Nematode survival rate identification method based on the combination of bright field and dark field images
By combining background brightness estimation segmentation algorithms with bright field and dark field images, nematode survival rate is identified, and the accuracy of nematode death recognition in high-throughput screening is solved, and the automation and efficiency of experiments are improved.
Patent Information
- Application Number
- CN202111280952.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2021-11-01
- Publication Date
- 2025-01-17
- Estimated Expiration
- 2041-11-01
AI Technical Summary
In high-throughput screening experiments, automatic and efficient identification of whether nematodes die is a technical difficulty, and the existing methods have the problem that live nematodes are misjudged as dead.
Using a combination of bright field and dark field images, the bright field and dark field images are processed separately through the image segmentation algorithm of background brightness estimation, and the total number of nematodes and the number of dead nematodes are identified, and the bright field and dark field images are compared. Determine the survival rate of nematodes.
Accurate identification of nematode survival rates is achieved, the problem of living nematodes being misjudged as dead in traditional methods is overcome, and the accuracy and efficiency of identification are improved.
Smart Images
Figure CN114004846B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of biological image processing, in particular to a nematode survival rate recognition method based on the combination of bright field and dark field images. Background Art
[0002] Caenorhabditis elegans, also known as nematodes, is a widely used model animal. It is small in size, about 1 mm long, and easy to culture. The average lifespan of ordinary Caenorhabditis elegans in a laboratory culture environment at 20°C is about two to three weeks, and the development time is about three days. In 1974, Brenner chose to use Caenorhabditis elegans as a model organism to study developmental and neuroscience issues. Image acquisition and image processing are used in many high-throughput screening studies using nematodes. The image data produced in high-throughput screening experiments far exceeds the ability of manual inspection and analysis, requiring researchers to use image processing tools. The purpose of image processing in biological experiments is to use advanced computer science and analysis methods to process large-scale biological image data to help solve biological problems.
[0003] The survival rate of nematodes is the most commonly used quantitative indicator in various experiments. However, in the current high-throughput compound screening experiments using nematodes, it is a very important technical difficulty to automatically and efficiently determine whether the nematodes are dead from the experimental images. Summary of the invention
[0004] The technical task of the present invention is to provide a nematode survival rate identification method based on the combination of bright field and dark field images to solve the problem of how to accurately identify whether the nematodes are dead from experimental images and accurately calculate the nematode survival rate.
[0005] The technical task of the present invention is achieved in the following way: a method for identifying the survival rate of nematodes based on the combination of bright field and dark field images, wherein the method is to place the nematodes in a 384-well plate, add PI dye solution to the 384-well plate after a preset time interval, and take bright field images and dark field images of the nematodes respectively; the bright field image and the dark field image are segmented respectively by an image segmentation algorithm based on background brightness estimation, and then the segmented bright field image and dark field image are combined to identify the total number of nematodes in the 384-well plate, the number of dead nematodes and the number of surviving nematodes, and then calculate the nematode survival rate.
[0006] Preferably, the method is as follows:
[0007] S1. Design the experimental environment of different chemicals in 384-well plates;
[0008] S2. Move adult nematodes to each well of a 384-well plate, with 10-20 nematodes per well;
[0009] S3, after an interval of 24 hours, PI dye solution was added and bright field images and dark field images were taken respectively;
[0010] S4, performing image segmentation and nematode identification on the captured bright field image;
[0011] S5, performing image segmentation and fluorescent nematode identification on the captured dark field image;
[0012] S6, comparing the nematodes identified in the bright field image with the fluorescent nematodes identified in the dark field image, so as to determine whether each nematode identified in the bright field image is dead;
[0013] S7. Obtain the number of dead nematodes through step S6, and then calculate the nematode survival rate.
[0014] More preferably, the PI dye solution is added into the nematode system at 1% volume, and after the PI dye solution is evenly dispersed in the system, bright field images and dark field images are immediately captured; wherein, the ambient temperature is 20-25° C., and the concentration of the PI dye solution is 1 mg / ml.
[0015] Preferably, the image segmentation and nematode identification of the bright field image captured in step S4 are specifically as follows:
[0016] S401, estimating the background brightness of different areas in the bright field image;
[0017] S402, background brightness correction: according to the background estimation value of the area where each pixel is located, the original gray value of each pixel is subtracted from the corresponding background estimation value to obtain the background brightness correction result;
[0018] S403, segmenting the image using a threshold segmentation method according to the result image after background correction;
[0019] S404, identifying nematodes in the bright field image: distinguishing the nematodes from the remaining areas by image segmentation; wherein the nematodes in the bright field image are white, and the remaining areas are all black.
[0020] Preferably, the background brightness of different areas in the image is estimated in step S401 as follows:
[0021] S40101. The grayscale value of each pixel in the bright field image is 0-255. The 256 grayscale intensity levels are divided into 16 levels (specifically: 0-15 is the first level, 16-31 is the second level, and so on until 240-255 is the 16th level);
[0022] S40102. Calculate the background brightness of different areas: Search for pixels with the same grayscale level in the bright field image, calculate the smallest convex polygon that can enclose these pixels, and set the background brightness within the convex polygon to the lowest brightness in the current level as the background estimate of the area where each pixel at that level is located.
[0023] Preferably, the image segmentation and fluorescent nematode identification of the captured dark field image in step S5 are specifically as follows:
[0024] S501, estimating the background brightness of different areas in the dark field image;
[0025] S502, background brightness correction: according to the background estimation value of the area where each pixel is located, the original gray value of each pixel is subtracted from the corresponding background estimation value to obtain the background brightness correction result;
[0026] S503, segmenting the image using a threshold segmentation method according to the result image after background correction;
[0027] S504, identifying nematodes in the dark field image: distinguishing the fluorescent nematodes from the remaining areas in the dark field image by image segmentation; wherein the white area in the dark field image is the identified area with high fluorescent brightness, wherein the location of the identified fluorescent nematodes, i.e., the location of the dead nematodes, is the location of the connected area close to the volume of the nematodes; and the remaining areas are black.
[0028] Preferably, in step S501, the background brightness of different areas in the dark field image is estimated as follows:
[0029] S50101. The grayscale value of each pixel in the dark field image is 0-255. The 256 grayscale intensity levels are divided into 16 levels (specifically: 0-15 is the first level, 16-31 is the second level, and so on until 240-255 is the 16th level);
[0030] S50102. Calculate the background brightness of different areas: Search for pixels with the same grayscale level in the dark field image, calculate the smallest convex polygon that can enclose these pixels, and set the background brightness within the convex polygon to the lowest brightness in the current level as the background estimation value of the area where each pixel at this level is located.
[0031] Preferably, the step S6 of determining whether each nematode identified in the bright field image is dead is specifically as follows:
[0032] After the bright field image and the dark field image are segmented respectively, all the nematodes in the hole are identified in the bright field image, and each nematode is compared with the corresponding position in the fluorescent image. If the position of a nematode coincides with that of the fluorescent nematode in the fluorescent image, it means that the nematode is dead.
[0033] Preferably, in the process of comparing the nematodes identified in the bright field image with the fluorescent nematodes identified in the dark field image in step S6, if there are multiple nematodes crossed together, the situation is as follows:
[0034] ① When all the crossed nematodes are alive: In the bright field image, after image segmentation, when multiple nematodes are crossed together, the number of the currently crossed nematodes A3 is calculated, and the formula is as follows:
[0035]
[0036] Wherein, A1 represents the total area of the crossed nematodes; A2 represents the area of a single nematode in the bright field image;
[0037] ② When all the crossed nematodes are dead nematodes: Dead nematodes have fluorescence in the dark field image. Therefore, when the bright field image and the dark field image are overlapped, the crossed dead nematodes will overlap with the fluorescent nematodes in the dark field, and the number of these nematodes will not be counted in the total number of live nematodes.
[0038] ③. When the crossed nematodes include both live and dead nematodes: first calculate the number of crossed nematodes A3 through the calculation process in case ①; then perform image segmentation in the dark field image to identify the total area A4 of the fluorescent nematodes in the current crossed area of the nematodes. The formula for the number A5 of fluorescent nematodes is as follows:
[0039]
[0040] The number of live nematodes in the nematode crossing area is A3-A5.
[0041] Preferably, the formula for calculating the nematode survival rate in step S7 is as follows:
[0042]
[0043] Among them, N1 represents the total number of nematodes in the hole; N2 represents the number of living nematodes in the hole; P represents the survival rate after living in the current chemical environment for 24 hours.
[0044] The nematode survival rate identification method based on the combination of bright field and dark field images of the present invention has the following advantages:
[0045] (i) Both bright-field and dark-field images were used to accurately identify the total number of nematodes in the experiment, the number of dead nematodes, and the number of live nematodes; this method accurately identified all nematodes through bright-field image segmentation, and then combined with dark-field images to more accurately determine whether each nematode was dead;
[0046] (ii) The traditional method for automatically distinguishing whether a nematode is dead is based on whether the nematode is active. The obvious defect of this method is that if the nematode happens not to be active during the shooting process, then the living nematode will also be judged as dead. The present invention overcomes this defect by designing an image segmentation algorithm based on background brightness estimation, performing graphic segmentation on the bright field image after adding PI dye and the fluorescent image under dark field, so as to accurately identify whether the nematode is dead; only dead nematodes will emit strong fluorescence after adding PI dye. By designing an image segmentation algorithm, the bright field image and the dark field image are segmented separately, and the bright field image and the dark field image are combined to accurately identify the total number of nematodes in the experiment, the number of dead nematodes and the number of living nematodes and calculate the nematode survival rate;
[0047] (iii) After image segmentation in the bright field image, all nematodes can be identified, but it is impossible to distinguish between the current live and dead nematodes; the present invention combines the bright field image and the dark field fluorescent image to identify the life and death of the nematodes; an image segmentation algorithm based on background brightness estimation is designed to segment the bright field image and the dark field image respectively, so as to identify all nematodes in the hole in the bright field image, and compare each nematode with the corresponding position in the fluorescent image. If it coincides with the fluorescent nematode in the fluorescent image, it means that the nematode is dead, so as to accurately judge whether the nematode is dead, and then accurately calculate the nematode survival rate. BRIEF DESCRIPTION OF THE DRAWINGS
[0048] The present invention is further described below in conjunction with the accompanying drawings.
[0049] Attached Figure 1 It is a flowchart of the nematode survival rate identification method based on the combination of bright field and dark field images;
[0050] Attached Figure 2 This is a bright field image taken after adding PI dye solution;
[0051] Attached Figure 3 This is a dark field image taken after adding PI dye solution;
[0052] Attached Figure 4 The nematode image is identified after image segmentation of the bright field image taken after adding PI dye solution;
[0053] Attached Figure 5 The image of fluorescent nematodes was identified after image segmentation of the dark field image taken after adding PI dye solution. DETAILED DESCRIPTION
[0054] The nematode survival rate identification method based on the combination of bright field and dark field images of the present invention is described in detail below with reference to the accompanying drawings and specific embodiments of the specification.
[0055] Embodiment 1:
[0056] As attached Figure 1 As shown, the present invention is a method for identifying the survival rate of nematodes based on the combination of bright field and dark field images. The method is to place the nematodes in a 384-well plate, add PI dye solution to the 384-well plate after a preset time interval, and take bright field images and dark field images of the nematodes respectively; the bright field image and the dark field image are respectively segmented by an image segmentation algorithm based on background brightness estimation, and then the segmented bright field image and dark field image are combined to identify the total number of nematodes in the 384-well plate, the number of dead nematodes, and the number of surviving nematodes, and then calculate the survival rate of the nematodes; the details are as follows:
[0057] S1. Design experimental environments for different chemicals in a 384-well plate. Multiple groups of different chemical experiments can be designed in the same 384-well plate. For example, there are 384 independent wells on a 384-well plate, and 32 wells can be selected for each chemical experiment. Twelve groups of different chemical experiments can be performed simultaneously on a 384-well plate; chemicals such as saffron, turmeric, cadmium chloride, etc.
[0058] S2. Move adult nematodes to each well of a 384-well plate, with 15 nematodes per well;
[0059] S3, after an interval of 24 hours, PI dye solution was added and bright field images and dark field images were taken respectively;
[0060] S4, performing image segmentation and nematode identification on the captured bright field image;
[0061] S5, performing image segmentation and fluorescent nematode identification on the captured dark field image;
[0062] S6, comparing the nematodes identified in the bright field image with the fluorescent nematodes identified in the dark field image, so as to determine whether each nematode identified in the bright field image is dead;
[0063] S7. Obtain the number of dead nematodes through step S6, and then calculate the nematode survival rate.
[0064] The PI dye solution in this example was added into the nematode system at 1% volume, and after the PI dye solution was evenly dispersed in the system, bright field images and dark field images were immediately taken; wherein, the ambient temperature was 20-25° C., and the concentration of the PI dye solution was 1 mg / ml.
[0065] The bright field image obtained after adding PI dye is shown in the attached figure. Figure 2 As shown in the figure, all nematodes in the hole can be photographed. The dark field image obtained after adding PI dye is shown in the figure below. Figure 3 As shown, after PI staining, the fluorescence brightness of the dead nematodes can be clearly seen under dark field conditions, while the living nematodes are almost invisible under dark field conditions. By combining the two images, it is possible to accurately determine whether each nematode identified in the bright field image is dead.
[0066] In this embodiment, the bright field image segmentation is as follows: for each well of the 384-well plate, the bright field image is as follows: Figure 2 As shown in the figure, this image has uneven brightness (the background gets darker from the center to the periphery). This image with uneven background brightness cannot be directly segmented by image segmentation methods such as threshold segmentation. To this end, we designed an image segmentation algorithm based on background brightness estimation to accurately segment this image with uneven background brightness.
[0067] In step S4 of this embodiment, the image segmentation and nematode identification of the captured bright field image are performed as follows:
[0068] S401, estimating the background brightness of different areas in the bright field image; the details are as follows:
[0069] S40101. The grayscale value of each pixel in the bright field image is 0-255. The 256 grayscale intensity levels are divided into 16 levels; specifically, 0-15 is the first level, 16-31 is the second level, and so on until 240-255 is the 16th level;
[0070] S40102, calculate the background brightness of different areas: search for pixels with the same grayscale level in the bright field image, calculate the smallest convex polygon that can enclose these pixels, and set the background brightness within the convex polygon to the lowest brightness in the current level as the background estimation value of the area where each pixel of this level is located. Specifically:
[0071] The process of calculating the background brightness of different areas is as follows: search for pixels with grayscale level 1 (i.e. grayscale values between 0 and 15) in the image (these darkest pixels are located in the outer part of the image); after finding these pixels, calculate the smallest convex polygon that can enclose these pixels, and set the background brightness within this convex polygon to the lowest brightness in the current level (0); next, search for pixels with grayscale level 2 (i.e. grayscale values between 16 and 31) in the image (the area where these pixels are located is smaller than the range of pixels in level 1). ); after finding these pixels, calculate the smallest convex polygon that can enclose these pixels, and set the background brightness within this convex polygon to the lowest brightness in the current level (16); the same goes for subsequent levels; finally, search for pixels with a grayscale level of level 16 (i.e., grayscale values between 240 and 255) in the image (these brightest pixels are concentrated in the brightest area in the middle); after finding these pixels, calculate the smallest convex polygon that can enclose these pixels, and set the background brightness within this convex polygon to the lowest brightness in the current level (240).
[0072] Dark field image segmentation in this embodiment: The dark field image is as shown in the attached Figure 3 As shown in the figure, in the dark field image, only the dead nematodes have high-brightness fluorescence. There is also the problem of uneven background brightness in the dark field image. The segmentation method for this image is the same as the processing method for the bright field image above.
[0073] S402, background brightness correction: according to the background estimation value of the area where each pixel is located, the original gray value of each pixel is subtracted from the corresponding background estimation value to obtain the background brightness correction result;
[0074] S403, segmenting the image using a threshold segmentation method according to the result image after background correction;
[0075] S404, identifying nematodes in the bright field image: distinguishing the nematodes from the remaining areas by image segmentation; wherein the nematodes in the bright field image are white, and the remaining areas are all black; specifically:
[0076] The results of image segmentation and nematode identification of the bright field image obtained after adding PI dye are shown in the attached figure. Figure 4 As shown in the figure, the nematodes are distinguished from other areas through image segmentation. In the figure, the nematodes are white and the other areas are all black. Among these identified nematodes, there are living nematodes and dead nematodes.
[0077] In step S401 of this embodiment, the background brightness of different areas in the image is estimated.
[0078] In step S5 of this embodiment, the image segmentation and fluorescent nematode identification of the captured dark field image are performed as follows:
[0079] S501, estimating the background brightness of different areas in the dark field image; the details are as follows:
[0080] S50101. The grayscale value of each pixel in the dark field image is 0-255. The 256 grayscale intensity levels are divided into 16 levels; specifically, 0-15 is the first level, 16-31 is the second level, and so on until 240-255 is the 16th level.
[0081] S50102, calculate the background brightness of different areas: search for pixels with the same grayscale level in the dark field image, calculate the minimum convex polygon that can enclose these pixels, and set the background brightness within the convex polygon to the minimum brightness in the current level as the background estimation value of the area where each pixel of the level is located; specifically:
[0082] The process of calculating the background brightness of different areas is as follows: search for pixels with grayscale level 1 (i.e. grayscale values between 0 and 15) in the image (these darkest pixels are located in the outer part of the image); after finding these pixels, calculate the smallest convex polygon that can enclose these pixels, and set the background brightness within this convex polygon to the lowest brightness in the current level (0); next, search for pixels with grayscale level 2 (i.e. grayscale values between 16 and 31) in the image (the area where these pixels are located is smaller than the range of pixels in level 1). ); after finding these pixels, calculate the smallest convex polygon that can enclose these pixels, and set the background brightness within this convex polygon to the lowest brightness in the current level (16); the same goes for subsequent levels; finally, search for pixels with a grayscale level of level 16 (i.e., grayscale values between 240 and 255) in the image (these brightest pixels are concentrated in the brightest area in the middle); after finding these pixels, calculate the smallest convex polygon that can enclose these pixels, and set the background brightness within this convex polygon to the lowest brightness in the current level (240).
[0083] S502, background brightness correction: according to the background estimation value of the area where each pixel is located, the original gray value of each pixel is subtracted from the corresponding background estimation value to obtain the background brightness correction result;
[0084] S503, segmenting the image using a threshold segmentation method according to the result image after background correction;
[0085] S504, identifying nematodes in the dark field image: distinguishing the fluorescent nematodes from the remaining areas in the dark field image by image segmentation; wherein the white area in the dark field image is the identified area with high fluorescent brightness, wherein the location of the identified fluorescent nematodes, i.e., the location of the dead nematodes, is the location of the connected area close to the volume of the nematodes; and the remaining areas are black.
[0086] The dark field images obtained after adding PI dye were segmented and the fluorescent nematode identification results are shown in the attached figure. Figure 5 As shown in the figure, the fluorescent nematodes in the dark field image are distinguished from other areas through image segmentation. In the figure, the white area is the area with high fluorescence brightness, and the area with a connected area close to the volume of the nematode is the location of the identified fluorescent nematode; the other areas are all black. Only dead nematodes have high brightness fluorescence intensity. The fluorescence of live nematodes is very weak, and the location of dead nematodes can be identified by the fluorescence intensity in the dark field image.
[0087] In step S6 of this embodiment, determining whether each nematode identified in the bright field image is dead is specifically as follows:
[0088] After the bright field image and the dark field image are segmented respectively, all the nematodes in the hole are identified in the bright field image, and each nematode is compared with the corresponding position in the fluorescent image. If the position of a nematode coincides with that of the fluorescent nematode in the fluorescent image, it means that the nematode is dead.
[0089] In step S6 of this embodiment, in the process of comparing the nematodes identified in the bright field image with the fluorescent nematodes identified in the dark field image, if multiple nematodes are crossed together, the situation is as follows:
[0090] ① When all the crossed nematodes are alive: In the bright field image, after image segmentation, when multiple nematodes are crossed together, the number of the currently crossed nematodes A3 is calculated, and the formula is as follows:
[0091]
[0092] Where A1 represents the total area of the crossed nematodes; A2 represents the area of a single nematode in the bright field image; because the crossed nematodes will reduce the total area due to overlap, the number of nematodes is rounded up when calculating the number of nematodes;
[0093] ② When all the crossed nematodes are dead nematodes: Dead nematodes have fluorescence in the dark field image. Therefore, when the bright field image and the dark field image are overlapped, the crossed dead nematodes will overlap with the fluorescent nematodes in the dark field, and the number of these nematodes will not be counted in the total number of live nematodes.
[0094] ③. When the crossed nematodes include both live and dead nematodes: first calculate the number of crossed nematodes A3 through the calculation process in case ①; then perform image segmentation in the dark field image to identify the total area A4 of the fluorescent nematodes in the current crossed area of the nematodes. The formula for the number A5 of fluorescent nematodes is as follows:
[0095]
[0096] The number of live nematodes in the nematode crossing area is A3-A5.
[0097] Calculation of nematode survival rate: Figure 4 Calculate the total number of nematodes (set as N1). Then, Figure 4 and 5 to compare; that is, Figure 4 Each nematode and its Figure 5 Compare the corresponding positions in Figure 4 A nematode and its Figure 5 If the fluorescent nematode in the attached Figure 4 A nematode and its Figure 5 If the fluorescent nematodes in the well do not overlap, the nematode is defined as a live nematode. Finally, the number of live nematodes in the current well is obtained by subtracting the number of dead nematodes from the value of N1 (set as N2).
[0098] The formula for calculating the nematode survival rate in step S7 of this embodiment is as follows:
[0099]
[0100] Among them, N1 represents the total number of nematodes in the hole; N2 represents the number of living nematodes in the hole; P represents the survival rate after living in the current chemical environment for 24 hours.
[0101] An experimental environment of three different chemicals (saffron, turmeric and chromium chloride) was designed in a 384-well plate, and nematodes were placed in different chemicals for 24 hours of culture. The survival rate of the nematodes was identified by the traditional method and the identification method of the present invention, as shown in the following table:
[0102]
[0103] It can be seen from the table that the recognition result of the survival rate of the present invention is close to the actual survival rate result, which proves that the image segmentation and recognition method adopted by the present invention greatly improves the accuracy of nematode survival rate recognition.
[0104] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.
Claims
1. A method for identifying nematode survival rate based on a combination of bright field and dark field images, characterized in that: The method is to place the nematodes in a 384-well plate, add PI dye solution to the 384-well plate after a preset time interval, and respectively take bright field images and dark field images of the nematodes; respectively perform image segmentation on the bright field image and the dark field image by an image segmentation algorithm based on background brightness estimation, and then combine the segmented bright field image and the dark field image to identify the total number of nematodes in the 384-well plate, the number of dead nematodes, and the number of surviving nematodes, and then calculate the nematode survival rate; the details are as follows: S1. Design the experimental environment of different chemicals in 384-well plates; S2. Move adult nematodes to each well of a 384-well plate, with 10-20 nematodes per well; S3, after an interval of 24 hours, PI dye solution was added and bright field images and dark field images were taken respectively; S4. Perform image segmentation and nematode identification on the bright field image obtained by shooting; the details are as follows: S401, estimating the background brightness of different areas in the bright field image; the details are as follows: S40101. The grayscale value of each pixel in the bright field image is 0-255, and these 256 grayscale intensity levels are divided into 16 levels; S40102, calculating the background brightness of different areas: searching for pixels with the same grayscale level in the bright field image, calculating the smallest convex polygon that can enclose these pixels, and setting the background brightness within the convex polygon to the lowest brightness in the current level as the background estimation value of the area where each pixel of the level is located; S402, background brightness correction: according to the background estimation value of the area where each pixel is located, the original gray value of each pixel is subtracted from the corresponding background estimation value to obtain the background brightness correction result; S403, segmenting the image using a threshold segmentation method according to the result image after background correction; S404, identifying nematodes in the bright field image: distinguishing the nematodes from the remaining areas by image segmentation; wherein the nematodes in the bright field image are white, and the remaining areas are all black; S5, performing image segmentation and fluorescent nematode identification on the captured dark field image; S6. Compare the nematodes identified in the bright field image with the fluorescent nematodes identified in the dark field image to determine whether each nematode identified in the bright field image is dead; wherein, during the comparison of the nematodes identified in the bright field image with the fluorescent nematodes identified in the dark field image, if multiple nematodes are crossed together, the situation is as follows: ① When all the crossed nematodes are alive: In the bright field image, after image segmentation, when multiple nematodes are crossed together, the number of the currently crossed nematodes A3 is calculated, and the formula is as follows: A3=⌈A1 / A2⌉; Wherein, A1 represents the total area of the crossed nematodes; A2 represents the area of a single nematode in the bright field image; ② When all the crossed nematodes are dead nematodes: Dead nematodes have fluorescence in the dark field image. Therefore, when the bright field image and the dark field image are overlapped, the crossed dead nematodes will overlap with the fluorescent nematodes in the dark field, and the number of these nematodes will not be counted in the total number of live nematodes. ③. When the crossed nematodes include both live and dead nematodes: first calculate the number of crossed nematodes A3 through the calculation process in case ①; then perform image segmentation in the dark field image to identify the total area A4 of the fluorescent nematodes in the current crossed area of the nematodes. The formula for the number A5 of fluorescent nematodes is as follows: A5=⌈A4 / A2⌉; The number of live nematodes in the nematode intersection area is A3-A5; S7. Obtain the number of dead nematodes through step S6 and calculate the nematode survival rate.
2. The method for identifying nematode survival rate based on the combination of bright field and dark field images according to claim 1, characterized in that: The PI dye solution was added into the nematode system at 1% volume, and after the PI dye solution was evenly dispersed in the system, bright field images and dark field images were immediately taken; wherein, the ambient temperature was 20-25° C., and the concentration of the PI dye solution was 1 mg / ml.
3. The method for identifying nematode survival rate based on the combination of bright field and dark field images according to claim 1, characterized in that: The image segmentation and fluorescent nematode identification of the captured dark field image in step S5 are specifically performed as follows: S501, estimating the background brightness of different areas in the dark field image; S502, background brightness correction: according to the background estimation value of the area where each pixel is located, the original gray value of each pixel is subtracted from the corresponding background estimation value to obtain the background brightness correction result; S503, segmenting the image using a threshold segmentation method according to the result image after background correction; S504, identifying nematodes in the dark field image: distinguishing the fluorescent nematodes from the remaining areas in the dark field image by image segmentation; wherein the white area in the dark field image is the identified area with high fluorescent brightness, wherein the location of the identified fluorescent nematodes, i.e., the location of the dead nematodes, is the location of the connected area close to the volume of the nematodes; and the remaining areas are black.
4. The method for identifying nematode survival rate based on the combination of bright field and dark field images according to claim 3, characterized in that: In step S501, the background brightness of different areas in the dark field image is estimated as follows: S50101. The grayscale value of each pixel in the dark field image is 0-255, and the 256 grayscale intensity levels are divided into 16 levels; S50102. Calculate the background brightness of different areas: Search for pixels with the same grayscale level in the dark field image, calculate the smallest convex polygon that can enclose these pixels, and set the background brightness within the convex polygon to the lowest brightness in the current level as the background estimation value of the area where each pixel at this level is located.
5. The method for identifying nematode survival rate based on the combination of bright field and dark field images according to claim 1, characterized in that: The specific method of determining whether each nematode identified in the bright field image is dead in step S6 is as follows: After the bright field image and the dark field image are segmented respectively, all the nematodes in the hole are identified in the bright field image, and each nematode is compared with the corresponding position in the fluorescent image. If the position of a nematode coincides with that of the fluorescent nematode in the fluorescent image, it means that the nematode is dead.
6. The method for identifying nematode survival rate based on the combination of bright field and dark field images according to claim 1, characterized in that: The formula for calculating the nematode survival rate in step S7 is as follows: ; Among them, N1 represents the total number of nematodes in the hole; N2 represents the number of living nematodes in the hole; P represents the survival rate after living in the current chemical environment for 24 hours.