Basic group identification method of fluorescence image, computer equipment and readable storage medium

Abstract

The invention provides a fluorescent image base recognition method, computer equipment and a computer readable storage medium. The method comprises the following steps: acquiring multiple frames of fluorescent images of which bases are to be detected; based on first intensity value information corresponding to pixel points in the multiple frames of fluorescence images, performing first segmentation processing on the multiple frames of fluorescence images to obtain candidate fluorescence light spot areas corresponding to the multiple frames of fluorescence images, and performing fusion processing on the candidate fluorescence light spot areas corresponding to the multiple frames of fluorescence images to obtain candidate fluorescence light spot areas corresponding to the multiple frames of fluorescence images; obtaining a fused fluorescent light spot region corresponding to the multiple frames of fluorescent images; performing image sharpening processing on the multiple frames of fluorescence images to obtain target fluorescence images corresponding to the multiple frames of fluorescence images, and performing second segmentation processing on the multiple frames of target fluorescence images by using the fused fluorescence light spot region to obtain target fluorescence light spot regions in the multiple frames of target fluorescence images; and identifying the base type based on the target fluorescent light spot area.

Description

Technical Field

[0001] The present disclosure relates to the field of gene sequencing technology, and in particular, to a base recognition method of a fluorescent image, a computer device, and a computer-readable storage medium. Background Art

[0002] In the field of gene sequencing technology, sequencing methods based on optical microscopic imaging are currently the mainstream sequencing methods. In sequencing methods based on optical microscopic imaging, specific fluorescent substances are attached to the gene fragments to be tested (e.g., DNA fragments) through chemical reactions, and fluorescent substances are excited by visible light (e.g., generally red light or green light) to generate fluorescent signals that can reflect the base characteristics on the gene fragments to be tested, and then an image sensor (e.g., generally an industrial camera) is used to collect the fluorescent signals and generate fluorescent images, and then base recognition software obtains the fluorescent images, extracts the grayscale value of the light spots in the fluorescent images, and identifies the base types corresponding to the light spots according to the grayscale value of the light spots, thereby identifying the base sequences corresponding to the gene fragments to be tested; wherein the base types include: adenine (A), thymine (T), cytosine (C) and guanine (G).

[0003] However, due to the limitations of optical microscopy imaging technology, the current base recognition in fluorescent images suffers from low recognition accuracy. Summary of the invention

[0004] The embodiments of the present disclosure at least provide a base recognition method for a fluorescent image, a computer device, and a computer-readable storage medium.

[0005] In a first aspect, an embodiment of the present disclosure provides a method for base recognition of a fluorescent image, the method comprising:

[0006] Acquire multiple frames of fluorescence images to be base-called;

[0007] Based on the first intensity value information respectively corresponding to the pixel points in the multiple frames of the fluorescence images, the multiple frames of the fluorescence images are subjected to a first segmentation process to obtain candidate fluorescence spot regions respectively corresponding to the multiple frames of the fluorescence images, and the candidate fluorescence spot regions respectively corresponding to the multiple frames of the fluorescence images are subjected to a fusion process to obtain a fused fluorescence spot region corresponding to the multiple frames of the fluorescence images;

[0008] Performing image sharpening processing on the multiple frames of fluorescence images respectively to obtain target fluorescence images corresponding to the multiple frames of fluorescence images respectively;

[0009] Performing a second segmentation process on the target fluorescence images of multiple frames by using the fused fluorescence spot region to obtain the target fluorescence spot regions in the target fluorescence images of multiple frames;

[0010] The base type is identified based on the target fluorescent spot area.

[0011] In this way, the candidate fluorescent spot areas corresponding to the multiple frames of fluorescent images are first used to determine the fused fluorescent spot area where the spot is more likely to exist. At the same time, the multiple frames of fluorescent images are respectively subjected to image sharpening processing to eliminate the blurred edges of the spots. Then, the fused fluorescent spot area is used to perform a second segmentation processing on the target fluorescent image with the blurred edges eliminated, so that the target fluorescent spot area determined for each target fluorescent image has higher accuracy, thereby improving the base recognition accuracy of the fluorescent image.

[0012] In an optional embodiment, the step of acquiring multiple frames of fluorescent images to be base called includes:

[0013] Image noise reduction processing is performed on multiple frames of original fluorescent images to obtain fluorescent images to be base-detected corresponding to the multiple frames of original fluorescent images.

[0014] In an optional embodiment, the first segmentation processing is performed on the multiple frames of the fluorescence images based on the first intensity value information corresponding to the pixel points in the multiple frames of the fluorescence images to obtain the candidate fluorescence spot areas corresponding to the multiple frames of the fluorescence images, including:

[0015] Determine, from a plurality of pixels in a plurality of frames of the fluorescent image, candidate pixels whose first intensity value information belongs to a first intensity threshold interval;

[0016] Based on the position information of the candidate pixel points in the multiple frames of the fluorescence images, the candidate fluorescence spot areas in the multiple frames of the fluorescence images are determined.

[0017] In an optional embodiment, the performing image sharpening processing on the multiple frames of the fluorescence images to obtain target fluorescence images corresponding to the multiple frames of the fluorescence images respectively includes:

[0018] Using a sharpening template of the target fluorescent image, convolution processing is performed on pixel points in multiple frames of the fluorescent image to obtain second intensity value information corresponding to the pixel points of the multiple frames of the fluorescent image;

[0019] Based on the second intensity value information corresponding to the pixel points in the multiple frames of the fluorescent images, target fluorescent images corresponding to the multiple frames of the fluorescent images are generated.

[0020] In an optional implementation manner, the sharpening template of the target fluorescence image is obtained by smoothing the weight coefficient of the image sharpening template using a Laplacian operator;

[0021] The image sharpening template is constructed based on the Riemann-Liouville RL fractional-order differential operator.

[0022] In an optional implementation, the performing second segmentation processing on multiple frames of the target fluorescence images using the fused fluorescence spot area includes:

[0023] Determining a target area corresponding to the fused fluorescent spot area from multiple frames of the target fluorescent image;

[0024] According to the second intensity value information corresponding to the pixel points in the target area, the target area is subjected to a second segmentation process to obtain the target fluorescent spot area.

[0025] In an optional embodiment, after identifying the base type based on the target fluorescent spot area, the method further includes:

[0026] Comparing the base type with the target base type to obtain a comparison result;

[0027] Based on the comparison result, determining whether the base recognition accuracy of the target fluorescent spot area meets the target accuracy condition;

[0028] When the base recognition accuracy of the target fluorescent spot area does not meet the target accuracy condition, the area of ​​the fused fluorescent spot area is adjusted, and the part of the target fluorescent image belonging to the fused fluorescent spot area is re-segmented for the second time to obtain the target fluorescent spot area in the target fluorescent image.

[0029] In an optional implementation, the adjusting the area of ​​the fused fluorescent spot region includes:

[0030] The fused fluorescent light spot area is subjected to area reduction processing from the edge position of the fused fluorescent light spot area.

[0031] In a second aspect, an optional implementation of the present disclosure further provides a computer device, a processor, and a memory, wherein the memory stores machine-readable instructions executable by the processor, and the processor is used to execute the machine-readable instructions stored in the memory, and when the machine-readable instructions are executed by the processor, the machine-readable instructions perform the steps in the above-mentioned first aspect, or any possible implementation of the first aspect.

[0032] In a third aspect, an optional implementation of the present disclosure further provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed, the steps of the above-mentioned first aspect, or any possible implementation of the first aspect are executed.

[0033] It should be understood that the above general description and the following detailed description are merely exemplary and explanatory, and are not intended to limit the technical solutions of the present disclosure.

[0034] In order to make the above-mentioned objectives, features and advantages of the present disclosure more obvious and easy to understand, preferred embodiments are specifically cited below and described in detail with reference to the attached drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the following is a brief introduction to the drawings required for use in the embodiments. The drawings herein are incorporated into the specification and constitute a part of the specification. These drawings illustrate embodiments consistent with the present disclosure and are used together with the specification to illustrate the technical solutions of the present disclosure. It should be understood that the following drawings only illustrate certain embodiments of the present disclosure and should not be regarded as limiting the scope. For ordinary technicians in this field, other relevant drawings can also be obtained based on these drawings without creative work.

[0036] Figure 1 A schematic diagram of a sequencing system / instrument provided by some embodiments of the present disclosure is shown;

[0037] Figure 2 A schematic diagram showing the structure of an optical detection system provided by some embodiments of the present disclosure is shown;

[0038] Figure 3 A flowchart of a method for base recognition of a fluorescent image provided by some embodiments of the present disclosure is shown;

[0039] Figure 4 A flowchart of aligning multiple original fluorescence images provided by some embodiments of the present disclosure is shown;

[0040] Figure 5 A flowchart showing a specific method for performing the first segmentation process provided by some embodiments of the present disclosure;

[0041] Figure 6 shows an example diagram of fusion fluorescence provided by some embodiments of the present disclosure;

[0042] Figure 7 A schematic diagram of a computer device provided by some embodiments of the present disclosure is shown. DETAILED DESCRIPTION

[0043] In order to make the purpose, technical scheme and advantages of the embodiments of the present disclosure clearer, the technical scheme in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are only part of the embodiments of the present disclosure, rather than all of the embodiments. The components of the embodiments of the present disclosure generally described and shown here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present disclosure is not intended to limit the scope of the present disclosure claimed for protection, but merely represents the selected embodiments of the present disclosure. Based on the embodiments of the present disclosure, all other embodiments obtained by those skilled in the art without making creative work belong to the scope of protection of the present disclosure.

[0044] Library construction

[0045] The genomic DNA or RNA molecules to be sequenced are broken by physical or chemical means, for example, by ultrasound, to form DNA or RNA fragments. The two ends of the DNA or RNA fragments are first filled with enzymes, and then the two ends of the fragments are connected to a specific DNA or RNA sequence (usually, this specific DNA sequence or RNA is also called a linker) with a specific enzyme to form a mixture of DNA or RNA. This mixture of DNA or RNA is also called a library in the industry.

[0046] In order to save sequencing costs, generally, multiple samples will be sequenced in the sequencer at the same time. In order to distinguish the sequencing results of different samples, when preparing libraries of different samples, the connectors will contain a DNA or RNA sequence (usually containing 6-8 bases) that can identify the source of the sample. This DNA or RNA sequence that can identify the source of the sample can also be called a sample tag (or Index, Barcode). It can be understood that each sample library connector contains its own sample tag.

[0047] Typically, library construction is done outside of the sequencer, for example, by experimental manipulation in the laboratory to obtain the library.

[0048] Amplification reaction

[0049] Taking the amplification of a DNA library by bridge PCR (polymerase chain reaction) as an example, after the library is constructed, the library can be inoculated onto a sequencing chip and amplified on the sequencing chip. The adapters at both ends of the library are complementary to the first amplification primer on the sequencing chip, so the library can be inoculated onto the sequencing chip through complementary hybridization.

[0050] After the library is inoculated onto the sequencing chip, an amplification reaction can be performed using the library as a template chain. For example, the amplification reaction process can be to first add dNP and polymerase to the sequencing chip. The polymerase will start from the first amplification primer and synthesize a new DNA chain along the template chain. The new DNA chain is completely complementary to the template chain, so it is also called the complementary chain of the template chain. The complementary chain is covalently linked to the sequencing chip. Next, a NaOH alkaline solution is added to the sequencing chip for washing. The template chain and the complementary chain are mutually untied in the presence of the NaOH alkaline solution, and the template chain is washed away with the alkaline solution, while the complementary chain covalently linked to the sequencing chip is retained. Then add neutral liquid to the sequencing chip to neutralize the NaOH alkaline solution. The entire environment inside the sequencing chip becomes neutral, and the other end of the complementary chain will continue to hybridize with the second amplification primer on the sequencing chip. Add dNP and polymerase, and the polymerase will start from the second amplification primer and synthesize a new DNA chain along the complementary chain. At this time, the new DNA chain is completely complementary to the complementary chain and is exactly the same as the template chain. Then add NaOH alkaline solution to untie the two chains from each other, and then you can get two chains that are covalently connected to the sequencing chip and complementary to each other. Repeat this process, and the number of DNA chains will grow exponentially.

[0051] After amplification, the sequencing chip will retain the DNA double strands identical to the template strand and the complementary strand, respectively. Then, a specific reaction reagent is added to the sequencing chip to cut off the DNA strand synthesized from one of the amplification primers, for example, the DNA strand identical to the complementary strand is cut off, and the DNA strand identical to the template strand is retained. Then, a NaOH alkaline solution is added to the sequencing chip for washing. The alkaline solution disentangles the DNA double strands from each other, and the cut DNA strands are also washed away with the alkaline solution, ultimately leaving only single DNA strands on the sequencing chip. At this time, the number of single DNA strands retained on the sequencing chip is exponentially times the number at the beginning of amplification, thereby forming a DNA cluster, and all single DNA strands in a DNA cluster are identical. Then, a neutral solution is added, and all single DNA strands in the DNA cluster can be sequenced in a neutral solution environment.

[0052] It should be noted that the above-mentioned amplification reaction can be completed outside the sequencer, such as amplifying the library through experimental operations in the laboratory, or it can be completed inside the sequencer. When the amplification reaction is completed inside the sequencer, the library involved in the amplification reaction and various reaction reagents (for example, dNP, polymerase, NaOH alkaline solution, neutral solution, etc.) can be added to the sequencing chip through the liquid path system of the sequencer. In addition, the above amplification reaction is only exemplary, and the present application is not limited to the use of bridge PCR amplification mode, and other amplification modes can also be used for amplification, for example, loop-mediated isothermal amplification (LAMP), nucleic acid-dependent amplification (NASBA, Nuclear acidsequence-based amplification), rolling circle amplification (RCA, Rolling Circle Amplification), multiplex probe amplification (MPA, Multiplex Probe Amplification), etc.

[0053] Sequencing reaction

[0054] Taking the principle of sequencing by synthesis as an example, when sequencing is performed, four dNTPs with fluorescent groups are added to the sequencing chip through the liquid system, each dNTP can only be synthesized with one of the four bases of ATCG, and the 3' end of the dNTP has been blocked by a blocking group (the blocking group includes but is not limited to an azide group), and then polymerase is added to the sequencing chip through the liquid system. Through the action of the polymerase, one of the four dNTPs will be synthesized with the complementary base on the single strand being sequenced, and because the 3' end of the dNTP is blocked by a blocking group, only one dNTP can be extended on the single strand being sequenced each time. After synthesis, specific chemical reagents are added to the sequencing chip through the liquid system to flush out excess dNTPs and polymerase. Next, the optical detection system can be used to excite the fluorescent group of the dNTP that has been synthesized on the single chain, causing the fluorescent group to emit a fluorescent signal. Since the fluorescent group of each dNTP on a single chain in a cluster will emit the same fluorescent signal, the fluorescent signal is amplified. Therefore, the optical detection system can collect the fluorescent signal and generate a fluorescent image.

[0055] The computer system processes and analyzes the fluorescence image to determine which dNTP is synthesized on the sequenced single strand, and then based on the principle of complementarity, it can be inferred which base is synthesized with the dNTP on the sequenced single strand. At this point, a sequencing cycle is completed.

[0056] Next, specific chemical reagents are added to the sequencing chip through the liquid system to cut off the blocking group and the fluorescent group, thereby exposing the hydroxyl group at the 3' end of the dNTP.

[0057] Then enter the next sequencing cycle and repeat the above process.

[0058] It is understood that one sequencing cycle can detect one base, and after multiple sequencing cycles, multiple bases in the sequenced single strand can be detected. Specifically, the number of sequencing cycles can be determined according to the set sequencing read length, for example, 150 or 300 sequencing cycles.

[0059] Of course, it should be noted that the above sequencing reactions are only exemplary, and the present application is not limited to the sequencing-by-synthesis principle, and other sequencing principles may also be used.

[0060] Sequencing system

[0061] See also Figure 1 As shown, the sequencing system includes: a sequencing chip 10, a chip platform 20, a reagent storage container 30, a liquid path system (or a flow guide system) 40, an optical detection system 50, a computer system 60 and a waste liquid storage container 70. Among them:

[0062] The sequencing chip 10 is configured to provide a reaction area for amplification reaction and sequencing reaction;

[0063] A chip platform 20 configured to fix and support the sequencing chip 10;

[0064] A reagent storage container 30, configured to store one or more mixed sample libraries, one or more reagents;

[0065] The liquid circuit system 40 is configured to controllably transport one or more mixed sample libraries and one or more reagents from the reagent storage container 30 to the sequencing chip 10 so as to perform an amplification reaction and a sequencing reaction in the sequencing chip 10, and controllably transport waste liquid after the reaction from the sequencing chip 10 to the waste liquid storage container 70;

[0066] An optical detection system 50 is configured to excite and collect fluorescent signals during a sequencing reaction and generate a fluorescent image based on the fluorescent signals;

[0067] A computer system 60 is configured to obtain a fluorescent image from the optical detection system 50 and identify a base sequence of the sample library based on the fluorescent image;

[0068] The waste liquid storage container 70 is configured to store the waste liquid generated after the reaction.

[0069] Optical inspection system

[0070] See also Figure 2 As shown, taking the dual-color channel as an example, the optical detection system 50 includes at least a light source component and an imaging component. The light source component includes at least a light source (5101, 5102), a field aperture plate (5201, 5202), a dichroic mirror 5301 and a filter element 540.

[0071] Here, in one embodiment, the light source (5101, 5102) can be a light emitting diode (LED), and the LED is an aspherical mirror, which can diverge the LED point light source into parallel light. Of course, in addition to LED, other forms of point or surface light sources can also be used. In another embodiment, a collimating element ( Figure 2 ), used to collimate the light emitted by the light source (5101, 5102) into a parallel light beam. The collimating element may include one or more lenses, including but not limited to any one or any combination of a single lens, a cemented lens, a spherical lens, and an aspherical lens.

[0072] The parallel light beam is emitted through the field aperture plates (5201, 5202), and the field aperture plates (5201, 5202) limit the field of view range of the light emitted by the light source (5101, 5102), thereby limiting the field of view range of the excitation light irradiated onto the sequencing chip 10. The light passing through the field aperture plates (5101, 5102) reaches the dichroic mirror 5301. The dichroic mirror 5301 can transmit the light emitted by one of the light sources (5101, 5102) and reflect the light emitted by the other light source. The light passing through the dichroic mirror 5301 further passes through the filter element 540. The filter element 540 allows light of a certain wavelength band emitted by the light sources (5101, 5102) to pass through and serve as excitation light, while preventing light of other wavelength bands from passing through. For example, light of the same wavelength band as the fluorescence emitted by the fluorescent group is prevented from passing through, thereby ensuring that the fluorescence emitted by the fluorescent group does not contain stray light introduced by the light source (5101, 5102), which helps to improve the optical imaging effect.

[0073] In one implementation, the dichroic mirror 5301 can be fixed at a certain angle, and illustratively, the dichroic mirror 5301 is fixed by dispensing glue. In another implementation, the field aperture piece (5201, 5202) can be a rectangular aperture or a circular aperture, but not limited to; the field aperture piece (5201, 5202) can be a single-hole or multi-hole aperture piece; the field aperture piece (5201, 5202) can be made of an opaque material, and illustratively, can be a metal piece.

[0074] Here, it should be noted that in the optical detection system 50, the light sources (5101, 5102) emit light alternately, not simultaneously. By operating the light sources (5101, 5102) in a time-sharing manner (i.e., the light wavelengths emitted by the light sources 5101 and 5102 are different, and they are turned on in turn when in use), only one light source is turned on at a time, which will reduce the light power of the excitation light irradiated to the chip, which is beneficial to protecting the fluorescence lifetime of the fluorescent group.

[0075] In addition, in order to obtain the best fluorescent image through the imaging component, the imaging component needs to be focused. The light source component also includes a light source 5103, a field aperture 5203, an attenuation plate 550, and a dichroic mirror 5302. In one embodiment, considering that using LED light as excitation light for focusing will cause damage to the DNA chain to be sequenced on the sequencing chip 10, affecting the subsequent sequencing quality, the light source 5103 is, for example, a semiconductor laser (Laser Diode, LD), and emits lasers used for focusing the imaging component.

[0076] When focusing the imaging component, the laser emitted by the semiconductor laser passes through the field aperture plate 5203, and the field aperture plate 5203 limits the field of view range of the laser emitted by the semiconductor laser, and further limits the field of view range of the excitation light irradiated onto the sequencing chip 10. The field aperture plate 5203 can be a single-hole or multi-hole aperture plate. The field aperture plate 5203 can be made of an opaque material, and can be a metal plate for example. The laser can only pass through the light-transmitting hole portion of the field aperture plate corresponding to it, and the laser will be blocked for the non-light-transmitting hole portion of the field aperture plate. The laser passing through the field aperture plate 5203 reaches the attenuation plate 550, and the laser reaches the dichroic mirror 5302 after being attenuated by the attenuation plate 550. The dichroic mirror 5302 can transmit the laser and reflect the LED light. The dichroic mirror 5302 can be tilted and fixed at a certain angle, and can be fixed by dispensing glue for example. The laser light transmitted through the dichroic mirror 5302 or the reflected LED light reaches the convex lens 560 , and the convex lens 560 collimates the laser light or the LED light into a parallel light beam.

[0077] The imaging assembly at least includes a dichroic mirror 5303, an objective lens 570, a tube lens 580, and an image sensor 590. The laser and LED light collimated into parallel light beams pass through the dichroic mirror 5303. The dichroic mirror 5303 reflects the laser and LED light toward the sequencing chip 10. The reflected laser and LED light are irradiated onto the sequencing chip 10 through the objective lens 570, and excite the fluorescent group to emit a fluorescent signal. The objective lens 570 collects the fluorescent signal, and the fluorescent signal reaches the dichroic mirror 5303 through the objective lens 570. The dichroic mirror 5303 transmits the fluorescent signal to the tube lens 580. The tube lens 580 projects the fluorescent signal to the image sensor 590. The image sensor 590 collects the fluorescent signal and generates a fluorescent image. The computer system 60 identifies the base sequence according to the fluorescent image. In one embodiment, a filter element ( Figure 2 ), a filter element (not shown) may be provided between the tube lens 580 and the image sensor 590 Figure 2 not shown).

[0078] In one embodiment, the image sensor 590 may be an industrial camera. In another embodiment, the collimating element may include one or more lenses, including but not limited to any one or any combination of a single lens, a cemented lens, a spherical lens, and an aspherical lens.

[0079] In order to achieve the focus of the imaging component, a motor 600 is configured on one side of the objective lens 570. In one implementation, the motor can be, for example, a voice coil motor or other linear motor. The image quality of the fluorescent image recognized by the computer system 60 can be controlled by the driver to move the objective lens 570 up and down, so that the objective lens 570 reaches the optimal focal plane and finally obtains the best fluorescent image.

[0080] Base sequence recognition (also known as base recognition) is to determine the spot area of ​​the fluorescent image obtained by processing the collected fluorescent image and the corresponding pixel value of the spot area; the fluorescent signal released by the chemical reaction is captured in real time by the high-sensitivity industrial camera configured in the gene sequencer, and the base recognition software records and performs base recognition on the acquired fluorescent image. However, due to the existence of mechanical errors, the image sensor 590 will produce positional deviations between the images after imaging the fluorescent signal generated by the same sequencing object in different sequencing cycles; at the same time, due to the difference in the intensity of the fluorescent signal in different areas of the light spot, such as the fluorescent signal at the edge of the light spot is usually weaker than the fluorescent signal intensity in the center of the fluorescence, which will cause the fluorescent image to be prone to unclear spots. The above two situations result in a certain difference between the spot area and the real spot area when the spot area is currently identified from the fluorescent image, and this difference will lead to the problem of low base recognition accuracy.

[0081] Based on the above research, the present disclosure provides a base recognition method for fluorescent images, a computer device and a computer-readable storage medium. First, a first segmentation process is performed on multiple frames of fluorescent images to obtain candidate fluorescent spot areas corresponding to the multiple frames of fluorescent images, and then the candidate fluorescent spot areas corresponding to the multiple frames of fluorescent images are fused, so as to determine the fused fluorescent spot area where the spot has a higher probability of existing. At the same time, image sharpening processing is performed on the multiple frames of fluorescent images to eliminate the blurred edges of the spots, and then the fused fluorescent spot area is used to perform a second segmentation process on the target fluorescent image with the blurred edges eliminated, so that the target fluorescent spot area determined for each target fluorescent image has higher accuracy, thereby improving the base recognition accuracy of the fluorescent image.

[0082] The defects existing in the above solutions are the results obtained by the inventor after practice and careful research. Therefore, the discovery process of the above problems and the solutions proposed by the present disclosure for the above problems below should be the contributions made by the inventor to the present disclosure during the disclosure process.

[0083] It should be noted that similar reference numerals and letters denote similar items in the following drawings, and therefore, once an item is defined in one drawing, further definition and explanation thereof is not required in subsequent drawings.

[0084] To facilitate understanding of this embodiment, the base recognition method for a fluorescent image provided by the embodiment of the present disclosure is first introduced in detail. The execution subject of the base recognition method for a fluorescent image provided by the embodiment of the present disclosure is generally a computer device with certain computing capabilities, for example, Figure 1 The computer system 60 is integrated in the gene sequencer. In some possible implementations, the base recognition method of the fluorescent image can be implemented by a processor calling a computer readable instruction stored in a memory.

[0085] The base recognition method of the fluorescent image provided by the embodiment of the present disclosure is described below.

[0086] See also Figure 3 FIG. 1 is a flowchart of a method for base recognition of a fluorescent image provided by an embodiment of the present disclosure, wherein the method comprises steps S301 to S305, wherein:

[0087] S301: Acquire multiple frames of fluorescence images to be base-detected.

[0088] Here, the fluorescence image to be base-checked can be obtained by, for example, performing noise reduction on the original fluorescence image, or performing noise reduction on the original fluorescence image after image alignment. The multiple frames of original fluorescence images are obtained by performing multiple sequencing cycles on the same DNA fragment to be tested.

[0089] a: The original fluorescence image is subjected to noise reduction processing:

[0090] The noise reduction process includes, for example, smoothing the original fluorescence image, the purpose of which is to filter out noise and avoid the influence of noise on the subsequent spot area identification and base recognition. Image smoothing is to adjust the intensity value of the target pixel to be close to the intensity value of the surrounding pixels when the intensity value difference between the target pixel and the surrounding pixels in the original fluorescence image is large. Among them, the large difference in the intensity value between the target pixel and the surrounding pixels is reflected in that the intensity value of the target pixel is significantly lower than the intensity value of the surrounding pixels, or the intensity value of the target pixel is significantly higher than the intensity value of the surrounding pixels.

[0091] Exemplarily, when the original fluorescent image is smoothed, the surrounding pixels of the target pixel are determined according to the size of the smoothing template used when the image smoothing method is adopted. For example, a 5×5 smoothing template may have surrounding pixels within a 3×3 range centered on the target pixel, or may have surrounding pixels within a 5×5 range centered on the target pixel. The size of the smoothing template should also be selected according to the actual application. The larger the size of the smoothing template, the better the smoothing effect, but the more serious the distortion. In a preferred embodiment of the present application, a 5×5 smoothing template is selected.

[0092] Common noise reduction methods may include, for example, GL fractional differential algorithm, Gaussian filtering algorithm, median filtering algorithm, etc.

[0093] Among them, Gaussian filtering is a linear smoothing filter, which uses a smoothing template to determine the weighted average grayscale of pixels in the neighborhood, replacing the value of the central pixel of the template; median filtering is a nonlinear smoothing filter, which sets the grayscale value of each pixel to the median of the grayscale values ​​of all pixels in the neighborhood window determined by the smoothing template; GL fractional differential filtering uses a template (or convolution, mask) to scan each pixel in the fluorescent image, and uses the weighted average grayscale value of the pixels in the neighborhood determined by the template to replace the grayscale value of the central pixel of the template.

[0094] Exemplarily, the present disclosure uses the GL fractional-order differential filtering algorithm as an example for illustration. The GL fractional-order differential filtering algorithm is a nonlinear filtering operator that can weaken noise, especially the influence of salt and pepper noise on image information. The specific operation of the GL fractional-order differential operator filtering is: use a template (or convolution, mask) to scan each pixel in the fluorescent image, and use the weighted average gray value of the pixels in the neighborhood determined by the template to replace the gray value of the center pixel of the template.

[0095] The template of the GL fractional differential operator filter is calculated by the Grümwald-Letnikov formula of the fractional v-order derivative:

[0096]

[0097] In formula (1), h is the step size, and Γ(·) represents the gamma function.

[0098] According to formula (1), for the one-dimensional function f(t) defined on [a, t], it is divided equally according to the unit interval h-1, that is, The fractional differential expression of the one-dimensional function f(t) is as follows:

[0099]

[0100] When processing fluorescent images, the shortest distance for the gray value of an image pixel to change is between two adjacent pixels, that is, the minimum interval of a two-dimensional image g(s,t) is h = 1, that is, Then the partial differentials of g(s,t) in the x and y directions are:

[0101]

[0102] In order to complete the fractional-order differential operation of the fluorescence image in 8 central symmetric directions, the above two equations are extended to the other 6 directions (0°, 45°, 90°, 135°, 225°, 315°) to obtain a fractional-order differential filter based on 8 directions.

[0103] The fractional-order differential filter is the smoothing template used in the GL fractional-order differential filter. In the embodiment of the present disclosure, the size of the smoothing template may be, for example, 5×5, and the smoothing template is shown in the following formula (5):

[0104]

[0105] Among them, w1=8×1; w2=-v;

[0106] In the above formula 1, v represents the differential order, w1 is the weight value of the central pixel, w2 is the weight value in the 3×3 area around the central pixel, and w3 is the weight value in the 5×5 range extending in 8 directions. All pixels in the fluorescent image are scanned using the smoothing template corresponding to the above formula 1, and the weighted average gray value of the smoothing template is used to replace the value of the central pixel of the template. The purpose of image smoothing is achieved. Multiple frames of fluorescent images to be detected are obtained.

[0107] In addition, image smoothing may also be achieved in other ways, such as the median filter and Gaussian filter mentioned in the above examples, which will not be described in detail in this disclosure.

[0108] b: The situation of noise reduction on the original fluorescence image after image alignment:

[0109] For example, the multiple frames of original fluorescence images may be first aligned to obtain the aligned multiple frames of original fluorescence images, and then the aligned multiple frames of original fluorescence images may be filtered.

[0110] Specifically, for example, the following method can be used to align multiple frames of original fluorescence images:

[0111] Determine a template original fluorescence image among the multiple frames of the original fluorescence images;

[0112] Taking the template original fluorescence image as a reference, other original fluorescence images are aligned with the template original fluorescence image to obtain aligned multiple frames of original fluorescence images.

[0113] Specifically, when aligning the other original fluorescence images with the template original fluorescence image, for example, a similarity calculation method, a feature matching method, and a phase correlation method may be used but are not limited to.

[0114] See also Figure 4 As shown, the embodiment of the present disclosure takes the feature matching method as an example to explain the alignment process in detail. The alignment process includes the following S401 to S403:

[0115] S401: Extracting feature points of the original fluorescent image of the template and feature points of the original fluorescent images of other frames to be aligned.

[0116] Feature points can be some prominent points that will not disappear due to factors such as lighting, scale, rotation, etc., such as corner points, edge points, bright spots in dark areas, and dark spots in bright areas in the image. When extracting, the image positions in all scale spaces can be searched and feature points can be identified through Gaussian differential functions.

[0117] When extracting feature points, for example, feature extraction processing is performed on the area where the feature points are located to obtain feature vectors of each feature point. Each feature point can be represented by a set of feature vectors.

[0118] S402: Matching feature points of the template original fluorescence image and other frame original fluorescence images to obtain successfully matched target feature points.

[0119] Here, the feature vectors of different feature points obtained in the above steps are used for representation. For the template original fluorescence image p0, a feature vector set A0 is established based on the feature vectors corresponding to the feature points extracted from the template original fluorescence image, and for each frame of other original fluorescence images p1~pn, feature vector sets A1~An are established based on the feature vectors corresponding to the feature points extracted from the other original fluorescence images of each frame, and then A0 is matched with each of A1~An in turn. For A0 and A1, the target feature points corresponding to the template original fluorescence image p0 and the original fluorescence image p1 are obtained; for A0 and A2, the target feature points corresponding to the template original fluorescence image p0 and the original fluorescence image p2 are obtained, ..., for A0 and An, the target feature points corresponding to the template original fluorescence image p0 and the original fluorescence image pn are obtained.

[0120] When A0 and Ai (i∈[1,n]) are matched, for example, the distances between each feature vector in A0 and each feature vector in Ai are calculated in sequence; if the distance between a feature line vector a from A0 and a feature vector b from Ai is less than a preset distance threshold, it means that the feature points corresponding to the feature vector a and the feature vector b are target feature points that have been successfully matched.

[0121] S403: According to the mapping positions of the target feature points in the template original fluorescence image and the original fluorescence images of other frames, the original fluorescence images of other frames are adjusted to obtain aligned original fluorescence images.

[0122] Here, there is a mapping relationship between the feature points and the original feature vectors. According to the mapping relationship, the matching area between the original fluorescence images of other frames and the original fluorescence images of the template can be obtained anyway. Then, the position difference of the area in the original fluorescence images of other frames and the original fluorescence images of the template is used to adjust the original fluorescence images of other frames so that the matching area can be as close as possible or consistent in the image coordinate system, thereby completing the alignment processing of multiple original fluorescence images.

[0123] When matching feature points, for example, the distance between feature point A from the original fluorescent image of the template and feature point B from the original fluorescent image of other frames can be calculated. If the distance between feature point A and feature point B is less than a certain distance threshold, the two feature points are considered to be matched successfully.

[0124] The distance between feature points may be, for example, Euclidean distance, Hamming distance, cosine distance, etc., which is not limited in the embodiments of the present disclosure.

[0125] Using the above method, multiple frames of original fluorescence images are aligned to obtain multiple frames of aligned original fluorescence images. Then, the noise reduction method shown in a above can be used to process the aligned multiple frames of original fluorescence images to obtain multiple frames of fluorescence images to be detected.

[0126] Following the above S301, after obtaining multiple frames of fluorescence images to be base-detected, the method further includes the following steps.

[0127] S302: Based on the first intensity value information respectively corresponding to the pixel points in the multiple frames of the fluorescence images, the multiple frames of the fluorescence images are subjected to a first segmentation process to obtain candidate fluorescence spot regions respectively corresponding to the multiple frames of the fluorescence images, and the candidate fluorescence spot regions respectively corresponding to the multiple frames of the fluorescence images are subjected to a fusion process to obtain fused fluorescence spot regions corresponding to the multiple frames of the fluorescence images.

[0128] In a specific implementation, based on the first intensity value information respectively corresponding to the pixel points in the multiple frames of the fluorescent images, the multiple frames of the fluorescent images are subjected to a first segmentation process to obtain the candidate fluorescent spot regions respectively corresponding to the multiple frames of the fluorescent images, for example, the following method can be used:

[0129] Determine, from a plurality of pixels in a plurality of frames of the fluorescent image, candidate pixels whose first intensity value information belongs to a first intensity threshold interval;

[0130] Based on the position information of the candidate pixel points in the multiple frames of the fluorescence images, the candidate fluorescence spot areas in the multiple frames of the fluorescence images are determined.

[0131] The first intensity threshold interval includes: a first intensity minimum value. When determining a candidate pixel point from a plurality of pixel points, for example, the first intensity information of each pixel point and the first intensity minimum value may be compared; if the first intensity information is greater than or equal to the first intensity minimum value, the pixel point is a candidate pixel point; if the first intensity information is less than the first intensity minimum value, the pixel point is a non-candidate pixel point.

[0132] A specific method for determining candidate pixel points whose first intensity value information belongs to a first intensity threshold interval from a plurality of pixel points in a plurality of frames of the fluorescent image is as follows: comparing the first intensity value information of each pixel point in each frame of the fluorescent image to be base-detected with the first intensity threshold interval, and determining the pixel points whose first intensity value information belongs to the first intensity threshold interval from a plurality of pixel points in each frame of the fluorescent image to be base-detected; and determining the candidate fluorescent spot area in each frame of the fluorescent image to be base-detected based on the position information of the pixel points whose first intensity value information belongs to the first intensity threshold interval.

[0133] Here, the first and second segmentation processes are performed on each frame of the fluorescence image to be base-called to determine the candidate fluorescence spot area in the fluorescence image to be base-called, and the fused fluorescence spot area is obtained based on the candidate fluorescence spot areas of multiple frames of the fluorescence images to be base-called.

[0134] The first intensity threshold interval is represented by an intensity value range that can determine whether a pixel point can be identified as a fluorescent spot area. In practical applications, the selection of the first intensity threshold interval is limited by the detection equipment, detection technology, etc., and depends on the actual application.

[0135] In addition, a certain threshold may also be set, and the first intensity value of each pixel in each frame of the fluorescent image is compared with the preset threshold, and the area composed of pixels corresponding to the first intensity value greater than the threshold is the area of ​​interest.

[0136] The candidate fluorescent spot region can be represented as a set of position information of candidate pixels. The region where non-candidate pixels are located is the background region.

[0137] Then, the candidate pixels and non-candidate pixels are binarized, for example, the pixel values ​​of the candidate pixels can be binarized to 0, indicating white, and the pixel values ​​of the non-candidate pixels can be binarized to 1, indicating black, to obtain a black and white image. The white area is the candidate fluorescent spot area, and the black area is the background area.

[0138] Afterwards, the candidate fluorescent spot regions corresponding to the multiple frames of fluorescent images can be fused, that is, the union of the candidate fluorescent spot regions corresponding to the multiple frames of fluorescent images to be detected can be obtained to obtain a fused fluorescent spot region.

[0139] During fusion, the candidate fluorescent spot region of each frame of the fluorescent image is converted into a binary image. Assuming that the size of the binary image is m×n, for each pixel position (xi, yi) in the image coordinate system, determine whether there is at least one pixel point with a pixel value of 0 at the pixel position (xi, yi) in the multi-frame binary image; if so, determine the pixel position (xi, yi) as a pixel position in the fused fluorescent spot region. If the pixel values ​​of the pixel points at the pixel position in the multi-frame binary image are all 1, it means that the pixel point is a background pixel point and does not belong to the fused fluorescent spot region.

[0140] In addition, when performing the first segmentation process on the multiple frames of the fluorescent images, other methods may also be used, such as using any one of the Otsu method, the K-means segmentation algorithm, the mean segmentation algorithm, etc. Taking the K-means segmentation algorithm as an example, see Figure 3 A second segmentation process flow chart shown in Figure 5The first segmentation process of each frame of the fluorescent image to be base-detected includes the following steps S501 to S506, wherein:

[0141] S501: Input image data.

[0142] Each frame of the fluorescence image to be base called is input into the image segmentation model.

[0143] S502: Randomly select cluster centers.

[0144] K pixels are selected from the fluorescent image to be base-called as cluster centers of K classes. The cluster centers of K classes include cluster centers of fluorescent spots and cluster centers of non-fluorescent spots.

[0145] S503: Iterative classification.

[0146] In the i-th iteration, for any pixel point, according to the first intensity value corresponding to the pixel point, determine the candidate cluster center whose difference with the first intensity value of the pixel point is less than a certain difference range, and determine the distance between each candidate cluster center and the pixel point, and classify any pixel point into the category where the candidate cluster center with the shortest distance is located.

[0147] S504: Calculate new cluster centers.

[0148] Use methods such as mean to update the cluster center of this class.

[0149] S505: Determine whether the iteration satisfies the termination condition.

[0150] Repeat the above S3003 to S3004. For the K cluster centers, if the cluster size remains unchanged or the difference is less than the preset threshold, stop the iteration and execute S3006. If it is greater than the preset threshold, jump to S3003.

[0151] S5006: Output multiple frames of fluorescence images to be base-detected.

[0152] The above steps S3001 to S3006 are repeatedly executed until all the fluorescent images to be base-called have undergone the first and second segmentation processes.

[0153] Subsequently, a union process is performed on the candidate fluorescent spot regions of each frame of the fluorescent image to be base called after the first segmentation process to obtain a fused fluorescent spot region.

[0154] See also Figure 6 An example of a fusion fluorescence diagram is shown, in which Figure 6The S41 region in the figure is the union of the candidate fluorescent spot regions in each frame of the fluorescent image to be base-called, and the S43 region is the background region. The S42 region shows the difference between the region of interest identified by the original fluorescent image without noise reduction and the fluorescent image to be base-called after noise reduction. It can be seen that the area of ​​the fused fluorescent spot region obtained by fusion using the original fluorescent image without noise reduction is larger than the fused fluorescent spot region obtained by fusion using the fluorescent image to be base-called after noise reduction. Therefore, it is relatively more accurate to obtain the fused fluorescent spot region by fusion processing of multiple frames of fluorescent images to be base-called after noise reduction.

[0155] In some possible embodiments provided by the present disclosure, after determining the candidate fluorescent spot region in each frame of the fluorescent image to be base-called based on the position information of the pixel points belonging to the first intensity threshold interval based on the first intensity value information, the method further includes: filtering the spot region smaller than the region recognition template in the candidate fluorescent spot region corresponding to each frame of the fluorescent image to be base-called based on the region recognition template. The region recognition template is used to indicate the minimum spot region included in the candidate fluorescent spot region.

[0156] Here, the candidate fluorescent spot region may include multiple sub-regions; among these sub-regions, there may be sub-regions that are not connected to other sub-regions and have a smaller area; and although these small sub-regions belong to the first intensity threshold interval in the first intensity value information, they are more likely to belong to the noise part, such as some sub-regions that only contain a few pixels appearing in the background region. In order to filter out this part of the noise, in the embodiment of the present disclosure, these small sub-regions are filtered using a region recognition template.

[0157] The region recognition template can be an image template of a preset size. In practical applications, the light spot has a minimum size, and the light spot is usually approximately a circle. When it is smaller than this size, the area is usually considered to be a non-light spot area. Based on this feature, a region recognition template is set. The template can be equal to the size of the minimum light spot, for example, a 7×7 template. This template is used to traverse all sub-areas included in the candidate fluorescent area and filter the fluorescent image to be base-detected. If the size of a sub-area is smaller than the size of the region recognition template, the type corresponding to the sub-area is changed from "light spot" to "background".

[0158] The base recognition method of a fluorescent image provided in the embodiment of the present disclosure further includes the following S303. In the embodiment of the present disclosure, there is no sequence in which S302 and S303 are executed, that is, S302 may be executed first, S303 may be executed first, or both may be executed at the same time.

[0159] S303: Perform image sharpening processing on the multiple frames of fluorescence images respectively to obtain target fluorescence images corresponding to the multiple frames of fluorescence images respectively.

[0160] Here, since the fluorescent image to be base-called is smoothed, although the influence of salt-and-pepper noise on base recognition of the fluorescent image is weakened, the image edge is not clear enough. Therefore, the fluorescent image to be base-called is sharpened, and the first intensity value of the pixel in the fluorescent image to be base-called is stretched to obtain the second intensity value information corresponding to the pixel of the fluorescent image to be base-called.

[0161] Specifically, performing image sharpening processing on the multiple frames of fluorescence images respectively to obtain target fluorescence images corresponding to the multiple frames of fluorescence images respectively may specifically include:

[0162] Using a sharpening template of the target fluorescent image, convolution processing is performed on pixel points in multiple frames of the fluorescent image to obtain second intensity value information corresponding to the pixel points of the multiple frames of the fluorescent image;

[0163] Based on the second intensity value information corresponding to the pixel points in the multiple frames of the fluorescent images, target fluorescent images corresponding to the multiple frames of the fluorescent images are generated.

[0164] The sharpening template of the target fluorescence image is obtained by smoothing the weight coefficient of the image sharpening template using the Laplacian operator;

[0165] The image sharpening template is constructed based on the Riemann-Liouville RL fractional-order differential operator.

[0166] Specifically, when the fluorescent image is processed using the above sharpening template, the specific process may include:

[0167] An image sharpening template constructed based on a Riemann-Liouville RL fractional differential operator is obtained; a weight coefficient of the image sharpening template is smoothed using a Laplacian operator to obtain a sharpening template of a target fluorescent image; and pixel points in multiple frames of the fluorescent image to be base-detected are respectively convolved using the sharpening template of the target fluorescent image to obtain second intensity value information corresponding to the pixel points of each frame of the fluorescent image to be base-detected.

[0168] Here, the RL fractional differential operator has a weak derivative property and can nonlinearly retain the low-frequency information in the image. The low-frequency information here refers to the low-frequency information of the frequency domain diagram of the fluorescent image to be base detected in this field, that is, the area with a small change in the gradient of the pixel intensity value in the image. The area with a small change in the gradient of the pixel intensity value in the image is generally the background area, or the center area of ​​the light spot, and the area with a large change in the gradient of the pixel intensity value in the image is generally the edge of the light spot, or the edge of some noise with higher intensity values.

[0169] In practical applications, isotropic filters and anisotropic filters can be used to identify which are the light spot edges and which are the noise in the image. The characteristic of the light spot edge is that the gradient change of the intensity value in the normal direction is large, and the gradient change of the intensity value in the tangent direction is small, while the noise is that the gradient change of the intensity value in all directions is large. Based on this characteristic, some noise can be filtered out.

[0170] For example, assuming that the signal f(t) is in the interval [a, t], a∈R, and the general signal starts from zero, let a=0, v is the fractional differential order, then the definition of RL fractional differential is shown in the following formula (6):

[0171]

[0172] Here, function Γ(t) represents the Gamma function, which is a generalization of the factorial n!. And n can be a non-integer, as defined in formula (7).

[0173]

[0174] The fractional differentials of constant signals and non-constant signals are classified as follows: Let h (h>0) represent a discrete constant, let FW(h)f(t): = f(t+h), where: = represents the formula on the right to define the left. At this time, the v-order differential of f(t) is shown in formula (8), which can be expressed by the limit, as shown in formula (9).

[0175]

[0176] From the above formula, we can see that when the signal is a constant, if the fractional order v is within the interval (0, 1), the v-order derivative of the signal is zero. If f(t) is a continuous function, the v-order derivative of f(t) is shown in formula (10).

[0177]

[0178] In order to avoid high-order complexity, differential operations are performed in the positive and negative directions of x and y and the four diagonal directions. A 3×3 template is selected and the fractional differential templates in the eight directions are merged together to form a 5×5 image sharpening template, as shown in formula (11):

[0179]

[0180] in,

[0181] In the above formula (11), it can be known that the corresponding coefficient of the central pixel of the template is 8a0, which has the greatest impact on image enhancement. The farther away from the central pixel, the smaller the impact on the image. The corresponding coefficient a1 has a slightly smaller impact on the image, and the corresponding coefficient a2 has the smallest impact on the image.

[0182] In practical applications, the image sharpening template constructed using only the RL fractional differential operator does not produce a sharpened fluorescence image with an obvious gradient change of low-frequency information, which often results in the intensity value of the center area of ​​the spot not being stretched well. Therefore, the present application adds the Laplace operator to optimize the image sharpening template. Let s1 = 8a0 + 8a1, then b0 = 8a0 / s1, b1 = a1 / s1; let s2 = 8a0 + 8a1 + 8a2, then b2 = a2 / s2. The sharpening template of the target fluorescence image is obtained. As shown in the following formula (12):

[0183]

[0184] Since the weight value of the pixel center point b0 in formula (12) is 8a0 / s1, that is, the weight value of the pixel center point in the image sharpening template is weighted and averaged with the weight values ​​of the surrounding pixels, the weight value of the image sharpening template is appropriately smoothed to obtain the sharpening template of the target fluorescence image. When the sharpening template is applied to convolve each pixel point in the fluorescence image to be base detected, the gradient change of the intensity value in the center area of ​​the spot can be retained while ensuring the edge of the spot is clear and sharp.

[0185] The sharpening template of the target fluorescent image is used to perform convolution processing on the pixel points of each frame of the fluorescent image to be base-called, so as to obtain the second intensity value information of each pixel point in each frame of the fluorescent image to be base-called.

[0186] Then, a target fluorescent image corresponding to each fluorescent image is constructed according to the second intensity value information corresponding to each pixel point.

[0187] In addition, the image sharpening algorithm may also include: a histogram equalization algorithm, a Laplace algorithm, an object-based LOG transformation algorithm, an algorithm based on a gamma transformation, etc. The present disclosure embodiment does not limit this.

[0188] Following the above S302 and S303, after obtaining the fused fluorescent spot area and the target fluorescent image, the method further includes the following steps.

[0189] S304: performing a second segmentation process on the multiple frames of the target fluorescence images using the fused fluorescence spot region to obtain the target fluorescence spot region in the multiple frames of the target fluorescence images.

[0190] S305: Identify the base type based on the target fluorescent spot area.

[0191] Here, the second intensity value information of the fluorescent image to be base detected retains the intensity value and gradient change characteristics of the spot area. The fused fluorescent spot area includes areas in multiple fluorescent images to be base detected that are more likely to belong to the spot. According to the second intensity value information of the fluorescent image to be base detected, the fused fluorescent spot area is assigned an intensity value to obtain a target template image. The target template image includes the fused fluorescent spot area, as well as the intensity value, gradient change and other characteristics of the fused fluorescent spot area.

[0192] The target template image includes multiple mapping light spots, each of which is composed of multiple pixels with intensity value information. The light spot mapping position information indicates the coordinate position of each mapping light spot on the target template image. The light spot mapping position information can be used to determine the position information of the light spot corresponding to the mapping light spot in multiple frames of original fluorescence images. The mapping light spot does not specifically refer to a light spot in a certain frame of original fluorescence image, but indicates the light spots in multiple frames of original fluorescence images that correspond to the same coordinate position as the mapping light spot. The light spot position information includes: the light spot centroid position, the light spot edge position, the light spot area and other information.

[0193] In some possible implementations provided by the present disclosure, the second segmentation processing is performed on the target template image to determine the spot mapping position information of the target template image, including: comparing target intensity information of multiple pixel points of the target template image with a second intensity threshold interval, and determining, from the multiple pixel points of the target template image, the pixel points whose target intensity information belongs to the second intensity threshold interval; and determining the spot mapping position information based on the position information of the pixel points whose target intensity information belongs to the second intensity threshold interval.

[0194] Here, the second intensity threshold interval is similar to the first intensity threshold interval, except that the second intensity threshold interval compares the second intensity information of the fluorescent image to be base-detected after sharpening, while the first intensity threshold interval compares the first intensity information of the fluorescent image to be base-detected. The image segmentation algorithm can also adopt an algorithm similar to the first segmentation process, and the present disclosure will not elaborate on the image segmentation process.

[0195] After determining the spot mapping position information in the target template image, the spot position in the original fluorescent image can be indicated according to the spot mapping position information. Exemplarily, the spot corresponding to the corresponding coordinate position can be found in the original fluorescent image according to the coordinate position of the mapped spot indicated by the spot mapping position information.

[0196] The spot mapping position information usually indicates the pixel coordinates of a group of pixels, so that base recognition can be performed according to the corresponding pixels in the original fluorescent image. There are four-color identification methods and two-color identification methods, which can be selected according to the actual application.

[0197] In another embodiment of the present disclosure, since the fused fluorescent spot area is obtained by using candidate fluorescent spot areas respectively identified by multiple frames of fluorescent images, and each frame of candidate fluorescent spot areas may have a certain error; after the multiple frames of candidate fluorescent spot areas are fused, the fused fluorescent spot area obtained may amplify the error, thereby resulting in limited improvement in the accuracy of base recognition of the fluorescent image. Therefore, in another embodiment of the present disclosure, in order to further improve the accuracy of base recognition, after identifying the base type based on the target fluorescent spot area, the method further includes:

[0198] Comparing the base type with the target base type to obtain a comparison result;

[0199] Based on the comparison result, determining whether the base recognition accuracy of the target fluorescent spot area meets the target accuracy condition;

[0200] When the base recognition accuracy of the target fluorescent spot area does not meet the target accuracy condition, the area of ​​the fused fluorescent spot area is adjusted, and the part of the target fluorescent image belonging to the fused fluorescent spot area is re-segmented for the second time to obtain the target fluorescent spot area in the target fluorescent image.

[0201] When adjusting the area of ​​the fused fluorescent spot region, for example, it may include: performing area limiting processing on the fused fluorescent spot region from the edge position of the fused fluorescent spot region.

[0202] In this way, by further shrinking the edge of the fused fluorescent spot area, the area of ​​the determined target fluorescent spot area is reduced to a certain extent, and the pixel points where errors may occur at the edge of the spot are eliminated, thereby further improving the recognition accuracy of the bases.

[0203] like Figure 7As shown, the embodiment of the present disclosure provides a specific example of a method for base recognition of a fluorescent image. In this example, an optical acquisition system is used to acquire images of a nucleic acid sequence to obtain a plurality of original fluorescent images;

[0204] The present disclosure also provides a computer device, such as Figure 7 FIG. 1 is a schematic diagram of a computer device structure provided in an embodiment of the present disclosure, including:

[0205] A processor 71 and a memory 72; the memory 72 stores machine-readable instructions executable by the processor 71, and the processor 71 is used to execute the machine-readable instructions stored in the memory 72. When the machine-readable instructions are executed by the processor 71, the processor 71 performs the following steps:

[0206] Acquire multiple frames of fluorescence images to be base-called;

[0207] Based on the first intensity value information respectively corresponding to the pixel points in the multiple frames of the fluorescence images, the multiple frames of the fluorescence images are subjected to a first segmentation process to obtain candidate fluorescence spot regions respectively corresponding to the multiple frames of the fluorescence images, and the candidate fluorescence spot regions respectively corresponding to the multiple frames of the fluorescence images are subjected to a fusion process to obtain a fused fluorescence spot region corresponding to the multiple frames of the fluorescence images;

[0208] Performing image sharpening processing on the multiple frames of fluorescence images respectively to obtain target fluorescence images corresponding to the multiple frames of fluorescence images respectively;

[0209] Performing a second segmentation process on the target fluorescence images of multiple frames by using the fused fluorescence spot region to obtain the target fluorescence spot regions in the target fluorescence images of multiple frames;

[0210] The base type is identified based on the target fluorescent spot area.

[0211] The above-mentioned memory 72 includes internal memory 721 and external memory 722; the memory 721 here is also called internal memory, which is used to temporarily store the calculation data in the processor 71, as well as the data exchanged with the external memory 722 such as the hard disk. The processor 71 exchanges data with the external memory 722 through the internal memory 721.

[0212] The specific execution process of the above instructions can refer to the steps of the base recognition method of the fluorescent image described in the embodiment of the present disclosure, which will not be repeated here.

[0213] The present disclosure also provides a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the method for base recognition of fluorescent images described in the above method embodiment are executed. The storage medium can be a volatile or non-volatile computer-readable storage medium.

[0214] The embodiments of the present disclosure also provide a computer program product, which carries a program code. The instructions included in the program code can be used to execute the steps of the base recognition method of fluorescent images described in the above method embodiment. For details, please refer to the above method embodiment, which will not be repeated here.

[0215] The computer program product may be implemented in hardware, software or a combination thereof. In one optional embodiment, the computer program product is implemented as a computer storage medium. In another optional embodiment, the computer program product is implemented as a software product, such as a software development kit (SDK).

[0216] Those skilled in the art can clearly understand that, for the convenience and simplicity of description, the specific working process of the system and device described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here. In the several embodiments provided in the present disclosure, it should be understood that the disclosed system, device and method can be implemented in other ways. The device embodiments described above are merely schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some communication interfaces, and the indirect coupling or communication connection of the device or unit can be electrical, mechanical or other forms.

[0217] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0218] In addition, each functional unit in each embodiment of the present disclosure may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0219] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a non-volatile computer-readable storage medium that is executable by a processor. Based on this understanding, the technical solution of the present disclosure, or the part that contributes to the prior art or the part of the technical solution, can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present disclosure. The aforementioned storage medium includes: various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0220] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present disclosure, which are used to illustrate the technical solutions of the present disclosure, rather than to limit them. The protection scope of the present disclosure is not limited thereto. Although the present disclosure is described in detail with reference to the aforementioned embodiments, ordinary technicians in the field should understand that any technician familiar with the technical field can still modify the technical solutions recorded in the aforementioned embodiments within the technical scope disclosed in the present disclosure, or can easily think of changes, or make equivalent replacements for some of the technical features therein; and these modifications, changes or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present disclosure, and should be included in the protection scope of the present disclosure. Therefore, the protection scope of the present disclosure shall be based on the protection scope of the claims.

Claims

1. A method for base recognition using a fluorescent image, characterized in that: The method comprises: Acquire multiple frames of fluorescence images to be base-called; Based on the first intensity value information respectively corresponding to the pixel points in the multiple frames of the fluorescence images, the multiple frames of the fluorescence images are subjected to a first segmentation process to obtain candidate fluorescence spot regions respectively corresponding to the multiple frames of the fluorescence images, and the candidate fluorescence spot regions respectively corresponding to the multiple frames of the fluorescence images are subjected to a fusion process to obtain a fused fluorescence spot region corresponding to the multiple frames of the fluorescence images; Performing image sharpening processing on the multiple frames of fluorescence images respectively to obtain target fluorescence images corresponding to the multiple frames of fluorescence images respectively; Performing a second segmentation process on the target fluorescence images of multiple frames by using the fused fluorescence spot region to obtain the target fluorescence spot regions in the target fluorescence images of multiple frames; The base type is identified based on the target fluorescent spot area.

2. The method according to claim 1, characterized in that The step of acquiring multiple frames of fluorescent images to be base-checked comprises: Image noise reduction processing is performed on multiple frames of original fluorescent images to obtain fluorescent images to be base-detected corresponding to the multiple frames of original fluorescent images.

3. The method according to claim 1, characterized in that The first segmentation process is performed on the multiple frames of the fluorescence images based on the first intensity value information respectively corresponding to the pixel points in the multiple frames of the fluorescence images to obtain the candidate fluorescence spot areas respectively corresponding to the multiple frames of the fluorescence images, including: Determine, from a plurality of pixels in a plurality of frames of the fluorescent image, candidate pixels whose first intensity value information belongs to a first intensity threshold interval; Based on the position information of the candidate pixel points in the multiple frames of the fluorescence images, the candidate fluorescence spot areas in the multiple frames of the fluorescence images are determined.

4. The method according to claim 1, characterized in that: The performing image sharpening processing on the multiple frames of the fluorescence images respectively to obtain target fluorescence images corresponding to the multiple frames of the fluorescence images respectively includes: Using a sharpening template of the target fluorescent image, convolution processing is performed on pixel points in multiple frames of the fluorescent image to obtain second intensity value information corresponding to the pixel points of the multiple frames of the fluorescent image; Based on the second intensity value information corresponding to the pixel points in the multiple frames of the fluorescent images, target fluorescent images corresponding to the multiple frames of the fluorescent images are generated.

5. The method according to claim 4, characterized in that The sharpening template of the target fluorescence image is obtained by smoothing the weight coefficient of the image sharpening template using the Laplacian operator; The image sharpening template is constructed based on the Riemann-Liouville RL fractional-order differential operator.

6. The method according to claim 1, characterized in that The second segmentation process of the target fluorescence images of multiple frames using the fused fluorescence spot area includes: Determining a target area corresponding to the fused fluorescent spot area from multiple frames of the target fluorescent image; According to the second intensity value information corresponding to the pixel points in the target area, the target area is subjected to a second segmentation process to obtain the target fluorescent spot area.

7. The method according to any one of claims 1 to 6, characterized in that: After identifying the base type based on the target fluorescent spot area, the method further includes: Comparing the base type with the target base type to obtain a comparison result; Based on the comparison result, determining whether the base recognition accuracy of the target fluorescent spot area meets the target accuracy condition; When the base recognition accuracy of the target fluorescent spot area does not meet the target accuracy condition, the area of ​​the fused fluorescent spot area is adjusted, and the part of the target fluorescent image belonging to the fused fluorescent spot area is re-segmented for the second time to obtain the target fluorescent spot area in the target fluorescent image.

8. The method according to claim 7, characterized in that The adjusting the area of ​​the fused fluorescent spot region includes: The fused fluorescent light spot area is subjected to area reduction processing from the edge position of the fused fluorescent light spot area.

9. A computer device, characterized in that: include: A processor and a memory, wherein the memory stores machine-readable instructions executable by the processor, and the processor is used to execute the machine-readable instructions stored in the memory. When the machine-readable instructions are executed by the processor, the processor performs the steps of the base recognition method of the fluorescent image as described in any one of claims 1 to 8.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program. When the computer program is executed by a computer device, the computer device performs the steps of the method for base recognition of a fluorescent image according to any one of claims 1 to 8.

Images