WO2020155043A1 - 荧光图像配准方法、基因测序仪及系统、存储介质 - Google Patents

荧光图像配准方法、基因测序仪及系统、存储介质 Download PDF

Info

Publication number
WO2020155043A1
WO2020155043A1 PCT/CN2019/074244 CN2019074244W WO2020155043A1 WO 2020155043 A1 WO2020155043 A1 WO 2020155043A1 CN 2019074244 W CN2019074244 W CN 2019074244W WO 2020155043 A1 WO2020155043 A1 WO 2020155043A1
Authority
WO
WIPO (PCT)
Prior art keywords
pixel
points
point
preset
fluorescence image
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Ceased
Application number
PCT/CN2019/074244
Other languages
English (en)
French (fr)
Inventor
申晋晋
李大卫
刘扬宝
冯歌
李美
黎宇翔
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
BGI Shenzhen Co Ltd
Original Assignee
BGI Shenzhen Co Ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by BGI Shenzhen Co Ltd filed Critical BGI Shenzhen Co Ltd
Priority to CN201980002455.8A priority Critical patent/CN112204615B/zh
Priority to PCT/CN2019/074244 priority patent/WO2020155043A1/zh
Priority to US17/426,885 priority patent/US11972583B2/en
Publication of WO2020155043A1 publication Critical patent/WO2020155043A1/zh
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
    • G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
    • G06T7/337—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods involving reference images or patches
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/30—Determination of transform parameters for the alignment of images, i.e. image registration
    • G06T7/33—Determination of transform parameters for the alignment of images, i.e. image registration using feature-based methods
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
    • G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
    • G01N21/64—Fluorescence; Phosphorescence
    • G01N21/6486—Measuring fluorescence of biological material, e.g. DNA, RNA, cells
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/60—Analysis of geometric attributes
    • G06T7/66—Analysis of geometric attributes of image moments or centre of gravity
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/70—Determining position or orientation of objects or cameras
    • G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
    • C—CHEMISTRY; METALLURGY
    • C12—BIOCHEMISTRY; BEER; SPIRITS; WINE; VINEGAR; MICROBIOLOGY; ENZYMOLOGY; MUTATION OR GENETIC ENGINEERING
    • C12Q—MEASURING OR TESTING PROCESSES INVOLVING ENZYMES, NUCLEIC ACIDS OR MICROORGANISMS; COMPOSITIONS OR TEST PAPERS THEREFOR; PROCESSES OF PREPARING SUCH COMPOSITIONS; CONDITION-RESPONSIVE CONTROL IN MICROBIOLOGICAL OR ENZYMOLOGICAL PROCESSES
    • C12Q1/00—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions
    • C12Q1/68—Measuring or testing processes involving enzymes, nucleic acids or microorganisms; Compositions therefor; Processes of preparing such compositions involving nucleic acids
    • C12Q1/6869—Methods for sequencing
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
    • G01N21/62—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light
    • G01N21/63—Systems in which the material investigated is excited whereby it emits light or causes a change in wavelength of the incident light optically excited
    • G01N21/64—Fluorescence; Phosphorescence
    • G01N21/645—Specially adapted constructive features of fluorimeters
    • G01N21/6456—Spatial resolved fluorescence measurements; Imaging
    • G01N21/6458—Fluorescence microscopy
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10056—Microscopic image
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10064—Fluorescence image
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30004—Biomedical image processing
    • G06T2207/30072—Microarray; Biochip, DNA array; Well plate
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30168—Image quality inspection

Definitions

  • the invention relates to the field of gene sequencing, and in particular to a fluorescent image registration method, a gene sequencer, a gene sequencing system and a storage medium.
  • Gene sequencing refers to the analysis of the base sequence of specific DNA fragments, namely the arrangement of adenine (A), thymine (T), cytosine (C) and guanine (G).
  • A adenine
  • T thymine
  • C cytosine
  • G guanine
  • One of the commonly used sequencing methods at present is: the above-mentioned four bases carry four different fluorescent molecules, and different fluorescent molecules emit fluorescence of different wavelengths (colors) after being excited. By identifying the fluorescent wavelength, the The type of synthetic base to read the base sequence.
  • the second-generation sequencing technology uses a high-resolution microscopy imaging system to take pictures and collect the fluorescent molecular images of DNA nanoballs (ie DNB, DNA Nanoballs) on the biochip (gene sequencing chip), and send the fluorescent molecular images to the base recognition software for decoding
  • the image signal obtains the base sequence.
  • the effect of different fluorescent dyes on a base will produce fluorescent molecular signals of different wavelengths.
  • To detect the type of base it is necessary to analyze the fluorescent molecular energy of the base in different scenes, and firstly locate the same position coordinates in different scenes. This is the work of the registration algorithm.
  • sequencer products are equipped with sequencing data real-time processing and analysis software, most of which are equipped with registration and positioning algorithms.
  • different base signals need to be matched.
  • One base signal corresponds to a specific position in the image. It is necessary to match the corresponding position of the base in the image. Therefore, the speed, accuracy and robustness of matching are all in the image matching.
  • the existing image registration algorithms are generally designed based on the CPU, and the efficiency is low.
  • the throughput of the sequencer increases, the time for processing a picture will also increase linearly, and the edge synthesis of the second-generation sequencer will not be guaranteed.
  • the need for sequencing. And the registration of other points of the algorithm depends on the best point extracted. If the best point is not found well, the registration quality of all points will be affected. Therefore, there is an urgent need to provide a higher-performance image registration method.
  • the first aspect of the embodiments of the present invention provides a fluorescence image registration method, which is applied to a biochip, and the fluorescence image registration method includes:
  • S2 Select a preset local area of the fluorescence image, where at least one boundary frame exists in the first direction and the second direction of the preset local area, and the boundary frame contains a number of fluorescent molecular image points, The first direction is perpendicular to the second direction;
  • Step S5 Determine whether the fluorescence image meets the preset standard according to the number of good points, if it is determined that the fluorescence image meets the preset standard, then go to step S6, if it is determined that the fluorescence image does not meet the preset standard, return Step S1, reacquire the fluorescence image;
  • S7 Detect a fluorescent molecular image point on the intersection of the boundary line frame of the first direction and the second direction where the pixel-level registration point is located, and obtain the center of gravity position of the fluorescent molecular image point according to the center of gravity method;
  • the method before acquiring the positions of the pixels and the minimum value in the first direction and the second direction, the method further includes:
  • the template boundary line frame is a template line frame preset on the biochip ;
  • the obtaining the positions of the pixels and the minimum value in the first direction and the second direction includes:
  • the dividing the pixel-level registration points into good points and bad points according to a preset classification rule includes:
  • the absolute value is greater than the preset abscissa threshold, it is determined that the pixel-level registration point is a dead pixel.
  • the dividing the pixel-level registration points into good points and bad points according to a preset classification rule further includes:
  • the correcting the position of the bad dot according to the position of the good dot includes:
  • the abscissa of the dead pixel is acquired, and the ordinate of the dead pixel is corrected according to the fitted straight line.
  • the acquiring the position of the center of gravity of the fluorescent molecular image point according to the center of gravity method includes:
  • the position of the center of gravity of the fluorescent molecule image point is obtained according to a preset center of gravity formula.
  • a second aspect of the embodiments of the present invention provides a gene sequencing system, which is applied to a biochip, and the gene sequencing system includes:
  • a fluorescence image acquisition module for acquiring at least one fluorescence image of the biochip
  • the local area acquisition module is used to select a preset local area of the fluorescence image, wherein the preset local area has at least one boundary line frame in a first direction and a second direction, and the boundary line frame contains several Fluorescent molecular image point, the first direction is perpendicular to the second direction;
  • the pixel-level registration point acquisition module is used to acquire the position of the pixel and the minimum value in the first direction and the second direction in the preset local area, and according to the position of the pixel and the minimum value Obtain pixel-level registration points;
  • the registration point classification module is used to classify the pixel-level registration points into good points and bad points according to preset classification rules
  • An image quality judgment module configured to judge whether the fluorescence image meets a preset standard according to the number of good spots
  • a bad pixel position correction module configured to correct the position of the bad pixel according to the position of the good pixel
  • the fluorescent molecule center of gravity acquisition module is used to detect the fluorescent molecule image point on the intersection of the first direction and the second direction where the pixel-level registration point is located, and obtain the fluorescent molecule according to the center of gravity method The position of the center of gravity of the image point;
  • the sub-pixel level registration point acquisition module is used to fit a straight line in the first direction and the second direction respectively according to the position of the center of gravity of the fluorescent molecular image point, and the straight line fitted in the first direction is The intersection of the fitted straight lines in the second direction is the sub-pixel position of the registration point;
  • the boundary point position acquisition module is configured to acquire the boundary point of the fluorescence image according to the straight line fitted in the first direction and the second direction, and calculate the position of the boundary point.
  • a third aspect of the embodiments of the present invention provides a gene sequencer, the gene sequencer includes a processor, and the processor is configured to implement the fluorescence image registration method described in any one of the foregoing when executing a computer program stored in a memory.
  • a fourth aspect of the embodiments of the present invention provides a non-volatile computer-readable storage medium on which a computer program is stored, and when the computer program is executed by a processor, the fluorescence image registration method described in any one of the above is implemented.
  • Embodiments of the present invention provide a fluorescence image registration method, a gene sequencing system, a gene sequencer, and a non-volatile computer-readable storage medium to obtain at least one fluorescence image of a biochip; select a preset local area of the fluorescence image , Wherein at least one boundary frame exists in the predetermined local area in a first direction and a second direction, the boundary frame contains fluorescent molecular image points, and the first direction is perpendicular to the second direction; In the preset local area, obtain the position of the pixel and the minimum in the first direction and the second direction, and obtain the pixel-level registration point according to the position of the pixel and the minimum; The classification rule divides the pixel-level registration points into good points and bad points; according to the number of good points, it is determined whether the fluorescence image meets the preset standard, and if it is determined that the fluorescence image meets the preset standard, then according to the Correct the position of the good point to correct the position of the bad point; detect the fluorescent mo
  • the present invention proposes a high-precision GPU parallel fluorescence image registration method based on point features, which can locate and register base information maps of different sequencing platforms with high accuracy and efficiency.
  • the embodiment of the present invention provides The fluorescence image registration method also judges whether the quality of the collected fluorescence image meets the standard by the number of good points in the classification, and discards the fluorescence image that does not meet the standard, so it has a strong anti-impurity interference ability.
  • FIG. 1 is a flowchart of a method for registering a fluorescent image according to an embodiment of the present invention.
  • Fig. 2 is a schematic diagram of a preset partial area of a fluorescence image provided by an embodiment of the present invention.
  • FIG. 3A is a schematic diagram of pixel-level positioning of the border line frame in the second direction.
  • FIG. 3B is a pixel and curve diagram of a preset rectangular area in the second direction.
  • Fig. 4A is a partial enlarged view of the border line frame in the fluorescence image.
  • Fig. 4B is a schematic diagram of the arrangement of fluorescent molecular image points on the border line frame shown in Fig. 4A.
  • Figure 4C is a schematic diagram of local area pixels and features.
  • FIG. 5A is a schematic diagram of pixel level correction in the first direction.
  • FIG. 5B is a schematic diagram of extreme correction in the first direction.
  • Fig. 6A is a schematic diagram of pixel-level positions.
  • Fig. 6B is a schematic diagram of sub-pixel level positions.
  • FIG. 7A is a schematic diagram of sub-pixel level registration.
  • FIG. 7B is another schematic diagram of sub-pixel level registration.
  • Figure 8 is a schematic diagram of the GPU programming model.
  • Fig. 9 is a schematic structural diagram of a gene sequencer according to an embodiment of the present invention.
  • Fig. 10 is an exemplary functional module diagram of the gene sequencer shown in Fig. 9.
  • Pixel-level registration point acquisition module 105 Registration point classification module 107 Image quality judgment module 109 Bad pixel location correction module 111 Fluorescent molecular center of gravity acquisition module 113 Sub-pixel level registration point acquisition module 115 Boundary point position acquisition module 117
  • FIG. 1 is a flowchart of a method for registering a fluorescent image according to an embodiment of the present invention. As shown in FIG. 1, the fluorescence image registration method may include the following steps:
  • the biochip may be a gene sequencing chip, and the fluorescence image may be a fluorescence signal image taken during sequencing.
  • the biochip has several track lines (Trackline, such as the track region in US Patent US9880089) in a first direction and a second direction, wherein the first direction is perpendicular to the second direction, and the first The direction may be a vertical direction, and the second direction may be a horizontal direction.
  • the area formed between two adjacent trajectory lines in the first and second directions is called a block, and the block is divided into an inner block and an outer block.
  • DNS DNA nanosphere molecules
  • the DNA nanosphere molecules carry fluorescent molecules during base synthesis, and the fluorescent molecules emit fluorescent signals when excited.
  • the fluorescent molecules can be fixedly arranged on the biochip according to preset rules, and through special design and processing, there are no sites on the biochip, that is, no fluorescent molecules exist.
  • the border line of non-luminescence The frame is highlighted.
  • the highlighted border frame may be composed of three fluorescent molecular image points, the fluorescent molecular image points in the middle row are bright spots, and the fluorescent molecular image points on both sides of the middle row are not bright.
  • the fluorescent molecular image points in the middle row form a trajectory line, and the fluorescent molecular image points on both sides of the middle row form a dark line.
  • the imaging point of the fluorescent molecule becomes the imaging point of the fluorescent molecule.
  • the boundary line frame may include the trajectory line and dark lines on both sides of the trajectory line.
  • a TDI camera can be used to scan the biochip according to a preset scanning sequence to obtain a fluorescent signal image.
  • the preset scanning order may be preset by the terminal user, and the preset scanning order includes, but is not limited to scanning in a top-down order.
  • S2 Select a preset local area of the fluorescence image, where at least one boundary frame exists in the preset local area in the first direction and the second direction, and the boundary frame contains fluorescent molecular image points, so The first direction is perpendicular to the second direction.
  • the preset local area may be preset by the end user, and the preset local area has at least one boundary frame in the first direction and the second direction, that is, the At least one registration point exists in the preset local area.
  • FIG. 2 is a schematic diagram of a preset partial area of a fluorescence image provided by an embodiment of the present invention.
  • the preset local area may be a 9*9 area
  • there are 9 boundary line frames in the first direction (indicated by the black solid line in FIG. 2)
  • there are 9 boundary line frames represented by black solid lines in Figure 2)
  • the line frame in the first direction intersects the line frame in the second direction
  • 81 (9*9) intersection points (registration points) are obtained .
  • the white solid line in FIG. 2 represents a template boundary line frame
  • the template boundary line frame is a template line frame preset on the biochip.
  • There are locations marked with asterisks on the four boundaries in FIG. 2, and the locations marked with white asterisks are the boundary point positions of the fluorescence image.
  • the method further includes: obtaining the second The position of the template boundary line frame in the first direction and the second direction.
  • the template boundary line frame is the template line frame preset on the biochip. It is the actual physical position of the boundary line frame of the biochip.
  • the camera imaging will produce distortion, so there will be a deviation between the boundary line frame of the fluorescent signal image and the template boundary line frame; select a preset rectangular area according to the position of the template boundary line frame; within the preset rectangular area, obtain The position of the pixel and the minimum value in the first direction and the second direction.
  • the preset rectangular area is preset by the terminal user.
  • the preset rectangular area includes a template registration point.
  • the preset rectangular area may be arranged symmetrically with respect to the position of the template registration point. It is understandable that, for a certain column in the first direction, a preset rectangular area needs to be selected at the position of the template wire frame where each template registration point on the column is located, and within each preset rectangular area To obtain the position of the pixel and the minimum value in the first direction, so as to obtain the position of the actual boundary line frame on the longitudinal column.
  • the “sum of pixels” referred to herein refers to the sum of the brightness values of the pixels. "The position of the pixel and the minimum in the first direction and the second direction” refers to the row with the smallest sum of the brightness values of the entire row of pixels in that direction.
  • FIG. 3A is a schematic diagram of pixel-level positioning of the boundary wire frame in the second direction
  • FIG. 3B is a pixel and curve diagram of a preset rectangular area in the second direction.
  • FIG. 3A there are two types of boundary line frames in the figure, black represents the actual boundary line frame, and white represents the template boundary line frame.
  • the second direction the position of the template boundary line frame in the second direction is obtained, and a preset rectangular area is selected according to the position of the template boundary line frame.
  • the preset rectangular area is represented by a white solid line
  • the center of the preset rectangular area is generally the intersection of the template boundary line frame in the first direction and the template boundary line frame in the second direction .
  • the length and width of the preset rectangular area cannot be greater than the template boundary line frame The shortest distance.
  • the size of the preset rectangular area is set to 128*128 pixels, which can simultaneously ensure program running speed and positioning accuracy.
  • the obtaining the position of the pixel and the minimum value in the first direction and the second direction includes: selecting a plurality of template lines; respectively sequentially arranging the plurality of template lines in the first direction and the second direction
  • the template line performs a translation operation in the preset rectangular area; the superimposed sum of the brightness values of pixels covered by the positions of several template lines on the preset rectangular area is calculated, and the superimposed sum of the brightness values is the template line
  • the obtaining the pixel-level registration point according to the position of the pixel and the minimum value includes: the intersection of the position of the pixel and the minimum value in the first direction and the position of the pixel and the minimum value in the second direction is Pixel-level registration points. As shown in FIG.
  • each trough in FIG. 3B corresponds to the position of the boundary line frame in FIG. 3A.
  • Figure 4A is a partial enlarged view of the border frame in the fluorescence image.
  • Figure 4B is a schematic diagram of the arrangement of fluorescent molecular image points on the border frame shown in Figure 4A.
  • the border frame can be composed of three fluorescent molecular image points, the fluorescent molecular image points in the middle row are bright spots, and the fluorescent molecular image points on both sides of the middle row are not bright.
  • the fluorescent molecular image points in the middle row form a trajectory line
  • the fluorescent molecular image points on both sides of the middle row form a dark line.
  • the highlighted boundary line frame may include the trajectory line and dark lines on both sides of the trajectory line.
  • the size of the fluorescent molecules corresponding to different biochips and different cameras is different.
  • the present invention takes the size of 3*3 pixels as an example.
  • a W-shaped line feature black The part selected by the ellipse
  • the two trough positions are the positions corresponding to the dark lines on both sides of the track line on the biochip, and the pixels and values of the positions corresponding to the dark lines are lower.
  • the position corresponding to the wave peak in the W-shaped line feature is the position corresponding to the trajectory line, and the pixel and value of the position corresponding to the trajectory line are higher.
  • the pixels at the positions (-4)(-3)(-2) and (2)(3)(4) on both sides of each point in the preset rectangular area are summed, and the sum is the smallest point Corresponds to the position of the boundary wire frame.
  • the position of the boundary line frame in the first direction can be obtained.
  • the intersection of each boundary line frame in the first direction and each boundary line frame in the second direction is the pixel-level registration point, that is, the common 81 (9*9) pieces.
  • the embodiment of the present invention provides a straight line fitting method to determine the quality of each pixel-level registration point, and at the same time correct the dead pixels, that is, correct while judging. Finally, the conditions of all pixel-level registration points are combined to determine the quality of the entire fluorescent image.
  • the specific process is as follows:
  • the dividing the pixel registration points into good points and bad points according to a preset classification rule includes: obtaining the pixel-level registration points on any column in a first direction of the preset local area; calculating one of them The absolute value of the abscissa difference between the pixel-level registration point and the other pixel-level registration points in the column; if there are at least two of the absolute values less than the preset abscissa threshold, it is determined that the pixel-level registration point is Good point, otherwise, the pixel-level registration point is determined to be a bad point.
  • the preset abscissa threshold may be preset by those skilled in the art according to experience values.
  • the range of searching for fluorescent molecules in the vicinity of the borderline frame is 3 pixels during sub-pixel level registration. As long as it is within 3 pixels, the best position can be positioned through sub-pixel registration. Therefore, the preset abscissa threshold is 3 pixels.
  • the preset local area In the second direction of the preset local area, obtain the pixel registration points on any row; select one of the pixel-level registration points, and calculate the pixel-level registration point and the other pixel-level registration points on the row
  • the slope of the punctual point; the slopes of the pixel-level registration points on the row are compared according to a preset sorting rule (the preset sorting order can be from small to large, or from large to small) Sort, obtain the first median value as the selected first slope of the pixel-level registration point; obtain the first slopes of the other pixel-level registration points on the row, and combine all the pixel-level registration points on the row
  • the slopes are sorted according to the preset sorting rule, the second median value is obtained as the second slope of the row, and the second slope corresponds to the point on the row as the reference point; and the other points on the preset local area are obtained
  • the registration point is determined to be a dead pixel, that is, because the boundary line (trajectory line) in the first direction has been previously framed
  • the pixel-level abscissa has been determined. According to the abscissa, the ordinate value corresponding to the second direction (slope) line is obtained, and then the pixel-level registration point and the ordinate value are determined as the absolute value. If the absolute value is greater than 3 Pixel, then the registration point is determined to be a dead pixel.
  • Step S5 Determine whether the fluorescence image meets the preset standard according to the number of good points, if it is determined that the fluorescence image meets the preset standard, then go to step S6, if it is determined that the fluorescence image does not meet the preset standard, return Step S1, reacquire the fluorescence image.
  • the judging whether the fluorescence image meets a preset standard according to the number of good points includes: in the first direction, if the number of good points in each column is greater than 3, then determining The fluorescence image meets the preset standard, otherwise, the fluorescence image is discarded.
  • the second direction if the number of good points on each line is greater than the preset number, it is determined that the fluorescence image meets the preset standard, otherwise, the fluorescence image is discarded, and step S1 is executed to select the corresponding image again .
  • the preset number is pre-adjusted and set according to the selected preset local area size. In this embodiment, the preset local area is a 9*9 area, and the preset number is 5. That is, in the second direction, if the number of good points on each line is greater than 5, it is determined that the fluorescence image meets the preset standard, otherwise, the fluorescence image is discarded.
  • the boundary line frame in the first direction always maintain the vertical orientation. Therefore, the bad pixels in the first direction may be corrected first, and the correction of the bad pixels in the second direction depends on the abscissa of the corrected bad pixels in the first direction.
  • the correcting the position of the bad point according to the position of the good point includes: obtaining the mean value of the abscissa of the good point in the first direction of the preset local area, and assigning the mean value of the abscissa to The dead pixel; in the second direction of the preset local area, the abscissa of the dead pixel is obtained, and the ordinate of the dead pixel is corrected according to the fitted straight line. It is understandable that after correcting the dead pixels in the first direction of the preset local area, the abscissa of the pixel-level registration point can be determined. Therefore, in the second direction of the preset local area, the ordinate of the dead point can be corrected by the fitted straight line according to the determined abscissa of the dead point.
  • FIG. 5A is a schematic diagram of pixel-level correction in the first direction
  • FIG. 5B is a schematic diagram of extreme correction in the first direction.
  • the left straight line is the straight line before correction
  • the gray solid circles represent good points
  • the left straight line there are six near the left straight line.
  • Dead pixels that need to be corrected black solid circles indicate dead pixels).
  • the dead points can be corrected to obtain the straight line on the right, the straight line on the right is the corrected straight line, and the straight line on the right is present 9 good points. It is understandable that there is also a more extreme case. As shown in FIG. 5B, there are 9 pixel-level registration points on the straight line on the left, and every three points are on the same straight line. At this time, according to the above-mentioned good point determination method, all 9 points are good points. However, these 9 pixel-level registration points are not on the same straight line.
  • the method further includes: obtaining the abscissa difference of any two of the 9 pixel-level registration points; judging whether the abscissa difference is greater than a preset abscissa threshold, and if the abscissa is determined If the difference is greater than the preset abscissa threshold, the fluorescence image is considered to be of poor quality, and the fluorescence image is discarded.
  • S7 Detect a fluorescent molecular image point on the intersection of the boundary line frame of the first direction and the second direction where the pixel-level registration point is located, and obtain the center of gravity position of the fluorescent molecular image point according to the center of gravity method.
  • the acquiring the center of gravity position of the fluorescent molecular image point according to the center of gravity method includes: acquiring the position coordinates of the pixel-level registration point; acquiring the pixel brightness value of the fluorescent molecular image point; acquiring the fluorescent molecule according to a preset center of gravity formula The position of the center of gravity of the image point.
  • the fluorescent molecular image points on the crossover line of the borderline frame are actually the coding sites on the trackline (such as fluorescent spots on the track region in US Patent US9880089).
  • FIG. 6A is a schematic diagram of pixel-level positions
  • FIG. 6B is a schematic diagram of sub-pixel-level positions.
  • the black line frame area represents a pixel
  • the pixel is further subdivided down to obtain the sub-pixel position, as shown in the gray solid circle in FIG. 6B.
  • the range of the intersection line is set to -64 to 64 pixels, that is, the length is 128 pixels, which is consistent with the length of the preset rectangular area.
  • the detection conditions of fluorescent molecular image points are shown in Figure 6A.
  • the brightness value of 1 position is the highest and greater than the image background value, and the brightness value of 2345 position is higher than that of 6789. Then use the center of gravity method to obtain the position of the center of gravity of the fluorescent molecular image point.
  • the formula of the center of gravity method is as follows:
  • (X 1 , Y 1 ) represents the pixel-level coordinates of 1, H 1 ⁇ 5 respectively represent the brightness value of the position corresponding to the number, and H b, q are the image background value.
  • a straight line is fitted in the first direction and the second direction by the least square method, and the fitted line in the first direction is The intersection of the straight line and the fitted straight line in the second direction is the sub-pixel position of the registration point.
  • FIG. 7A is a schematic diagram of sub-pixel level registration
  • FIG. 7B is another schematic diagram of sub-pixel level registration.
  • the center of gravity of the detected fluorescent molecule image point is fitted to a straight line in the first direction and the second direction by the least square method, and the intersection of the straight line is the sub-pixel position of the registration point.
  • the black solid circle represents the sub-pixel level registration point
  • the gray solid circle represents the center of gravity of the fluorescent molecular image point.
  • the boundary point of the fluorescence image is acquired according to the straight line fitted in the first direction and the second direction, and the position of the boundary point is calculated.
  • Figure 2 there are positions marked with asterisks on the four boundaries.
  • the position marked with asterisks is the position of the boundary points of the fluorescence image.
  • the fitted straight line can be calculated according to the sub-pixel level registration point correction. Draw out the points on the border of the fluorescent image. It is understandable that since a coordinate value of the boundary point is known (for example, the abscissa of the leftmost column of boundary points are all 0), so only the known coordinate values need to be substituted into the fitted straight line. Another coordinate value can be calculated to get the position of the boundary point. It is understandable that for a region where the preset local region is 9*9, a total of 121 (11*11) sub-pixel-level registration points can be obtained by the above method.
  • the embodiment of the present invention provides a fluorescence image registration method to obtain at least one fluorescence image of a biochip; select a preset local area of the fluorescence image, wherein the preset local area is in a first direction and a second direction
  • the boundary line frame contains fluorescent molecular image points, the first direction is perpendicular to the second direction; in the preset local area, the first direction and the According to the position of the pixel and the minimum value in the second direction, the pixel-level registration point is obtained according to the position of the pixel and the minimum value; the pixel-level registration point is divided into good points and bad points according to preset classification rules;
  • the number of the good points determines whether the fluorescence image meets a preset standard, and if it is determined that the fluorescence image meets the preset standard, the position of the bad point is corrected according to the position of the good point; Detect the fluorescent molecular image point on the intersection of the border line frame of the first direction and the second direction where the quasi-point
  • the present invention proposes a high-precision GPU parallel fluorescence image registration method based on point features, which can locate and register base information maps of different sequencing platforms with high accuracy and efficiency.
  • the embodiment of the present invention provides The fluorescence image registration method also judges whether the quality of the collected fluorescence image meets the standard by the number of good points in the classification, and discards the fluorescence image that does not meet the standard, so it has a strong anti-impurity interference ability.
  • Figure 8 is a schematic diagram of the GPU programming model.
  • the fluorescence image registration method provided by the embodiment of the present invention may be designed based on GPU, thereby improving the fluorescence image registration rate.
  • the first box from the left refers to the CPU, and the second and third boxes refer to the GPU.
  • the host contains three core functions, namely core function 1, core function 2, and core function 3.
  • Each running core function contains a thread network (grid), and a thread network can have multiple thread blocks (block) , There can be multiple threads (threads) in a thread block.
  • Thread blocks can be understood as coarse-grained parallelism, and threads can be understood as fine-grained parallelism.
  • fine-grained parallelism can make multiple fluorescent images and multiple registration points of each fluorescent image be processed in parallel, and fine-grained parallelism can make each registration point pixel-level registration , Calibration and other processes are done in parallel.
  • 9*9 thread blocks can be designed and used on the GPU, which are respectively responsible for the registration of each region.
  • the design idea of using thread blocks and the number of threads in the subsequent fluorescent image registration step is consistent with this, and will not be described in detail.
  • An embodiment of the present invention also provides a gene sequencer, including a memory, a processor, and a computer program stored in the memory and capable of running on the processor.
  • the processor executes the program when the program is executed.
  • the gene sequencer may include a chip platform, an optical system, and a liquid path system.
  • the chip platform can be used to load biochips
  • the optical system can be used to acquire fluorescence images
  • the liquid path system can be used to perform biochemical reactions using preset reagents.
  • Fig. 9 is a schematic structural diagram of a gene sequencer according to an embodiment of the present invention.
  • the gene sequencer 1 includes a memory 10 in which a gene sequencing system 100 is stored.
  • the gene sequencing system 100 can obtain at least one fluorescence image of the biochip; select a preset partial area of the fluorescence image, wherein at least one boundary frame exists in the preset partial area in the first direction and the second direction ,
  • the boundary line frame contains fluorescent molecular image points, the first direction is perpendicular to the second direction; in the preset local area, the sum of pixels in the first direction and the second direction is acquired
  • the position of the minimum value is located, and the pixel-level registration point is obtained according to the position of the pixel and the minimum value; the pixel-level registration point is divided into good points and bad points according to preset classification rules; according to the number of good points Determine whether the fluorescence image meets the preset standard, if it is determined that the fluorescence image meets the preset standard, correct the position of the bad
  • the present invention proposes a high-precision GPU parallel fluorescence image registration method based on point features, which can locate and register base information maps of different sequencing platforms with high accuracy and efficiency.
  • the embodiment of the present invention provides The fluorescence image registration method also judges whether the quality of the collected fluorescence image meets the standard by the number of good points in the classification, and discards the fluorescence image that does not meet the standard, so it has a strong anti-impurity interference ability.
  • the gene sequencer 1 may also include a display screen 20 and a processor 30.
  • the memory 10 and the display screen 20 may be electrically connected to the processor 30 respectively.
  • the memory 10 may be different types of storage devices for storing various types of data.
  • it can be the memory or internal memory of the gene sequencer 1, or can be a memory card that can be connected to the gene sequencer 1, such as flash memory, SM card (Smart Media Card, smart media card), SD card (Secure Digital Card, Secure digital card) etc.
  • the memory 10 may include a high-speed random access memory, and may also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart media card (SMC), and a secure digital (SD) Card, Flash Card, at least one magnetic disk storage device, flash memory device, or other volatile solid-state storage device.
  • the memory 10 is used to store various types of data, for example, various types of applications (Applications) installed in the gene sequencer 1, and data set and obtained by applying the above-mentioned fluorescence image registration method.
  • the display screen 20 is installed in the gene sequencer 1 for displaying information.
  • the processor 30 is configured to execute the fluorescence image registration method and various software installed in the gene sequencer 1, such as an operating system and application display software.
  • the processor 30 includes, but is not limited to, a processor (Central Processing Unit, CPU), a Micro Controller Unit (Micro Controller Unit, MCU) and other devices for interpreting computer instructions and processing data in computer software.
  • a processor Central Processing Unit, CPU
  • MCU Micro Controller Unit
  • the gene sequencing system 100 may include one or more modules, and the one or more modules are stored in the memory 10 of the gene sequencer 1 and configured to be operated by one or more processors (this embodiment is A processor 30) executes to complete the embodiment of the present invention.
  • the gene sequencing system 100 may include a fluorescence image acquisition module 101, a local area acquisition module 103, a pixel-level registration point acquisition module 105, a registration point classification module 107, an image quality judgment module 109, and dead pixels
  • the module referred to in the embodiment of the present invention may be a program segment that completes a specific function, and is more suitable than a program to describe the execution process of software in a processor.
  • the gene sequencer 1 may include some or all of the functional modules shown in FIG. 10, and the functions of each module will be described in detail below. It should be noted that the same nouns and related nouns and specific explanations in the above embodiments of the fluorescence image registration method can also be applied to the following introduction to the functions of each module. To save space and avoid repetition, I won’t repeat them here.
  • the fluorescence image acquisition module 101 can be used to acquire at least one fluorescence image of the biochip.
  • the local area acquisition module 103 may be used to select a preset local area of the fluorescence image, where at least one boundary line frame exists in the first direction and the second direction of the preset local area, and the boundary line frame contains A number of fluorescent molecule image points, the first direction is perpendicular to the second direction.
  • the pixel-level registration point acquisition module 105 can be used to acquire the position of the pixel and the minimum value in the first direction and the second direction in the preset local area, and according to the position of the pixel and the minimum value. The position gets the pixel-level registration point.
  • the registration point classification module 107 may be configured to classify the pixel-level registration points into good points and bad points according to preset classification rules.
  • the image quality judgment module 109 can be used to judge whether the fluorescence image meets a preset standard according to the number of good points.
  • the bad pixel position correction module 111 may be used to correct the position of the bad pixel according to the position of the good pixel.
  • the fluorescence molecular center of gravity acquisition module 113 can be used to detect fluorescent molecular image points on the intersection of the first direction and the second direction where the pixel-level registration points are located, and obtain the fluorescence according to the center of gravity method. The position of the center of gravity of the molecular image point.
  • the sub-pixel level registration point acquisition module 115 may be used to fit a straight line in the first direction and the second direction respectively according to the position of the center of gravity of the fluorescent molecular image point, and the fitted straight line in the first direction The intersection with the straight line fitted in the second direction is the sub-pixel position of the registration point.
  • the boundary point position acquiring module 117 may be configured to acquire the boundary point of the fluorescence image according to the straight line fitted in the first direction and the second direction, and calculate the position of the boundary point.
  • the embodiment of the present invention also provides a non-volatile 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 fluorescence image registration method in any of the above embodiments are realized.
  • the integrated modules/units of the gene sequencing system/gene sequencer/computer equipment are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer readable storage medium.
  • the present invention implements all or part of the processes in the above-mentioned implementation methods, and can also be completed by instructing relevant hardware through a computer program.
  • the computer program can be stored in a computer-readable storage medium. When the program is executed by the processor, it can implement the steps of the foregoing method embodiments.
  • the computer program includes computer program code, and the computer program code may be in the form of source code, object code, executable file, or some intermediate forms.
  • the computer-readable storage medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory) ), random access memory (RAM, Random Access Memory), electrical carrier signal, telecommunications signal, and software distribution media.
  • the so-called processor can be a central processing unit (Central Processing Unit, CPU), other general-purpose processors, digital signal processors (Digital Signal Processor, DSP), application specific integrated circuits (Application Specific Integrated Circuit, ASIC), ready-made Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
  • the general-purpose processor can be a microprocessor, or the processor can also be any conventional processor, etc.
  • the processor is the control center of the gene sequencing system/gene sequencer, and various interfaces and lines are used to connect the entire gene sequencing Various parts of the system/gene sequencer.
  • the memory is used to store the computer program and/or module, and the processor realizes the gene by running or executing the computer program and/or module stored in the memory, and calling data stored in the memory.
  • the memory may mainly include a program storage area and a data storage area, where the program storage area may store an operating system, an application program required by at least one function (such as a sound playback function, an image playback function, etc.), and the like.
  • the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disks, memory, plug-in hard disks, Smart Media Card (SMC), Secure Digital (SD) cards , Flash Card, at least one magnetic disk storage device, flash memory device, or other volatile solid-state storage device.
  • non-volatile memory such as hard disks, memory, plug-in hard disks, Smart Media Card (SMC), Secure Digital (SD) cards , Flash Card, at least one magnetic disk storage device, flash memory device, or other volatile solid-state storage device.
  • the disclosed terminal and method may be implemented in other ways.
  • the system implementation described above is only illustrative.
  • the division of the modules is only a logical function division, and there may be other division modes in actual implementation.

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Theoretical Computer Science (AREA)
  • General Health & Medical Sciences (AREA)
  • Organic Chemistry (AREA)
  • Analytical Chemistry (AREA)
  • Proteomics, Peptides & Aminoacids (AREA)
  • Immunology (AREA)
  • Biochemistry (AREA)
  • Zoology (AREA)
  • Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
  • Wood Science & Technology (AREA)
  • Pathology (AREA)
  • Molecular Biology (AREA)
  • Microbiology (AREA)
  • General Engineering & Computer Science (AREA)
  • Biotechnology (AREA)
  • Genetics & Genomics (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Biophysics (AREA)
  • Biomedical Technology (AREA)
  • Geometry (AREA)
  • Investigating, Analyzing Materials By Fluorescence Or Luminescence (AREA)

Abstract

本发明提供一种荧光图像配准方法，其包括获取生物芯片的荧光图像；选取荧光图像的预设局部区域；获取第一方向与第二方向上像素和最小值所在的位置，并得到像素级配准点；根据预设分类规则将像素级配准点分为好点与坏点；根据好点的数量判断荧光图像是否符合预设标准，若符合预设标准，则根据好点的位置校正所述坏点的位置；根据重心法获取荧光分子像点的重心位置；根据荧光分子像点的重心位置分别在第一方向与第二方向上拟合直线，直线的交点即为配准点的亚像素级位置；获取荧光图像的边界点，并计算边界点的位置。本发明实施例还提供一种基因测序系统、基因测序仪与存储介质。利用本发明实施例，可优化荧光图像中荧光分子的定位与配准操作。

Description

荧光图像配准方法、基因测序仪及系统、存储介质 技术领域
本发明涉及基因测序领域，具体的，涉及一种荧光图像配准方法、基因测序仪、基因测序系统以及存储介质。
背景技术
本部分旨在为权利要求书及具体实施方式中陈述的本发明实施例的实施方式提供背景或上下文。此处的描述不因为包括在本部分中就承认是现有技术。
基因测序是指分析特定DNA片段的碱基序列，即腺嘌呤(A),胸腺嘧啶(T)，胞嘧啶(C)与鸟嘌呤(G)的排列方式。目前常用的测序方法之一是：上述四种碱基分别携带四种不同的荧光分子，不同的荧光分子受激发后发射出不同波长(颜色)的荧光，通过识别该荧光波长就能够识别出被合成碱基的类型，从而读取碱基序列。二代测序技术采用高分辨显微成像系统，拍照采集生物芯片(基因测序芯片)上的DNA纳米球分子(即DNB，DNA Nanoballs)的荧光分子图像，将荧光分子图像送入碱基识别软件解码图像信号得到碱基序列。不同荧光染料对一个碱基作用会产生不同波长的荧光分子信号。想要检测出碱基的类别，就需要分析碱基在不同场景下的荧光分子能量，就先要将不同场景的同一位置坐标定位，这就是配准算法的工作。
随着二代测序技术的发展，测序仪产品均配套有测序数据实时处理分析软件，其中大部分配套有配准和定位算法。首先需要将不同碱基信号对应起来，一个碱基信号对应图像中的一个具体位置，需要对碱基在图像中的对应位置进行匹配，因而，匹配的速度、精度和鲁棒性均为图像配准的基本要求。然而，现有的图像配准算法一般是基于CPU设计的，效率较低，随着测序仪通量的增加，处理一张图的时 间也将线性增长，将无法保证二代测序仪的边合成边测序的需求。且该算法其他点的配准情况依赖提取出的最佳点，如果最佳点没有找好将会影响所有点的配准质量。因而亟需提供一种更高性能的图像配准方法。
发明内容
鉴于此，有必要提供一种荧光图像配准方法、基因测序仪、基因测序系统以及存储介质，可优化荧光图像中荧光分子的定位与配准操作。
本发明实施例第一方面提供一种荧光图像配准方法，应用于生物芯片，所述荧光图像配准方法包括：
S1：获取生物芯片的至少一荧光图像；
S2：选取所述荧光图像的预设局部区域，其中，所述预设局部区域在第一方向与第二方向上至少存在一条边界线框，所述边界线框中包含若干荧光分子像点，所述第一方向垂直于所述第二方向；
S3：在所述预设局部区域中，获取所述第一方向与所述第二方向上像素和最小值所在的位置，并根据所述像素和最小值所在的位置得到像素级配准点；
S4：根据预设分类规则将所述像素级配准点分为好点与坏点；
S5：根据所述好点的数量判断所述荧光图像是否符合预设标准，若判定所述荧光图像符合预设标准，则进入步骤S6，若判定所述荧光图像不符合预设标准，则返回步骤S1，重新获取荧光图像；
S6：根据所述好点的位置校正所述坏点的位置；
S7：在所述像素级配准点所在的所述第一方向与所述第二方向的边界线框交叉线上检测荧光分子像点，并根据重心法获取所述荧光分子像点的重心位置；
S8：根据所述荧光分子像点的重心位置分别在所述第一方向与所述第二方向上拟合直线，所述第一方向上拟合出的直线与所述第二方 向上拟合出的直线的交点即为配准点的亚像素级位置；
S9：根据所述第一方向与所述第二方向上拟合出的直线获取所述荧光图像的边界点，并计算所述边界点的位置。
进一步的，在本发明实施例提供的上述荧光图像配准方法中，在获取所述第一方向与所述第二方向上像素和最小值所在的位置之前，所述方法还包括：
在所述预设局部区域内，获取在所述第一方向与所述第二方向上模板边界线框所在的位置，所述模板边界线框为预先设置在所述生物芯片上的模板线框；
根据所述模板边界线框的位置选取预设矩形区域；
在所述预设矩形区域内，获取所述第一方向与所述第二方向上像素和最小值所在的位置。
进一步的，在本发明实施例提供的上述荧光图像配准方法中，所述获取所述第一方向与所述第二方向上像素和最小值所在的位置包括：
选取若干条模板线；
分别在所述第一方向与所述第二方向上依次将所述若干条模板线在所述预设矩形区域进行平移操作；
计算在所述预设矩形区域上若干模板线所在位置覆盖到的像素的亮度值的叠加和，所述亮度值的叠加和即为所述模板线所在位置覆盖到的像素和。
进一步的，在本发明实施例提供的上述荧光图像配准方法中，所述根据预设分类规则将所述像素级配准点分为好点与坏点包括：
在所述预设局部区域的第一方向上，获取任意一列上的所述像素级配准点；
计算其中一个所述像素级配准点与该列上其他所述像素级配准点的横坐标差值的绝对值；
若存在至少2个所述绝对值小于预设横坐标阈值，则判定该像素级配准点为好点；
若所述绝对值大于预设横坐标阈值，则判定该像素级配准点为坏点。
进一步的，在本发明实施例提供的上述荧光图像配准方法中，所述根据预设分类规则将所述像素级配准点分为好点与坏点还包括：
在所述预设局部区域的第二方向上，获取任意一行上的所述像素配准点；
选取其中一个所述像素级配准点，计算该像素级配准点与该行上其他所述像素级配准点的斜率；
将所述与该行上其他所述像素级配准点的斜率按照预设排序规则进行排序，获取第一中位值作为选取的所述像素级配准点的第一斜率；
获取该行上其他所述像素级配准点的第一斜率，并将该行上所有所述像素级配准点的斜率按照所述预设排序规则进行排序，获取第二中位值作为该行的第二斜率，所述第二斜率对应该行上的点为参考点；
获取所述预设局部区域上其他行的第二斜率，并将所有行的斜率按照所述预设排序规则进行排序，获取第三中位值作为所有行的公共斜率；
根据所述公共斜率与所述参考点对所有行进行直线拟合，并根据拟合的直线将所述像素级配准点分为好点与坏点。
进一步的，在本发明实施例提供的上述荧光图像配准方法中，所述根据所述好点的位置校正所述坏点的位置包括：
在所述预设局部区域的第一方向上，获取所述好点的横坐标均值，并将所述横坐标均值赋值给所述坏点；
在所述预设局部区域的第二方向上，获取所述坏点的横坐标，并根据拟合的直线校正所述坏点的纵坐标。
进一步的，在本发明实施例提供的上述荧光图像配准方法中，所述根据重心法获取所述荧光分子像点的重心位置包括：
获取所述像素级配准点的位置坐标；
获取所述荧光分子像点的像素亮度值；
根据预设重心公式获取所述荧光分子像点的重心位置。
本发明实施例第二方面提供一种基因测序系统，应用于生物芯片，所述基因测序系统包括：
荧光图像获取模块，用于获取生物芯片的至少一荧光图像；
局部区域获取模块，用于选取所述荧光图像的预设局部区域，其中，所述预设局部区域在第一方向与第二方向上至少存在一条边界线框，所述边界线框中包含若干荧光分子像点，所述第一方向垂直于所述第二方向；
像素级配准点获取模块，用于在所述预设局部区域中，获取所述第一方向与所述第二方向上像素和最小值所在的位置，并根据所述像素和最小值所在的位置得到像素级配准点；
配准点分类模块，用于根据预设分类规则将所述像素级配准点分为好点与坏点；
图像质量判断模块，用于根据所述好点的数量判断所述荧光图像是否符合预设标准；
坏点位置校正模块，用于根据所述好点的位置校正所述坏点的位置；
荧光分子重心获取模块，用于在所述像素级配准点所在的所述第一方向与所述第二方向的边界线框交叉线上检测荧光分子像点，并根据重心法获取所述荧光分子像点的重心位置；
亚像素级配准点获取模块，用于根据所述荧光分子像点的重心位置分别在所述第一方向与所述第二方向上拟合直线，所述第一方向上拟合出的直线与所述第二方向上拟合出的直线的交点即为配准点的亚像素级位置；
边界点位置获取模块，用于根据所述第一方向与所述第二方向上拟合出的直线获取所述荧光图像的边界点，并计算所述边界点的位置。
本发明实施例第三方面提供一种基因测序仪，所述基因测序仪包 括处理器，所述处理器用于执行存储器中存储的计算机程序时实现上述任意一项所述的荧光图像配准方法。
本发明实施例第四方面提供一种非易失性计算机可读存储介质，其上存储有计算机程序，所述计算机程序被处理器执行时实现上述任意一项所述的荧光图像配准方法。
本发明实施例提供一种荧光图像配准方法、基因测序系统、基因测序仪以及非易失性计算机可读存储介质，获取生物芯片的至少一荧光图像；选取所述荧光图像的预设局部区域，其中，所述预设局部区域在第一方向与第二方向上至少存在一条边界线框，所述边界线框中包含荧光分子像点，所述第一方向垂直于所述第二方向；在所述预设局部区域中，获取所述第一方向与所述第二方向上像素和最小值所在的位置，并根据所述像素和最小值所在的位置得到像素级配准点；根据预设分类规则将所述像素级配准点分为好点与坏点；根据所述好点的数量判断所述荧光图像是否符合预设标准，若判定所述荧光图像符合预设标准，则根据所述好点的位置校正所述坏点的位置；在所述像素级配准点所在的所述第一方向与所述第二方向的边界线框交叉线上检测荧光分子像点，并根据重心法获取所述荧光分子像点的重心位置；根据所述荧光分子像点的重心位置分别在所述第一方向与所述第二方向上拟合直线，所述第一方向上拟合出的直线与所述第二方向上拟合出的直线的交点即为配准点的亚像素级位置；根据所述第一方向与所述第二方向上拟合出的直线获取所述荧光图像的边界点，并计算所述边界点的位置。本发明提出了一种基于点特征的高精度GPU并行荧光图像配准方法，可以对不同测序平台的碱基信息图定位和配准都可以达到较高的精确度和效率，本发明实施例提供的荧光图像配准方法还通过分类的好点的数量判断采集到的荧光图像的质量是否符合标准，舍弃不符合标准的荧光图像，因而具有较强的抗杂质干扰能力。
附图说明
图1是本发明实施例提供荧光图像配准方法的流程图。
图2是本发明实施例提供的荧光图像的预设局部区域示意图。
图3A是第二方向上边界线框像素级定位示意图。
图3B是第二方向上预设矩形区域像素和曲线图。
图4A是荧光图像中边界线框局部放大图。
图4B是图4A所示边界线框上荧光分子像点排列示意图。
图4C是局部区域像素和特征示意图。
图5A是在第一方向上像素级校正示意图。
图5B是在第一方向上极端校正情况示意图。
图6A为像素级位置示意图。
图6B为亚像素级位置示意图。
图7A为亚像素级配准的一原理图。
图7B为亚像素级配准的另一原理图。
图8是GPU编程模型示意图。
图9是本发明一实施方式的基因测序仪的结构示意图。
图10是图9所示的基因测序仪的示例性的功能模块图。
主要元件符号说明
基因测序仪 1
存储器 10
显示屏 20
处理器 30
基因测序系统 100
荧光图像获取模块 101
局部区域获取模块 103
像素级配准点获取模块 105
配准点分类模块 107
图像质量判断模块 109
坏点位置校正模块 111
荧光分子重心获取模块 113
亚像素级配准点获取模块 115
边界点位置获取模块 117
如下具体实施方式将结合上述附图进一步说明本发明实施例。
具体实施方式
为了能够更清楚地理解本发明实施例的上述目的、特征和优点，下面结合附图和具体实施方式对本发明进行详细描述。需要说明的是，在不冲突的情况下，本申请的实施方式中的特征可以相互组合。
在下面的描述中阐述了很多具体细节以便于充分理解本发明实施例，所描述的实施方式仅仅是本发明一部分实施方式，而不是全部的实施方式。基于本发明中的实施方式，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施方式，都属于本发明实施例保护的范围。
除非另有定义，本文所使用的所有的技术和科学术语与属于本发明实施例的技术领域的技术人员通常理解的含义相同。本文中在本发明的说明书中所使用的术语只是为了描述具体的实施方式的目的，不是旨在于限制本发明实施例。
理解本发明部分内容需参阅美国专利US9880089及其同族专利，但是这些内容并不是本发明所要求保护的技术方案，因此文中不重复解释这些技术内容。
图1是本发明实施例提供荧光图像配准方法的流程图。如图1所示，所述荧光图像配准方法可以包括如下步骤：
S1：获取生物芯片的至少一荧光图像。
在本实施方式中，所述生物芯片可以是基因测序芯片，所述荧光图像可以是测序时拍摄的荧光信号图像。所述生物芯片在第一方向与第二方向上存在若干条轨迹线(Trackline，如美国专利US9880089中的track region)，其中，所述第一方向与所述第二方向垂直，所述第一方向可以是垂直方向，所述第二方向可以是水平方向。相邻2条第一和第二方向上的轨迹线之间形成的区域称为一个区块(block)，区块分为内部区块和外部区块。所述生物芯片上每个区块内均匀分布有若干个位点，所述位点可以吸附DNA纳米球分子(DNB)，所述DNA纳米球分子可以是包括DNA片段的扩增产物。所述DNA纳米球分子在合成碱基时携带有荧光分子，荧光分子受激发时会发出荧光信号。所述荧光分子可以按照预设规则固定排列在所述生物芯片上，通过特殊的设计和处理，所述生物芯片上某些位置没有位点存在，也即没有荧光分子存在。在所述荧光分子大于25％(腺嘌呤(A),胸腺嘧啶(T)，胞嘧啶(C)与鸟嘌呤(G)四种碱基均衡)随机位置发光的情况下，不发光的边界线框就凸显出来。凸显出来的所述边界线框可以由三个荧光分子像点位置组成，中间排的荧光分子像点为亮点，中间排的两边位置的荧光分子像点均不亮。中间排的荧光分子像点形成轨迹线，中间排的两边位置的荧光分子像点形成暗线。荧光分子成像的点成为荧光分子像点。可以理解的是，所述边界线框可以包括所述轨迹线和所述轨迹线两侧的暗线。
在测序过程中，可以利用TDI相机对所述生物芯片按照预设扫描顺序进行扫描获得荧光信号图像。所述预设扫描顺序可以是终端用户预先设置的，所述预设扫描顺序包括，但不限于按照自上而下的顺序进行扫描。
S2：选取所述荧光图像的预设局部区域，其中，所述预设局部区域在第一方向与第二方向上至少存在一条边界线框，所述边界线框中包含荧光分子像点，所述第一方向垂直于所述第二方向。
在本实施方式中，所述预设局部区域可以是终端用户预先设置 的，所述预设局部区域在所述第一方向与所述第二方向上至少存在一条边界线框，也即所述预设局部区域内至少存在一个配准点。
请参阅图2，图2是本发明实施例提供的荧光图像的预设局部区域示意图。如图2所示，所述预设局部区域可以是9*9的区域，所述第一方向上存在9条边界线框(图2中用黑色实线表示)，所述第二方向上存在9条边界线框(图2中用黑色实线表示)，所述第一方向上的线框与所述第二方向上的线框相交，得到81(9*9)个交点(配准点)。图2中的白色实线表示模板边界线框，所述模板边界线框为预先设置在所述生物芯片上的模板线框。图2中四个边界上均存在标注有星号的位置，标注白色星号的位置即为荧光图像的边界点位置。
S3：在所述预设局部区域中，获取所述第一方向与所述第二方向上像素和最小值所在的位置，并根据所述像素和最小值所在的位置得到像素级配准点。
在本实施方式中，在所述获取所述第一方向与所述第二方向上像素和最小值所在的位置之前，所述方法还包括：在所述预设局部区域内，获取所述第一方向与所述第二方向上模板边界线框所在的位置，所述模板边界线框为预先设置在所述生物芯片上的模板线框，它是生物芯片的边界线框实际物理位置，由于相机成像会产生畸变，所以荧光信号图像的边界线框与模板边界线框之间会有偏差；根据所述模板边界线框的位置选取预设矩形区域；在所述预设矩形区域内，获取所述第一方向与所述第二方向上像素和最小值所在的位置。其中，所述预设矩形区域为终端用户预先设置的。根据所述第一方向与所述第二方向上模板边界线框所在的位置，可以获得所述预设局部区域内模板配准点的位置。所述预设矩形区域中包含一模板配准点，优选的，所述预设矩形区域可以相对于所述模板配准点的位置呈中心对称设置。可以理解的是，对于所述第一方向上的某一列上，需对该列上每一模板配准点所在的模板线框位置选取预设矩形区域，并在每一所述预设矩形区域内，获取所述第一方向上像素和最小值所在的位置，从而得到该纵向列上的实际的边界线框所在的位置。
需要说明的是，本文中所称的“像素和”指的是像素的亮度值之和。“第一方向与第二方向上像素和最小值所在的位置”指的就是该方向上整排像素的亮度值之和最小的那一排。
请参阅图3A、图3B，图3A是第二方向上边界线框像素级定位示意图，图3B是第二方向上预设矩形区域像素和曲线图。如图3A所示，图中存在两种边界线框，黑色表示实际边界线框，白色表示模板边界线框。以第二方向为例，获取所述第二方向上的模板边界线框所在的位置，根据所述模板边界线框所在的位置选取预设矩形区域，图3A中每个模板边界线框的位置都存在一个预设矩形区域(所述预设矩形区域用白色实线表示)，预设矩形区域的中心一般是第一方向上的模板边界线框和第二方向上的模板边界线框的交点。为了避免在获取所述第一方向与所述第二方向上像素和最小值所在的位置时，搜索的边界线框发生重叠，所述预设矩形区域的长度与宽度均不能大于模板边界线框的最短距离。且所述预设矩形区域面积设置较大时，会影响程序运行速度；所述预设矩形区域面积设置较小时，会影响暗线定位精度。因而，在本实施方式中，所述预设矩形区域的大小设置为128*128像素，能够同时保证了程序运行速度和定位精度。
所述获取所述第一方向与所述第二方向上像素和最小值所在的位置包括：选取若干条模板线；分别在所述第一方向与所述第二方向上依次将所述若干条模板线在所述预设矩形区域进行平移操作；计算所述预设矩形区域上若干模板线所在位置覆盖到的像素的亮度值的叠加和，所述亮度值的叠加和即为所述模板线所在位置覆盖到的像素和。所述根据所述像素和最小值所在的位置得到像素级配准点包括：所述第一方向上像素和最小值所在的位置与所述第二方向上像素和最小值所在的位置的交点即为像素级配准点。如图3B所示，所述边界线框所在位置朝向的所有像素的亮度值之和保持最低，图3B中的所述边界线框表现出明显的波谷。图3B中每个波谷都对应图3A中边界线框所在位置。
请参阅图4A、4B以及4C，图4A是荧光图像中边界线框局部放 大图，图4B是图4A所示边界线框上荧光分子像点排列示意图，图4C是局部区域像素和特征示意图。如图4A以及4B所示，图4A中在所述第一方向与所述第二方向上分别有两条边界线框(箭头指向的位置是边界线框所在的位置)，凸显出来的所述边界线框可以由三个荧光分子像点位置组成，中间排的荧光分子像点为亮点，中间排的两边位置的荧光分子像点均不亮。中间排的荧光分子像点形成轨迹线，中间排的两边位置的荧光分子像点形成暗线。可以理解的是，凸显出来的所述边界线框可以包括所述轨迹线和所述轨迹线两侧的暗线。可以理解的是，不同生物芯片、不同相机对应的荧光分子大小不一样，本发明以大小为3*3像素为例，如图4C所示，所述边界线框附近出现W型线特征(黑色椭圆圈选中的部分)，W型线特征中有两处波谷位置以及一处波峰位置。可以理解的是，两处波谷位置为所述生物芯片上所述轨迹线两侧暗线对应的位置，暗线对应的位置的像素和值较低。W型线特征中波峰对应的位置为所述轨迹线对应的位置，所述轨迹线对应的位置的像素和值较高。根据这一特征，对所述预设矩形区域内每个点两边(-4)(-3)(-2)和(2)(3)(4)位置的像素和累加，累加和最小的点对应边界线框的位置。使用同样的方法可以求得第一方向上边界线框的位置，所述第一方向上各个边界线框与所述第二方向上各个边界线框的交点即为像素级配准点，也即共81(9*9)个。
S4：根据预设分类规则将所述像素级配准点分为好点与坏点。
在本实施方式中，在获得像素级配准点之后，需要对每个像素级配准点进行好坏点判定和校正。本发明实施例提供一种直线拟合方法来判定每个像素级配准点的质量，同时对坏点进行校正，即边判定边校正。最后综合所有像素级配准点的情况来判定整张荧光图像的质量。具体过程如下：
所述根据预设分类规则将所述像素配准点分为好点与坏点包括：在所述预设局部区域的第一方向上，获取任意一列上的所述像素级配准点；计算其中一个所述像素级配准点与该列上其他所述像素级配准 点的横坐标差值的绝对值；若存在至少2个所述绝对值小于预设横坐标阈值，则判定该像素级配准点为好点，否则，则判定该像素级配准点为坏点。所述预设横坐标阈值可以是本领域技术人员根据经验值预先设置的。在本实施方式中，由于在亚像素级配准时，在边界线框附近搜索荧光分子的范围是3像素。只要在3像素范围内，通过亚像素配准可以定位到最佳位置。因而所述预设横坐标阈值为3像素。
在所述预设局部区域的第二方向上，获取任意一行上的所述像素配准点；选取其中一个所述像素级配准点，计算该像素级配准点与该行上其他所述像素级配准点的斜率；将所述与该行上其他所述像素级配准点的斜率按照预设排序规则(所述预设排列顺序可以是从小到大的顺序，或者是从大到小的顺序)进行排序，获取第一中位值作为选取的所述像素级配准点的第一斜率；获取该行上其他所述像素级配准点的第一斜率，并将该行上所有所述像素级配准点的斜率按照所述预设排序规则进行排序，获取第二中位值作为该行的第二斜率，所述第二斜率对应该行上的点为参考点；获取所述预设局部区域上其他行的第二斜率，并将所有行的斜率按照所述预设排序规则进行排序，获取第三中位值作为所有行的公共斜率；根据所述公共斜率与所述参考点对所有行进行直线拟合，并根据拟合的直线将所述像素级配准点分为好点与坏点。具体的，当像素级配准点与直线在第一方向上的距离大于3个像素时，判定该配准点为坏点，也就是说，由于先前已经把第一方向的边界线框(轨迹线)的像素级横坐标已经确定，根据该横坐标获得上述第二方向(斜率)直线对应的纵坐标值，再将像素级配准点与该纵坐标值作差取绝对值，如果绝对值大于3个像素，那么判定该配准点为坏点。
S5：根据所述好点的数量判断所述荧光图像是否符合预设标准，若判定所述荧光图像符合预设标准，则进入步骤S6，若判定所述荧光图像不符合预设标准，则返回步骤S1，重新获取荧光图像。
在本实施方式中，所述根据所述好点的数量判断所述荧光图像是否符合预设标准包括：在所述第一方向上，若每一列上好点的个数大 于3，则判定所述荧光图像符合预设标准，否则，丢弃所述荧光图像。在所述第二方向上，若每一行上好点的个数大于预设个数，则判定所述荧光图像符合预设标准，否则，丢弃所述荧光图像，执行步骤S1，重新选取应该图像。所述预设个数为根据选取的所述预设局部区域大小预先调整设置好的。在本实施方式中，所述预设局部区域为9*9的区域，则所述预设个数为5个。也即，在所述第二方向上，若每一行上好点的个数大于5，则判定所述荧光图像符合预设标准，否则，丢弃所述荧光图像。
S6：根据所述好点的位置校正所述坏点的位置。
在本实施方式中，以利用TDI相机对所述生物芯片按照自上而下的扫描顺序进行扫描获得荧光信号图像为例，由于TDI模式的光学特征，在所述第一方向上的边界线框始终保持为垂直方向。因而可以先对所述第一方向上的坏点进行校正，所述第二方向上坏点的校正有依赖于所述第一方向上校正后的坏点的横坐标。
所述根据所述好点的位置校正所述坏点的位置包括：在所述预设局部区域的第一方向上，获取所述好点的横坐标均值，并将所述横坐标均值赋值给所述坏点；在所述预设局部区域的第二方向上，获取所述坏点的横坐标，并根据拟合的直线校正所述坏点的纵坐标。可以理解的是，在对所述预设局部区域第一方向上的坏点进行校正后，所述像素级配准点的横坐标可以确定。因而，在所述预设局部区域第二方向上，可以根据确定的坏点的横坐标，通过拟合的直线校正所述坏点的纵坐标。
请参阅图5A以及图5B，图5A是在第一方向上像素级校正示意图，图5B是在第一方向上极端校正情况示意图。如图5A所示，在所述第一方向上存在两条直线，左边直线为校正前的直线，左边直线上存在三个好点(用灰色实心圆表示好点)，左边直线附近存在六个需要校正的坏点(用黑色实心圆表示坏点)。通过获取好点的横坐标均值，并将横坐标均值赋值给所述坏点，可以将所述坏点进行校正，从而得到右边的直线，右边的直线为校正后的直线，右边的直线上存 在9个好点。可以理解的是，还存在一种较极端的情况，如图5B所示，左边的直线上存在9个像素级配准点，其中，每三个点分别在同一条直线上。此时，若按照上述好点判定方法，则9个点均为好点。然而这9个像素级配准点并不在同一直线上，在不依赖模板边界线框的情况下，无法确定哪三个点所在的直线是边界线框所在的真实位置，如右边直线所示，校正前后的点分布情况未发生变化。为了避免这种情况，所述方法还包括：获取9个像素级配准点中任意两点的横坐标差值；判断所述横坐标差值是否大于预设横坐标阈值，若判定所述横坐标差值大于预设横坐标阈值，则认为所述荧光图像质量较差，舍弃该荧光图像。
S7：在所述像素级配准点所在的所述第一方向与所述第二方向的边界线框交叉线上检测荧光分子像点，并根据重心法获取所述荧光分子像点的重心位置。
在本实施方式中，在获取所述像素级配准点后，需要对所述像素级配准点进行更高精度的定位，即亚像素级配准。所述根据重心法获取所述荧光分子像点的重心位置包括：获取所述像素级配准点的位置坐标；获取所述荧光分子像点的像素亮度值；根据预设重心公式获取所述荧光分子像点的重心位置。
边界线框交叉线上的荧光分子像点实际上就是轨迹线(Trackline)上的编码位点(如美国专利US9880089中track region上的fluorescent spots)。
请参阅图6A以及图6B，图6A为像素级位置示意图，图6B为亚像素级位置示意图。如图6A所示，黑色线框区域表示一个像素，将像素再向下细分，则得到亚像素位置，如图6B中所示的灰色实心圆点。在像素级配准点所在交叉线上检测荧光分子像点，交叉线的范围设置为-64到64像素，也即长度为128像素，这边和预设矩形区域的长度大小设置一致。荧光分子像点的检测条件如图6A所示，在3*3像素范围内，①位置的亮度值最高且大于图像背景值，②③④⑤位置的亮度值高于⑥⑦⑧⑨位置的亮度值。再采用重心法获取荧光分子像 点的重心位置，重心法的公式如下：
X＝X 1+(H 3-H 5)/(H 1+H 3+H 5-H b，q×3)
Y＝Y 1+(H 4-H 2)/(H 1+H 2+H 4-H b，q×3）
上式中，(X 1，Y 1)表示①的像素级坐标，H 1～５分别表示编号对应位置的亮度值，H b，q为图像背景值。
S8：根据所述荧光分子像点的重心位置分别在所述第一方向与所述第二方向上拟合直线，所述第一方向上拟合出的直线与所述第二方向上拟合出的直线的交点即为配准点的亚像素级位置。
在本实施方式中，根据获取的所述荧光分子像点的重心位置分别在所述第一方向与所述第二方向上通过最小二乘法拟合直线，所述第一方向上拟合出的直线与所述第二方向上拟合出的直线的交点即为配准点的亚像素级位置。
请参阅图7A以及图7B，图7A为亚像素级配准的一原理图，图7B为亚像素级配准的另一原理图。将检测到的荧光分子像点重心位置分别在所述第一方向与所述第二方向上通过最小二乘法拟合直线，直线的交点即为配准点的亚像素级位置。在图7B中，黑色实心圆表示亚像素级配准点，灰色实心圆表示荧光分子像点的重心位置。
S9：根据所述第一方向与所述第二方向上拟合出的直线获取所述荧光图像的边界点，并计算所述边界点的位置。
在本实施方式中，根据所述第一方向与所述第二方向上拟合出的直线获取所述荧光图像的边界点，并计算所述边界点的位置。请参阅图2，图2中四个边界上均存在标注有星号的位置，标注星号的位置即为荧光图像的边界点位置，依据亚像素级配准点校正时拟合好的直线可以求出荧光图像边界上的点。可以理解的是，由于所述边界点的一个坐标值已知(例如，最左边的一列边界点的横坐标均为0)，所以只需将已知的坐标值代入拟合好的直线中即可求出另一个坐标值，从而得到边界点的位置。可以理解的是，对于预设局部区域为9*9的区域，通过上述方法一共可以获取到121(11*11)个亚像素级配准点。
本发明实施例提供一种荧光图像配准方法，获取生物芯片的至少 一荧光图像；选取所述荧光图像的预设局部区域，其中，所述预设局部区域在第一方向与第二方向上至少存在一条边界线框，所述边界线框中包含荧光分子像点，所述第一方向垂直于所述第二方向；在所述预设局部区域中，获取所述第一方向与所述第二方向上像素和最小值所在的位置，并根据所述像素和最小值所在的位置得到像素级配准点；根据预设分类规则将所述像素级配准点分为好点与坏点；根据所述好点的数量判断所述荧光图像是否符合预设标准，若判定所述荧光图像符合预设标准，则根据所述好点的位置校正所述坏点的位置；在所述像素级配准点所在的所述第一方向与所述第二方向的边界线框交叉线上检测荧光分子像点，并根据重心法获取所述荧光分子像点的重心位置；根据所述荧光分子像点的重心位置分别在所述第一方向与所述第二方向上拟合直线，所述第一方向上拟合出的直线与所述第二方向上拟合出的直线的交点即为配准点的亚像素级位置；根据所述第一方向与所述第二方向上拟合出的直线获取所述荧光图像的边界点，并计算所述边界点的位置。本发明提出了一种基于点特征的高精度GPU并行荧光图像配准方法，可以对不同测序平台的碱基信息图定位和配准都可以达到较高的精确度和效率，本发明实施例提供的荧光图像配准方法还通过分类的好点的数量判断采集到的荧光图像的质量是否符合标准，舍弃不符合标准的荧光图像，因而具有较强的抗杂质干扰能力。
请参阅图8，图8为GPU编程模型示意图。本发明实施例提供的荧光图像配准方法可以基于GPU设计，从而提高荧光图像配准速率。如图8所示，从左数起第1个框指的是CPU，第2个和第3个框指的是GPU。主机上包含三个核函数，分别为核函数1、核函数2以及核函数3，每个运行的核函数包含一个线程网络(grid)，一个线程网络内可以设置有多个线程块(block)，一个线程块内可以设置有多个线程(thread)。线程块之间可以理解为粗粒度并行，线程之间可以理解为细粒度并行。在本发明实施例提供的荧光图像配准方法设计中，粗粒度并行可以使得多张荧光图像和每张荧光图像的多个配准点 并行处理，细粒度并行可以使得每个配准点像素级配准、校正等过程做到并行。在本实施方式中，在像素级配准过程中，可以在GPU上设计使用9*9个线程块，分别负责每个区域的配准。每个线程块内可以设置有512个线程，分别负责第二方向和第一方向在第一方向和第二方向上的像素和累加，最后得到每个方向上波谷位置。后续荧光图像配准步骤中使用线程块和线程的数量设计思路与此一致，将不再详述。
以上是对本发明实施例所提供的方法进行的详细描述。下面对本发明实施例所提供的基因测序仪进行描述。
本发明实施例还提供一种基因测序仪，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机程序，所述处理器执行所述程序时实现上述任一实施方式中所述的荧光图像配准方法的步骤。需要说明的是，所述基因测序仪可以包括芯片平台、光学系统、液路系统。其中，所述芯片平台可以用于装载生物芯片，所述光学系统可以用于获取荧光图像，所述液路系统可以用于利用预设的试剂进行生化反应。
图9是本发明一实施方式的基因测序仪的结构示意图。如图9所示，基因测序仪1包括存储器10，存储器10中存储有基因测序系统100。所述基因测序系统100可以获取生物芯片的至少一荧光图像；选取所述荧光图像的预设局部区域，其中，所述预设局部区域在第一方向与第二方向上至少存在一条边界线框，所述边界线框中包含荧光分子像点，所述第一方向垂直于所述第二方向；在所述预设局部区域中，获取所述第一方向与所述第二方向上像素和最小值所在的位置，并根据所述像素和最小值所在的位置得到像素级配准点；根据预设分类规则将所述像素级配准点分为好点与坏点；根据所述好点的数量判断所述荧光图像是否符合预设标准，若判定所述荧光图像符合预设标准，则根据所述好点的位置校正所述坏点的位置；在所述像素级配准点所在的所述第一方向与所述第二方向的边界线框交叉线上检测荧光分子像点，并根据重心法获取所述荧光分子像点的重心位置；根据 所述荧光分子像点的重心位置分别在所述第一方向与所述第二方向上拟合直线，所述第一方向上拟合出的直线与所述第二方向上拟合出的直线的交点即为配准点的亚像素级位置；根据所述第一方向与所述第二方向上拟合出的直线获取所述荧光图像的边界点，并计算所述边界点的位置。本发明提出了一种基于点特征的高精度GPU并行荧光图像配准方法，可以对不同测序平台的碱基信息图定位和配准都可以达到较高的精确度和效率，本发明实施例提供的荧光图像配准方法还通过分类的好点的数量判断采集到的荧光图像的质量是否符合标准，舍弃不符合标准的荧光图像，因而具有较强的抗杂质干扰能力。
本实施方式中，基因测序仪1还可以包括显示屏20及处理器30。存储器10、显示屏20可以分别与处理器30电连接。
所述的存储器10可以是不同类型存储设备，用于存储各类数据。例如，可以是基因测序仪1的存储器、内存，还可以是可外接于该基因测序仪1的存储卡，如闪存、SM卡(Smart Media Card，智能媒体卡)、SD卡(Secure Digital Card，安全数字卡)等。此外，存储器10可以包括高速随机存取存储器，还可以包括非易失性存储器，例如硬盘、内存、插接式硬盘，智能存储卡(Smart Media Card,SMC)，安全数字(Secure Digital,SD)卡，闪存卡(Flash Card)、至少一个磁盘存储器件、闪存器件、或其他易失性固态存储器件。存储器10用于存储各类数据，例如，所述基因测序仪1中安装的各类应用程序(Applications)、应用上述荧光图像配准方法而设置、获取的数据等信息。
显示屏20安装于基因测序仪1，用于显示信息。
处理器30用于执行所述荧光图像配准方法以及所述基因测序仪1内安装的各类软件，例如操作系统及应用显示软件等。处理器30包含但不限于处理器(Central Processing Unit，CPU)、微控制单元(Micro Controller Unit，MCU)等用于解释计算机指令以及处理计算机软件中的数据的装置。
所述的基因测序系统100可以包括一个或多个的模块，所述一个 或多个模块被存储在基因测序仪1的存储器10中并被配置成由一个或多个处理器(本实施方式为一个处理器30)执行，以完成本发明实施例。例如，参阅图10所示，所述基因测序系统100可以包括荧光图像获取模块101、局部区域获取模块103、像素级配准点获取模块105、配准点分类模块107、图像质量判断模块109、坏点位置校正模块111、荧光分子重心获取模块113、亚像素级配准点获取模块115、边界点位置获取模块117。本发明实施例所称的模块可以是完成一特定功能的程序段，比程序更适合于描述软件在处理器中的执行过程。
可以理解的是，对应上述荧光图像配准方法中的各实施方式，基因测序仪1可以包括图10中所示的各功能模块中的一部分或全部，各模块的功能将在以下具体介绍。需要说明的是，以上荧光图像配准方法的各实施方式中相同的名词相关名词及其具体的解释说明也可以适用于以下对各模块的功能介绍。为节省篇幅及避免重复起见，在此就不再赘述。
荧光图像获取模块101可以用于获取生物芯片的至少一荧光图像。
局部区域获取模块103可以用于选取所述荧光图像的预设局部区域，其中，所述预设局部区域在第一方向与第二方向上至少存在一条边界线框，所述边界线框中包含若干荧光分子像点，所述第一方向垂直于所述第二方向。
像素级配准点获取模块105可以用于在所述预设局部区域中，获取所述第一方向与所述第二方向上像素和最小值所在的位置，并根据所述像素和最小值所在的位置得到像素级配准点。
配准点分类模块107可以用于根据预设分类规则将所述像素级配准点分为好点与坏点。
图像质量判断模块109可以用于根据所述好点的数量判断所述荧光图像是否符合预设标准。
坏点位置校正模块111可以用于根据所述好点的位置校正所述 坏点的位置。
荧光分子重心获取模块113可以用于在所述像素级配准点所在的所述第一方向与所述第二方向的边界线框交叉线上检测荧光分子像点，并根据重心法获取所述荧光分子像点的重心位置。
亚像素级配准点获取模块115可以用于根据所述荧光分子像点的重心位置分别在所述第一方向与所述第二方向上拟合直线，所述第一方向上拟合出的直线与所述第二方向上拟合出的直线的交点即为配准点的亚像素级位置。
边界点位置获取模块117可以用于根据所述第一方向与所述第二方向上拟合出的直线获取所述荧光图像的边界点，并计算所述边界点的位置。
本发明实施例还提供一种非易失性计算机可读存储介质，其上存储有计算机程序，所述计算机程序被处理器执行时实现上述任一实施方式中的荧光图像配准方法的步骤。
所述基因测序系统/基因测序仪/计算机设备集成的模块/单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本发明实现上述实施方式方法中的全部或部分流程，也可以通过计算机程序来指令相关的硬件来完成，所述的计算机程序可存储于一计算机可读存储介质中，该计算机程序在被处理器执行时，可实现上述各个方法实施例的步骤。其中，所述计算机程序包括计算机程序代码，所述计算机程序代码可以为源代码形式、对象代码形式、可执行文件或某些中间形式等。所述计算机可读存储介质可以包括：能够携带所述计算机程序代码的任何实体或装置、记录介质、U盘、移动硬盘、磁碟、光盘、计算机存储器、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、电载波信号、电信信号以及软件分发介质等。
所称处理器可以是中央处理单元(Central Processing Unit，CPU)，还可以是其他通用处理器、数字信号处理器(Digital Signal Processor， DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现成可编程门阵列(Field-Programmable Gate Array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等，所述处理器是所述基因测序系统/基因测序仪的控制中心，利用各种接口和线路连接整个基因测序系统/基因测序仪的各个部分。
所述存储器用于存储所述计算机程序和/或模块，所述处理器通过运行或执行存储在所述存储器内的计算机程序和/或模块，以及调用存储在存储器内的数据，实现所述基因测序系统/基因测序仪的各种功能。所述存储器可主要包括存储程序区和存储数据区，其中，存储程序区可存储操作系统、至少一个功能所需的应用程序(比如声音播放功能、图像播放功能等)等。此外，存储器可以包括高速随机存取存储器，还可以包括非易失性存储器，例如硬盘、内存、插接式硬盘，智能存储卡(Smart Media Card,SMC)，安全数字(Secure Digital,SD)卡，闪存卡(Flash Card)、至少一个磁盘存储器件、闪存器件、或其他易失性固态存储器件。
在本发明所提供的几个具体实施方式中，应该理解到，所揭露的终端和方法，可以通过其它的方式实现。例如，以上所描述的系统实施方式仅仅是示意性的，例如，所述模块的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式。
对于本领域技术人员而言，显然本发明实施例不限于上述示范性实施例的细节，而且在不背离本发明实施例的精神或基本特征的情况下，能够以其他的具体形式实现本发明实施例。因此，无论从哪一点来看，均应将实施例看作是示范性的，而且是非限制性的，本发明实施例的范围由所附权利要求而不是上述说明限定，因此旨在将落在权利要求的等同要件的含义和范围内的所有变化涵括在本发明实施例内。不应将权利要求中的任何附图标记视为限制所涉及的权利要求。系统、装置或终端权利要求中陈述的多个单元、模块或装置也可以由同一个单元、模块或装置通过软件或者硬件来实现。
以上实施方式仅用以说明本发明实施例的技术方案而非限制，尽管参照以上较佳实施方式对本发明实施例进行了详细说明，本邻域的普通技术人员应当理解，可以对本发明实施例的技术方案进行修改或等同替换都不应脱离本发明实施例的技术方案的精神和范围。

Claims (10)

  1. 一种荧光图像配准方法，应用于生物芯片，其特征在于，所述荧光图像配准方法包括：
    S1：获取生物芯片的至少一荧光图像；
    S2：选取所述荧光图像的预设局部区域，其中，所述预设局部区域在第一方向与第二方向上至少存在一条边界线框，所述边界线框中包含若干荧光分子像点，所述第一方向垂直于所述第二方向；
    S3：在所述预设局部区域中，获取所述第一方向与所述第二方向上像素和最小值所在的位置，并根据所述像素和最小值所在的位置得到像素级配准点；
    S4：根据预设分类规则将所述像素级配准点分为好点与坏点；
    S5：根据所述好点的数量判断所述荧光图像是否符合预设标准，若判定所述荧光图像符合预设标准，则进入步骤S6，若判定所述荧光图像不符合预设标准，则返回步骤S1，重新获取荧光图像；
    S6：根据所述好点的位置校正所述坏点的位置；
    S7：在所述像素级配准点所在的所述第一方向与所述第二方向的边界线框交叉线上检测荧光分子像点，并根据重心法获取所述荧光分子像点的重心位置；
    S8：根据所述荧光分子像点的重心位置分别在所述第一方向与所述第二方向上拟合直线，所述第一方向上拟合出的直线与所述第二方向上拟合出的直线的交点即为配准点的亚像素级位置；
    S9：根据所述第一方向与所述第二方向上拟合出的直线获取所述荧光图像的边界点，并计算所述边界点的位置。
  2. 根据权利要求1所述的荧光图像配准方法，其特征在于，在获取所述第一方向与所述第二方向上像素和最小值所在的位置之前，所述方法还包括：
    在所述预设局部区域内，获取在所述第一方向与所述第二方向上模板边界线框所在的位置，所述模板边界线框为预先设置在所述生物芯片上的模板线框；
    根据所述模板边界线框的位置选取预设矩形区域；
    在所述预设矩形区域内，获取所述第一方向与所述第二方向上像素和最小值所在的位置。
  3. 根据权利要求2所述的荧光图像配准方法，其特征在于，所述获取所述第一方向与所述第二方向上像素和最小值所在的位置包括：
    选取若干条模板线；
    分别在所述第一方向与所述第二方向上依次将所述若干条模板线在所述预设矩形区域进行平移操作；
    计算在所述预设矩形区域上若干模板线所在位置覆盖到的像素的亮度值的叠加和，所述亮度值的叠加和即为所述模板线所在位置覆盖到的像素和。
  4. 根据权利要求1所述的荧光图像配准方法，其特征在于，所述根据预设分类规则将所述像素级配准点分为好点与坏点包括：
    在所述预设局部区域的第一方向上，获取任意一列上的所述像素级配准点；
    计算其中一个所述像素级配准点与该列上其他所述像素级配准点的横坐标差值的绝对值；
    若存在至少2个所述绝对值小于预设横坐标阈值，则判定该像素级配准点为好点。
  5. 根据权利要求4所述的荧光图像配准方法，其特征在于，所述根据预设分类规则将所述像素级配准点分为好点与坏点还包括：
    在所述预设局部区域的第二方向上，获取任意一行上的所述像素配准点；
    选取其中一个所述像素级配准点，计算该像素级配准点与该行上其他所述像素级配准点的斜率；
    将所述与该行上其他所述像素级配准点的斜率按照预设排序规则进行排序，获取第一中位值作为选取的所述像素级配准点的第一斜率；
    获取该行上其他所述像素级配准点的第一斜率，并将该行上所有所述像素级配准点的斜率按照所述预设排序规则进行排序，获取第二中位值作为该行的第二斜率，所述第二斜率对应该行上的点为参考点；
    获取所述预设局部区域上其他行的第二斜率，并将所有行的斜率按照所述预设排序规则进行排序，获取第三中位值作为所有行的公共斜率；
    根据所述公共斜率与所述参考点对所有行进行直线拟合，并根据拟合的直线将所述像素级配准点分为好点与坏点。
  6. 根据权利要求5所述的荧光图像配准方法，其特征在于，所述根据所述好点的位置校正所述坏点的位置包括：
    在所述预设局部区域的第一方向上，获取所述好点的横坐标均值，并将所述横坐标均值赋值给所述坏点；
    在所述预设局部区域的第二方向上，获取所述坏点的横坐标，并根据拟合的直线校正所述坏点的纵坐标。
  7. 根据权利要求5所述的荧光图像配准方法，其特征在于，所述根据重心法获取所述荧光分子像点的重心位置包括：
    获取所述像素级配准点的位置坐标；
    获取所述荧光分子像点的像素亮度值；
    根据预设重心公式获取所述荧光分子像点的重心位置。
  8. 一种基因测序系统，应用于生物芯片，其特征在于，所述基因测序系统包括：
    荧光图像获取模块，用于获取生物芯片的至少一荧光图像；
    局部区域获取模块，用于选取所述荧光图像的预设局部区域，其中，所述预设局部区域在第一方向与第二方向上至少存在一条边界线框，所述边界线框中包含若干荧光分子像点，所述第一方向垂直于所述第二方向；
    像素级配准点获取模块，用于在所述预设局部区域中，获取所述第一方向与所述第二方向上像素和最小值所在的位置，并根据所述像 素和最小值所在的位置得到像素级配准点；
    配准点分类模块，用于根据预设分类规则将所述像素级配准点分为好点与坏点；
    图像质量判断模块，用于根据所述好点的数量判断所述荧光图像是否符合预设标准；
    坏点位置校正模块，用于根据所述好点的位置校正所述坏点的位置；
    荧光分子重心获取模块，用于在所述像素级配准点所在的所述第一方向与所述第二方向的边界线框交叉线上检测荧光分子像点，并根据重心法获取所述荧光分子像点的重心位置；
    亚像素级配准点获取模块，用于根据所述荧光分子像点的重心位置分别在所述第一方向与所述第二方向上拟合直线，所述第一方向上拟合出的直线与所述第二方向上拟合出的直线的交点即为配准点的亚像素级位置；
    边界点位置获取模块，用于根据所述第一方向与所述第二方向上拟合出的直线获取所述荧光图像的边界点，并计算所述边界点的位置。
  9. 一种基因测序仪，其特征在于，所述基因测序仪包括处理器，所述处理器用于执行存储器中存储的计算机程序时实现如权利要求1-7任意一项所述的荧光图像配准方法。
  10. 一种非易失性计算机可读存储介质，其上存储有计算机程序，其特征在于，所述计算机程序被处理器执行时实现如权利要求1-7任意一项所述的荧光图像配准方法。
PCT/CN2019/074244 2019-01-31 2019-01-31 荧光图像配准方法、基因测序仪及系统、存储介质 Ceased WO2020155043A1 (zh)

Priority Applications (3)

Application Number Priority Date Filing Date Title
CN201980002455.8A CN112204615B (zh) 2019-01-31 2019-01-31 荧光图像配准方法、基因测序仪及系统、存储介质
PCT/CN2019/074244 WO2020155043A1 (zh) 2019-01-31 2019-01-31 荧光图像配准方法、基因测序仪及系统、存储介质
US17/426,885 US11972583B2 (en) 2019-01-31 2019-01-31 Fluorescence image registration method, gene sequencing instrument, and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2019/074244 WO2020155043A1 (zh) 2019-01-31 2019-01-31 荧光图像配准方法、基因测序仪及系统、存储介质

Publications (1)

Publication Number Publication Date
WO2020155043A1 true WO2020155043A1 (zh) 2020-08-06

Family

ID=71841504

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/074244 Ceased WO2020155043A1 (zh) 2019-01-31 2019-01-31 荧光图像配准方法、基因测序仪及系统、存储介质

Country Status (3)

Country Link
US (1) US11972583B2 (zh)
CN (1) CN112204615B (zh)
WO (1) WO2020155043A1 (zh)

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114549600A (zh) * 2022-01-28 2022-05-27 赛纳生物科技(北京)有限公司 一种荧光图像配准方法
CN115602245A (zh) * 2022-09-09 2023-01-13 郑州思昆生物工程有限公司（Cn） 一种荧光图筛选的方法、装置、设备以及存储介质
WO2024001051A1 (zh) * 2022-06-29 2024-01-04 深圳华大生命科学研究院 空间组学单细胞数据获取方法及装置和电子设备

Families Citing this family (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114845260B (zh) * 2022-04-22 2024-05-24 华北水利水电大学 一种基于物联网的水文监测数据采集系统
CN115187643B (zh) * 2022-06-20 2023-05-09 深圳赛陆医疗科技有限公司 图像配准和模板构建方法、装置、电子设备及存储介质
CN119343698A (zh) * 2022-06-29 2025-01-21 深圳华大生命科学研究院 图像中的点坐标的确定方法和装置及相应设备和介质
CN119422168A (zh) * 2022-06-29 2025-02-11 深圳华大生命科学研究院 一种图像处理方法、装置、设备及介质
CN119422162A (zh) * 2022-06-29 2025-02-11 深圳华大生命科学研究院 图像拼接方法、基因测序系统及相应的基因测序仪
CN120344996A (zh) * 2022-12-05 2025-07-18 深圳华大生命科学研究院 一种图像处理方法、装置、设备及介质
CN115937282B (zh) * 2023-01-10 2024-08-02 郑州思昆生物工程有限公司 一种荧光图像的配准方法、装置、设备和存储介质
CN116503471B (zh) * 2023-06-28 2023-08-29 南开大学 K次近邻位置云图的单分子定位成像漂移校正方法与系统
CN118279362A (zh) * 2024-04-19 2024-07-02 郑州思昆生物工程有限公司 荧光图像的配准方法、计算机设备及计算机可读存储介质
CN119991817B (zh) * 2025-04-15 2025-06-17 真健康(广东横琴)医疗科技有限公司 一种x角点定位方法及设备
CN120636551B (zh) * 2025-06-05 2026-04-10 上海赛陆生命科学有限公司 基因测序的模板构建方法、产品、基因测序仪及介质

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104992433A (zh) * 2015-06-19 2015-10-21 北京邮电大学 基于线段匹配的多光谱图像配准的方法及装置
CN105408939A (zh) * 2013-07-23 2016-03-16 皇家飞利浦有限公司 用于配准成像设备与跟踪设备的配准系统
CN105427327A (zh) * 2015-12-10 2016-03-23 北京中科紫鑫科技有限责任公司 一种dna测序的图像配准方法及装置
CN108537830A (zh) * 2017-03-02 2018-09-14 广州康昕瑞基因健康科技有限公司 图像配准方法和系统及图像拍摄对位方法和系统

Family Cites Families (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP3576523B2 (ja) * 2000-12-26 2004-10-13 オリンパス株式会社 蛍光輝度測定方法及び装置
US8703422B2 (en) * 2007-06-06 2014-04-22 Pacific Biosciences Of California, Inc. Methods and processes for calling bases in sequence by incorporation methods
US8774494B2 (en) * 2010-04-30 2014-07-08 Complete Genomics, Inc. Method and system for accurate alignment and registration of array for DNA sequencing
CN105427328B (zh) * 2015-12-10 2017-05-31 北京中科紫鑫科技有限责任公司 一种dna测序的互信息图像配准方法及装置
US20200359550A1 (en) * 2019-05-13 2020-11-19 Bao Tran Farm ecosystem

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105408939A (zh) * 2013-07-23 2016-03-16 皇家飞利浦有限公司 用于配准成像设备与跟踪设备的配准系统
CN104992433A (zh) * 2015-06-19 2015-10-21 北京邮电大学 基于线段匹配的多光谱图像配准的方法及装置
CN105427327A (zh) * 2015-12-10 2016-03-23 北京中科紫鑫科技有限责任公司 一种dna测序的图像配准方法及装置
CN108537830A (zh) * 2017-03-02 2018-09-14 广州康昕瑞基因健康科技有限公司 图像配准方法和系统及图像拍摄对位方法和系统

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN114549600A (zh) * 2022-01-28 2022-05-27 赛纳生物科技(北京)有限公司 一种荧光图像配准方法
WO2024001051A1 (zh) * 2022-06-29 2024-01-04 深圳华大生命科学研究院 空间组学单细胞数据获取方法及装置和电子设备
CN115602245A (zh) * 2022-09-09 2023-01-13 郑州思昆生物工程有限公司（Cn） 一种荧光图筛选的方法、装置、设备以及存储介质
CN115602245B (zh) * 2022-09-09 2023-10-03 郑州思昆生物工程有限公司 一种荧光图筛选的方法、装置、设备以及存储介质

Also Published As

Publication number Publication date
US20220108462A1 (en) 2022-04-07
US11972583B2 (en) 2024-04-30
CN112204615B (zh) 2023-05-02
CN112204615A (zh) 2021-01-08

Similar Documents

Publication Publication Date Title
US11972583B2 (en) Fluorescence image registration method, gene sequencing instrument, and storage medium
US12039744B2 (en) Fluorescence image registration method, gene sequencing instrument and system
CN110111842A (zh) 图像清晰度分析及对焦方法、测序仪、系统与存储介质
CN118967798B (zh) 一种显示屏子像素亮团的定位方法、装置和存储介质
CN107016389B (zh) 一种车牌定位的方法以及装置
CN115661774A (zh) 一种车道线特征检测方法
US20150317536A1 (en) System and method for evaluating data
CN116824122A (zh) 基于深度学习的led芯片定位方法及led芯片定位装置
CN114820530A (zh) 一种数字pcr荧光图像的像素级别识别方法
CN114792383A (zh) 一种微流控芯片数字pcr荧光图像识别方法和装置
CN116228856A (zh) 定位微流控生物芯片所有磁头点样点的方法、系统及设备
HK40035019A (zh) 荧光图像配准方法、基因测序仪及系统、存储介质
HK40035019B (zh) 荧光图像配准方法、基因测序仪及系统、存储介质
CN120726075B (zh) 一种自动绘制感兴趣区域的方法及相关产品
CN106778780B (zh) 一种基于gpu的边缘检测图像二值化方法
CN121582555B (zh) 一种背景区域数据的获取方法、电子设备及存储介质
CN119672013B (zh) 灯珠缺陷检测方法、电子设备及存储介质
US12277468B2 (en) Magnetic bead-based detection method, storage medium, and detection device
HK40033773B (zh) 荧光图像配准方法、基因测序仪及系统、存储介质
CN113762259B (zh) 文本定位方法、装置、计算机系统和可读存储介质
CN115984280A (zh) 基于igbt器件的表痕分析方法、装置、设备及介质
CN119273720A (zh) 图像配准的方法、装置、设备和存储介质
HK40033773A (zh) 荧光图像配准方法、基因测序仪及系统、存储介质
CN121169789A (zh) 基于计算机视觉的检验记录识别的方法及装置
CN119963552A (zh) 一种用于微流控微滴阵列荧光比例信息统计的方法及装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19912762

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19912762

Country of ref document: EP

Kind code of ref document: A1