flood filling is also 255. This image mask is then used to crop the originally directly obtained microfluidic chip image to obtain a target image containing only the microfluidic chip channels.

[0055] Further, in one embodiment, step S20 includes:

[0056] The MSRCP algorithm, developed based on the Retine microfluidic chip channel recognition algorithm, is used to enhance the microfluidic chip image to obtain a second image.

[0057] The second image is binarized based on the adaptive equalization algorithm to obtain the third image. During the binarization process, the gray value of the microfluidic chip channel contour part in the microfluidic chip image is set to the target pixel value.

[0058] The first image is obtained by removing background impurities from the third image based on the small target removal algorithm. In the background impurity removal process, connected regions with an overall pixel value not greater than a preset threshold are removed.

[0059] In this embodiment, when preprocessing the microfluidic chip image, including image enhancement, image binarization, and small target removal, common image enhancement algorithms such as histogram equalization, gray-world algorithm, Retinex algorithm, automatic white balance, and automatic color equalization are compared. The MSRCP algorithm, developed based on the Retinex algorithm, provides the most significant brightness improvement to the chip channel contour (shadow area) in the microfluidic chip image. Therefore, the MSRCP algorithm, developed based on the Retinex microfluidic chip channel recognition algorithm, is used first to enhance the microfluidic chip image, resulting in a second image. After image enhancement, the adaptive equalization algorithm in OpenCV can be directly used to binarize the enhanced microfluidic chip image. In the binarized microfluidic chip image, the gray value of the chip channel contour can be 255, and the rest can be 0. At this time, the target pixel value can be 255. After binarization, all white parts other than the channel parts are regarded as background impurities, and the gray value corresponding to the white parts (including the channel contour and background impurities) is 255. To address the background impurities that appear after binarization, this embodiment directly uses the small target removal algorithm in OpenCV. A function is used to retain regions where the total number of pixel values ​​in a connected region is greater than a preset threshold. Through experimentation, the preset threshold (corresponding to the minimum pixel value in the connected region) is preferably set to 4000. By using the function to retain regions where the total number of pixel values ​​in a connected region is greater than 4000, and removing surrounding isolated pixels, the goal of making the channel contour clearer and more accurate is achieved.

[0060] Further, in one embodiment, step S40 includes:

[0061] Check all first target points in the target coordinate list. The check includes emitting line segments with a length not greater than the minimum length and width of the sliding window in the four directions of up, down, left and right, centered on the first target point, and determining the intersection points of the line segments with the microfluidic chip channel contour in the first image.

[0062] If line segments in the up-down or left-right directions intersect simultaneously, and the absolute value of the difference between the number of intersections on both sides of the line segments in the up-down or left-right directions is less than a preset difference, then the first target point is determined to be within the microfluidic chip channel.

[0063] In this embodiment, to determine whether all first target points in the target coordinate list are within the microfluidic chip channel, all first target points in the target coordinate list are checked. Using the first target point as the center, line segments with lengths no greater than the minimum length and width of the window are emitted in four directions: up, down, left, and right. The intersection points of these line segments with the microfluidic chip channel outline in the first image are determined to confirm whether the first target point is within the channel. Only when the line segments in the up / down or left / right directions simultaneously intersect, and the absolute value of the difference between the number of intersection points on the up / down or left / right sides is less than a preset difference, can the first target point be determined to be within the microfluidic chip channel outline. All other first target points are outside the microfluidic chip channel outline. The preset difference is preferably set to 3. Through this check, seed points not within the microfluidic chip channel outline can be eliminated. The image mask obtained after flooding processing based on the filtered first target points can more accurately distinguish between the microfluidic chip channel portion and other parts outside the channel. Furthermore, this filtering method does not require additional image processing, consumes fewer computational resources, and has a faster filtering speed.

[0064] Furthermore, in one embodiment, the preset size is set to one-sixteenth of the size of the first image.

[0065] In this embodiment, when using a sliding window to process the first image, the preset size of the sliding window is set to 1 / 16 of the size of the first image. This is mainly to account for the possibility that the size of the first image being processed may vary. In the experiment, the image size was mainly 2048x3096, but it could also be 1024x72048. To accommodate image processing of different sizes, it is set to 1 / 16 of the image size. In addition, the sliding step size of the sliding window is preferably set to 50px, i.e., 50 pixel values. The window slides from left to right and from top to bottom. When the moving distance is less than 50 pixels, it will directly move to the end position, thereby completing the sliding processing of the entire first image.

[0066] This application also provides a microfluidic chip channel identification device.

[0067] Reference Figure 3 A schematic diagram of the functional modules of the first embodiment of the microfluidic chip channel identification device.

[0068] In this embodiment, the microfluidic chip channel identification device includes:

[0069] The acquisition module 10 is used to acquire images of the microfluidic chip collected by the imaging sensor;

[0070] The image preprocessing module 20 is used to preprocess the microfluidic chip image to obtain a first image, wherein the preprocessing includes image enhancement, image binarization, and small target removal;

[0071] The sliding window processing module 30 is used to slide a sliding window of a preset size on the first image and save the center coordinates of the sliding window that meets the preset conditions as the first target point to the target coordinate list. The preset conditions are the coordinates of the target pixel value appearing in the sliding window.

[0072] The judgment module 40 is used to determine whether all the first target points in the target coordinate list are within the microfluidic chip channel;

[0073] The flood filling processing module is used to perform flood filling processing on the first target point in the chip channel as the seed point to obtain the image mask of the channel image in the microfluidic chip image. The gray value of the seed point after flood filling processing is the target pixel value.

[0074] The cropping module 50 is used to crop the microfluidic chip image using the image mask to obtain a target image that contains only the microfluidic chip channels.

[0075] Furthermore, in one embodiment, the image preprocessing module 20 is specifically used for:

[0076] The MSRCP algorithm, developed based on the Retine microfluidic chip channel recognition algorithm, is used to enhance the microfluidic chip image to obtain a second image.

[0077] The second image is binarized based on the adaptive equalization algorithm to obtain the third image. During the binarization process, the gray value of the microfluidic chip channel contour part in the microfluidic chip image is set to the target pixel value.

[0078] The first image is obtained by removing background impurities from the third image based on the small target removal algorithm. In the background impurity removal process, connected regions with an overall pixel value not greater than a preset threshold are removed.

[0079] Furthermore, in one embodiment, the determining module 40 is specifically used for:

[0080] Check all first target points in the target coordinate list. The check includes emitting line segments with a length not greater than the minimum length and width of the sliding window in the four directions of up, down, left and right, centered on the first target point, and determining the intersection points of the line segments with the microfluidic chip channel contour in the first image.

[0081] If line segments in the up-down or left-right directions intersect simultaneously, and the absolute value of the difference between the number of intersections on both sides of the line segments in the up-down or left-right directions is less than a preset difference, then the first target point is determined to be within the microfluidic chip channel.

[0082] Furthermore, in one embodiment, the preset size is set to one-sixteenth of the size of the first image.

[0083] The functions of each module in the microfluidic chip channel identification device correspond to the steps in the embodiment of the microfluidic chip channel identification method, and their functions and implementation processes will not be described in detail here.

[0084] This application provides a microfluidic chip channel identification device, which can be a personal computer (PC), laptop computer, server or other device with data processing capabilities.

[0085] Reference Figure 4 , Figure 4 This is a schematic diagram of the hardware structure of the microfluidic chip channel identification device involved in the embodiments of this application. In this embodiment, the microfluidic chip channel identification device may include a processor 1001 (e.g., a central processing unit, CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to realize the connection and communication between these components; the user interface 1003 may include a display screen or an input unit such as a keyboard; the network interface 1004 may optionally include a standard wired interface or a wireless interface (e.g., Wireless Fidelity, Wi-Fi interface); the memory 1005 may be high-speed random access memory (RAM) or stable memory (non-volatile memory), such as a disk storage device, and the memory 1005 may also optionally be a storage device independent of the aforementioned processor 1001. Those skilled in the art will understand that the hardware structure shown in the microfluidic chip channel identification diagram does not constitute a limitation of the present invention, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0086] Continue to refer to Figure 4 , Figure 4 The memory 1005, which serves as a computer-readable storage medium, may include an operating system, a network communication module, a user interface module, and a microfluidic chip channel identification program. The processor 1001 can call the microfluidic chip channel identification program stored in the memory 1005 and execute the steps of the microfluidic chip channel identification method provided in this embodiment.

[0087] The method implemented when the microfluidic chip channel identification program is executed can be referred to in various embodiments of the microfluidic chip channel identification method of this application, and will not be repeated here.

[0088] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0089] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A method for identifying channels in a microfluidic chip, characterized in that, The method includes: Acquire images of the microfluidic chip captured by the imaging sensor; The microfluidic chip image is preprocessed to obtain a first image, wherein the preprocessing includes image enhancement, image binarization, and small target removal; Slide a sliding window of a preset size on the first image, and save the center coordinates of the sliding window that meets the preset conditions as the first target point to the target coordinate list. The preset conditions are the coordinates of the target pixel value appearing in the sliding window. Determine whether all first target points in the target coordinate list are within the microfluidic chip channel; The first target point within the chip channel is used as the seed point for flood filling processing to obtain the image mask of the channel image in the microfluidic chip image. The gray value of the seed point after flood filling processing is the target pixel value. The microfluidic chip image is cropped using the image mask to obtain a target image containing only the microfluidic chip channels; The step of preprocessing the microfluidic chip image to obtain the first image includes: The MSRCP algorithm, developed based on the Retinex algorithm, is used to enhance images of microfluidic chips to obtain a second image. The second image is binarized based on the adaptive equalization algorithm to obtain the third image. During the binarization process, the gray value of the microfluidic chip channel contour part in the microfluidic chip image is set to the target pixel value. The first image is obtained by removing background impurities from the third image based on the small target removal algorithm. In the background impurity removal process, connected regions with an overall pixel value not greater than a preset threshold are removed.

2. The method according to claim 1, characterized in that, The step of determining whether all first target points in the target coordinate list are within the microfluidic chip channel includes: Check all first target points in the target coordinate list. The check includes emitting line segments with a length not greater than the minimum length and width of the sliding window in the four directions of up, down, left and right, centered on the first target point, and determining the intersection points of the line segments with the microfluidic chip channel contour in the first image. If line segments in the up-down or left-right directions intersect simultaneously, and the absolute value of the difference between the number of intersections on both sides of the line segments in the up-down or left-right directions is less than a preset difference, then the first target point is determined to be within the microfluidic chip channel.

3. The method according to claim 1, characterized in that: The preset size is set to 1 / 16th of the size of the first image.

4. A microfluidic chip channel identification device, characterized in that, The device includes: The acquisition module is used to acquire images of the microfluidic chip collected by the imaging sensor; An image preprocessing module is used to preprocess the microfluidic chip image to obtain a first image, wherein the preprocessing includes image enhancement, image binarization, and small target removal; The sliding window processing module is used to slide a sliding window of a preset size on the first image and save the center coordinates of the sliding window that meets the preset conditions as the first target point to the target coordinate list. The preset conditions are the coordinates of the target pixel value appearing in the sliding window. The judgment module is used to determine whether all the first target points in the target coordinate list are within the microfluidic chip channel; The flood filling processing module is used to perform flood filling processing on the first target point in the chip channel as the seed point to obtain the image mask of the channel image in the microfluidic chip image. The gray value of the seed point after flood filling processing is the target pixel value. The cropping module is used to crop the microfluidic chip image using the image mask to obtain a target image that contains only the microfluidic chip channels; The image preprocessing module is further used for: The MSRCP algorithm, developed based on the Retine microfluidic chip channel recognition algorithm, is used to enhance the microfluidic chip image to obtain a second image. The second image is binarized based on the adaptive equalization algorithm to obtain the third image. During the binarization process, the gray value of the microfluidic chip channel contour part in the microfluidic chip image is set to the target pixel value. The first image is obtained by removing background impurities from the third image based on the small target removal algorithm. In the background impurity removal process, connected regions with an overall pixel value not greater than a preset threshold are removed.

5. The apparatus according to claim 4, characterized in that, The judgment module is specifically used for: Check all first target points in the target coordinate list. The check includes emitting line segments with a length not greater than the minimum length and width of the sliding window in the four directions of up, down, left and right, centered on the first target point, and determining the intersection points of the line segments with the microfluidic chip channel contour in the first image. If line segments in the up-down or left-right directions intersect simultaneously, and the absolute value of the difference between the number of intersections on both sides of the line segments in the up-down or left-right directions is less than a preset difference, then the first target point is determined to be within the microfluidic chip channel.

6. The apparatus according to claim 4, characterized in that: The preset size is set to 1 / 16th of the size of the first image.

7. A microfluidic chip channel identification device, characterized in that, The microfluidic chip channel identification device includes a processor, a memory, and a microfluidic chip channel identification program stored in the memory and executable by the processor, wherein when the microfluidic chip channel identification program is executed by the processor, it implements the steps of the microfluidic chip channel identification method as described in any one of claims 1 to 3.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a microfluidic chip channel identification program, wherein when the microfluidic chip channel identification program is executed by a processor, it implements the steps of the microfluidic chip channel identification method as described in any one of claims 1 to 3.

Images