Fluorescent probe detection image processing method based on noise analysis

Abstract

The invention relates to the field of image processing, in particular to a fluorescence probe detection image processing method based on noise analysis, and the method comprises the steps: obtaining a fluorescence microscopic grayscale image; obtaining a window first optimization factor corresponding to the pixel point according to the local gray scale standard deviation in the pixel point initial window; obtaining a preliminary self-adaptive window side length according to the window first optimization factor; acquiring a local contrast of the pixel points according to the preliminary adaptive window, and acquiring a second optimization factor of the window according to the first optimization factor of the window and the local contrast; obtaining a self-adaptive local window side length according to the window first optimization factor and the window second optimization factor; acquiring a local self-adaptive contrast threshold value of the pixel point according to the first window optimization factor, the second window optimization factor and the local contrast of the pixel point; and image segmentation processing is completed according to the self-adaptive local window and the local self-adaptive contrast threshold value of the pixel points, so that the accuracy of image segmentation is improved, and the weak signal detection capability is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and particularly to an image processing method for fluorescence probe detection based on noise analysis. Background Art

[0002] Fluorescence microscopy imaging technology plays a crucial role in modern biomedical research. By using fluorescence probes to label specific cell structures or biomolecules, researchers can observe the dynamic changes of various biological processes in living cells in real time. For example, by labeling mitochondrial probes, the behaviors of mitochondrial fusion, fission, and movement can be observed; by labeling lysosome probes, the processes of intracellular material degradation and transport can be studied. However, fluorescence microscopy images, especially images of live cell imaging, often have problems such as complex background noise, uneven signal intensity, and diverse organelle morphologies and sizes, which pose challenges to the subsequent processing and quantitative analysis of the images. In order to accurately extract the target region of interest from fluorescence microscopy images, image segmentation is required. Currently, common image segmentation methods include threshold segmentation, edge detection, region growing, clustering analysis, etc. Among them, the segmentation method based on local adaptive threshold has certain advantages in processing images with non-uniform illumination and complex backgrounds because it can dynamically adjust the threshold according to the local features of the image. The Bernsen Algorithm is a classic local adaptive threshold segmentation algorithm, which determines the threshold of each pixel by calculating the maximum and minimum gray values within the local window around the pixel. However, the traditional Bernsen algorithm has some problems, such as being sensitive to noise and prone to generating artifacts in the noise region. In recent years, researchers have proposed many improved Bernsen algorithms in an attempt to improve its robustness to noise. Some methods introduce contrast limitation, and only when the local contrast is greater than a preset value, the threshold calculated by the Bernsen algorithm is used, otherwise the global threshold or other methods are used to calculate the threshold. There are also some methods that combine neighborhood information, and when calculating the local threshold, not only the local window around the current pixel is considered, but also the threshold information of its neighboring pixel points is considered. These improved methods have improved the segmentation accuracy to a certain extent, but there is still a core problem: they usually use a local window of a fixed size, and the contrast threshold is also preset, making it difficult to adapt to problems such as the diversity of organelle sizes and the non-uniformity of signal intensity commonly existing in fluorescence microscopy images. The existing improved Bernsen algorithms use a local window of a fixed size and a fixed contrast threshold, making it difficult to adapt to the characteristics of the diversity of organelle sizes and the non-uniformity of signal intensity in fluorescence microscopy images, resulting in inaccurate segmentation in regions with high noise and low contrast, and being prone to losing weak signals or generating artifacts.

[0003] Specifically, it is difficult for a local window with a fixed size to accommodate organelles of different sizes. For larger organelles, a smaller window cannot completely contain their entirety, resulting in an incomplete segmentation result; for smaller organelles, a larger window contains too much background noise, leading to inaccurate threshold calculation. In addition, there are often significant differences in the background noise levels and signal intensities in different regions of fluorescence microscopy images. A fixed-size window and a fixed contrast threshold cannot adapt to these changes simultaneously, resulting in poor segmentation effects in some regions. Especially in applications such as organelle tracking, the loss of weak signals will cause interruptions or errors in the tracking trajectory, seriously affecting subsequent quantitative analysis.

[0004] Therefore, there is an urgent need for a processing method that can adaptively adjust the size of the local window and the contrast threshold according to the local characteristics of fluorescence microscopy images, thereby improving the accuracy of image segmentation and enhancing the detection ability for weak signals. Summary of the Invention

[0005] In view of this, the embodiments of the present invention provide a fluorescence probe detection image processing method based on noise analysis to solve the problem that a fixed-size window and a fixed contrast threshold cannot adapt to these changes simultaneously, resulting in poor segmentation effects in some regions.

[0006] The embodiments of the present invention provide a fluorescence probe detection image processing method based on noise analysis. A fluorescence probe detection image processing method based on noise analysis includes the following steps:

[0007] Collect the cell fluorescence microscopy image to be detected; perform preprocessing and grayscale conversion on the collected cell fluorescence microscopy image to obtain a fluorescence microscopy grayscale image. By analyzing the local grayscale changes within the initial window of the pixel points in the fluorescence microscopy grayscale image, obtain the first window optimization factor of the pixel points; adjust the initial window of the pixel points through the first window optimization factor to obtain the preliminary adaptive window side length of the pixel points in the fluorescence microscopy grayscale image; analyze the contrast of the preliminary adaptive window of the pixel points in the fluorescence microscopy grayscale image to obtain the local contrast of the pixel points; analyze the local contrast of the pixel points to obtain the second window optimization factor of the pixel points in the fluorescence microscopy grayscale image; evaluate the window of the pixel points in the fluorescence microscopy grayscale image through the first window optimization factor and the second window optimization factor to obtain the adaptive local window side length of the pixel points; analyze the window content of the adaptive local window of the pixel points to obtain the local adaptive contrast threshold of the pixel points; segment the fluorescence microscopy grayscale image through the adaptive local window and the local adaptive contrast threshold of the pixel points to obtain the binary segmentation image corresponding to the fluorescence microscopy grayscale image.

[0008] Preferably, by analyzing the local gray-scale change within the initial window of the pixel points in the fluorescence microscopic gray-scale image, a first window optimization factor of the pixel points is obtained. The specific steps include:

[0009] Obtain a set window adjustment factor, a window optimization regularization factor, and the initial window size. Divide the pixel points in the fluorescence microscopic image into local windows through the initial window size, obtain the first local window corresponding to the pixel points in the fluorescence microscopic image, and take the calculation result of dividing the window adjustment factor by the sum of the standard deviation of the pixel gray-scale values in the first local window corresponding to the pixel points in the fluorescence microscopic image and the window optimization regularization factor as the first window optimization factor corresponding to the pixel points.

[0010] Preferably, adjust the initial window of the pixel points through the first window optimization factor to obtain the preliminary adaptive window side length of the pixel points in the fluorescence microscopic gray-scale image. The specific steps include:

[0011] Obtain a set window limit constant. Take the calculation result of multiplying the window limit constant by the first window optimization factor and rounding down as the first window side length, and take the calculation result of adding the constant 3 to twice the first window side length as the preliminary adaptive window side length corresponding to the pixel points.

[0012] Preferably, by analyzing the contrast of the preliminary adaptive window of the pixel points in the fluorescence microscopic gray-scale image, the local contrast of the pixel points is obtained. The specific steps include:

[0013] Obtain the preliminary adaptive window side length corresponding to the pixel points. Take the pixel points with gray-scale values greater than the average gray-scale value of all pixel points within the window in the preliminary adaptive window of the pixel points as the first pixel point set corresponding to the pixel points, and take the average gray-scale value of the pixel points in the first pixel point set corresponding to the pixel points as the first average gray-scale value corresponding to the pixel points; take the pixel points with gray-scale values less than the average gray-scale value of all pixel points within the window in the preliminary adaptive window of the pixel points as the second pixel point set corresponding to the pixel points, and take the average gray-scale value of the pixel points in the second pixel point set corresponding to the pixel points as the second average gray-scale value corresponding to the pixel points; take the calculation result of subtracting the second average gray-scale value corresponding to the pixel points from the first average gray-scale value corresponding to the pixel points as the local contrast corresponding to the pixel points.

[0014] Preferably, by analyzing the local contrast of the pixel points, a second window optimization factor of the pixel points in the fluorescence microscopic gray-scale image is obtained. The specific steps include:

[0015] Obtain the first window optimization factor corresponding to the pixel and the local contrast corresponding to the pixel, and use the calculation result of dividing the local contrast corresponding to the pixel by the first window optimization factor corresponding to the pixel as the second window optimization factor corresponding to the pixel.

[0016] Preferably, perform window evaluation on the pixels in the fluorescence microscopic grayscale image according to the first window optimization factor and the second window optimization factor of the window to obtain the adaptive local window side length of the pixel. The specific steps include:

[0017] Obtain the set window limit constant, the first window optimization factor corresponding to the pixel and the second window optimization factor corresponding to the pixel; use the calculation result of multiplying the window limit constant, the first window optimization factor corresponding to the pixel and the second window optimization factor corresponding to the pixel and rounding down as the second window side length, and use the calculation result of adding the constant 2 multiplied by the second window side length to the constant 3 as the adaptive local window side length corresponding to the pixel.

[0018] Preferably, obtaining the local adaptive contrast threshold of the pixel according to the first window optimization factor corresponding to the pixel, the second window optimization factor corresponding to the pixel and the local contrast corresponding to the pixel includes:

[0019] Obtain the first window optimization factor corresponding to the pixel, the second window optimization factor corresponding to the pixel and the local contrast corresponding to the pixel; use the calculation result of dividing the first window optimization factor corresponding to the pixel by the second window optimization factor corresponding to the pixel as the contrast threshold evaluation factor, and use the calculation result of adding the constant one to the contrast threshold evaluation factor and multiplying by the local contrast corresponding to the pixel as the local adaptive contrast threshold of the pixel.

[0020] Preferably, segmenting the fluorescence microscopic grayscale image according to the adaptive local window side length corresponding to the pixel and the local adaptive contrast threshold of the pixel to obtain a binary segmentation image corresponding to the fluorescence microscopic grayscale image, includes:

[0021] Obtain the side length of the adaptive local window corresponding to the pixel and the local adaptive contrast threshold of the pixel. Determine the local adaptive window of the pixel in the fluorescence microscopic grayscale image according to the side length of the adaptive local window corresponding to the pixel. Use the calculation result of adding the average grayscale value of all pixels in the local adaptive window corresponding to the pixel to the local adaptive contrast threshold of the pixel as the segmentation threshold of the pixel. Compare the grayscale value of the pixel with the segmentation threshold of the pixel to obtain the binary segmentation label of the pixel, complete the binary segmentation of the fluorescence microscopic image, and obtain the binary segmentation image corresponding to the fluorescence microscopic grayscale image.

[0022] Preferably, the comparing the grayscale value of the pixel with the segmentation threshold of the pixel to obtain the binary segmentation label of the pixel includes:

[0023] Obtain the segmentation threshold of the pixel. If the grayscale value of the pixel is greater than or equal to the segmentation threshold of the pixel, mark the pixel as a target pixel; if the grayscale value of the pixel is less than the segmentation threshold of the pixel, mark the pixel as a background pixel.

[0024] The beneficial effects of the embodiments of the present invention compared with the prior art are:

[0025] The present invention acquires the fluorescence microscopic image of the cell to be detected; preprocesses and performs gray conversion on the acquired fluorescence microscopic image of the cell to obtain a fluorescence microscopic gray image, analyzes the local gray change within the initial window of the pixel points in the fluorescence microscopic gray image to obtain the first window optimization factor of the pixel points; adjusts the initial window of the pixel points through the first window optimization factor to obtain the preliminary adaptive window side length of the pixel points in the fluorescence microscopic gray image; analyzes the contrast of the preliminary adaptive window of the pixel points in the fluorescence microscopic gray image to obtain the local contrast of the pixel points; analyzes the local contrast of the pixel points to obtain the second window optimization factor of the pixel points in the fluorescence microscopic gray image; performs window evaluation on the pixel points in the fluorescence microscopic gray image through the first window optimization factor and the second window optimization factor to obtain the adaptive local window side length of the pixel points; analyzes the window content of the adaptive local window of the pixel points to obtain the local adaptive contrast threshold of the pixel points; segments the fluorescence microscopic gray image through the adaptive local window and the local adaptive contrast threshold of the pixel points to obtain the binary segmentation image corresponding to the fluorescence microscopic gray image. Among them, the first window optimization factor is obtained through the standard deviation of the gray values of all pixel points within the initial window of the pixel points in the fluorescence microscopic image, the preliminary adaptive window is obtained according to the first window optimization factor, the second window optimization factor is evaluated through the preliminary adaptive window, and then the local adaptive window for Bernsen binary segmentation of the fluorescence microscopic image is obtained through the first window optimization factor and the second window optimization factor, and the adaptive image segmentation threshold is obtained according to the first window optimization factor, the second window optimization factor and the local contrast in the local adaptive window, so as to improve the accuracy of image segmentation and enhance the detection ability of weak signals. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The drawings constituting a part of the present invention are used to provide a further understanding of the present invention. The schematic embodiments and descriptions thereof of the present invention are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0027] Figure 1 is a flowchart of a method for processing an image detected by a fluorescence probe based on noise analysis according to an embodiment of the present invention; DETAILED DESCRIPTION OF THE EMBODIMENTS

[0028] The embodiments of the present disclosure will be described in detail below. The examples of the embodiments are shown in the drawings, where the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the drawings are exemplary and are intended to explain the present disclosure and should not be construed as a limitation of the present disclosure.

[0029] It should be noted that the terms "first", "second", etc. in the specification of the present disclosure and the above-mentioned drawings are used to distinguish similar objects, and do not necessarily need to be used to describe a specific order or sequence. It should be understood that the data used for this toothpaste can be interchanged under appropriate circumstances, so that the embodiments of the present disclosure described herein can be implemented in an order other than those illustrated or described herein. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. On the contrary, they are merely examples of devices and methods consistent with some aspects of the present disclosure.

[0030] In order to illustrate the technical solution of the present invention, the following will be described by means of specific embodiments.

[0031] The specific scenario targeted by the present invention is: image segmentation of microscopic images detected by fluorescence probes.

[0032] See Figure 1 , which is a method flowchart of an image processing method for fluorescence probe detection based on noise analysis provided by Embodiment 1 of the present invention. As Figure 1 shown, an image processing method for fluorescence probe detection based on noise analysis may include:

[0033] Step S101, collect a fluorescence microscopic image of a cell, preprocess the collected fluorescence microscopic image of the cell and convert it into a grayscale image to obtain a fluorescence microscopic grayscale image.

[0034] Preprocess the fluorescence microscopic image of the cell by using the top-hat transformation method to obtain a fluorescence microscopic image of the cell after background correction, and perform grayscale transformation on the fluorescence microscopic image of the cell after background correction to obtain a fluorescence microscopic grayscale image.

[0035] Step S102, obtain the first optimization factor of the window corresponding to the pixel point according to the local grayscale standard deviation in the initial window corresponding to the pixel point in the fluorescence microscopic grayscale image; obtain the preliminary adaptive window side length corresponding to the pixel point according to the first optimization factor of the window corresponding to the pixel point.

[0036] In the fluorescence microscopic image processing of cells, especially for the dynamic tracking of organelles, accurate image segmentation is the basis for subsequent quantitative analysis. However, the sizes and morphologies of organelles are significantly diverse, and there are also large differences in the background noise levels in different regions of the image. It is difficult for traditional image processing based on the Bernsen image segmentation method with a fixed window size to take both situations into account. In regions with high noise, using a larger window can better estimate the background, but it will sacrifice the details of the image and may misclassify some weak organelle signals as the background; while near the edges of organelles or near smaller organelle structures, using a smaller window can better retain the edge details, but it is easily interfered by noise, resulting in inaccurate segmentation results. Therefore, after obtaining the fluorescence microscopic grayscale image, it is necessary to obtain the first window optimization factor corresponding to the pixel point according to the local grayscale standard deviation of the pixel point in the initial window in the fluorescence microscopic grayscale image, including:

[0037] Obtain the set window adjustment factor, the window optimization regularization factor and the initial window size, divide the pixel points in the fluorescence microscopic image into local windows through the initial window size, obtain the first local window corresponding to the pixel points in the fluorescence microscopic image, and take the calculation result of dividing the window adjustment factor by the sum of the standard deviation of the pixel point gray values in the first local window corresponding to the pixel points in the fluorescence microscopic image and the window optimization regularization factor as the first window optimization factor corresponding to the pixel point.

[0038] In one embodiment, assume that the standard deviation of the gray values in the initial window of the pixel point with coordinates in the fluorescence microscopic grayscale image is , then the calculation expression of the first window optimization factor corresponding to the pixel point is:

[0039]

[0040] Among them, represents the first window optimization factor of the pixel point with coordinates in the image; represents the window adjustment factor; represents the window optimization regularization factor.

[0041] It should be noted that this step aims to initially solve the problem of the fixed local window size. The design idea of the first window optimization factor for pixel points is based on the analysis of the local grayscale statistical characteristics of the image, especially the local grayscale standard deviation . This is because the local grayscale standard deviation can be used as an important indicator to measure the local region noise level. Specifically, for each pixel point in the image, first calculate the standard deviation of the gray values within an initial window around it. In this embodiment, the initial window is set to 5x5. A larger value usually means that the gray level in this area changes drastically, there is strong noise, or it contains edge information of organelles; A smaller value usually means that the gray level in this area changes gently, mainly containing background signals or relatively uniform areas inside the organelles.

[0042] The solution based on this understanding is to dynamically adjust the window size according to the local gray standard deviation around each pixel point When is larger, the window size is appropriately reduced to avoid introducing too much noise while better retaining the detailed information of the edges of the organelles; when is smaller, the window size is appropriately increased to obtain a more stable background estimate, so as to calculate the threshold more accurately. In order to control the influence degree of the gray standard deviation on the window size, a regulation factor is introduced. The role of is to adjust the sensitivity of the noise level to the window size adjustment, similar to a "magnification factor". When is set to be larger, a slight change in the gray standard deviation will also cause a significant adjustment of the window size.

[0043] Through the above method, the preliminary adaptive window size adjustment based on the local gray standard deviation is realized. The factor is inversely proportional to . The preliminary adjustment of the window size according to the local noise level enables the algorithm to initially adapt to the diversity of organelle sizes and the spatial heterogeneity of the noise level.

[0044] After obtaining the first optimization factor of the window for the pixel point in the fluorescence microscopy gray image, it is necessary to obtain the preliminary adaptive window side length corresponding to the pixel point according to the first optimization factor of the window corresponding to the pixel point, including:

[0045] Obtain the set window limit constant, take the calculation result of multiplying the window limit constant by the first optimization factor of the window and rounding down as the first window side length, and take the calculation result of adding the product of the constant 2 and the first window side length to the constant 3 as the preliminary adaptive window side length corresponding to the pixel point.

[0046] In an embodiment, assuming that the window limit constant is , the calculation expression of the preliminary adaptive window side length of the pixel point is:

[0047]

[0048] Among them, represents the coordinate in the fluorescence microscopy gray image as The initial adaptive window side length of the pixel points; Indicates the window limit constant; Indicates the coordinates in the fluorescence microscopy grayscale image as The window first optimization factor of the pixel points; Indicates the floor function.

[0049] It should be noted that in this embodiment, the window limit constant is set to to ensure that the minimum window size is and, through ensure that the window side length is odd. Introducing the window limit constant plays a role in controlling the overall range of the window because the value of is greater than so the minimum value of is then when the result of this part is To make the window size at least it is necessary .

[0050] Step S103: Obtain the local contrast corresponding to the pixel points according to the initial adaptive window corresponding to the pixel points, and obtain the window second optimization factor corresponding to the pixel points according to the window first optimization factor corresponding to the pixel points and the local contrast corresponding to the pixel points; obtain the adaptive local window side length corresponding to the pixel points according to the window first optimization factor corresponding to the pixel points and the window second optimization factor corresponding to the pixel points.

[0051] In fluorescence microscopy images, due to factors such as the concentration of fluorescent dyes for organelle labeling, the spatial distribution of organelles, and the characteristics of the optical imaging system, the signal intensities of organelles often vary greatly. There are both bright regions with strong signals and dim regions with weak signals. If only based on the gray standard deviation to adjust the window size, it will cause the window size to be too small in regions with low contrast to accurately estimate the gray level of the local background, thereby affecting the accuracy of subsequent threshold calculation. Especially in applications such as organelle tracking, it is usually desired to retain as much weak signal of organelles as possible because these weak signals often correspond to the starting, ending, or interaction regions with other organelles and have important biological significance.

[0052] Therefore, after obtaining the initial adaptive window side length of the pixel points, further optimization is required on the basis of the initial adaptive window. Specifically, first, obtain the local contrast corresponding to the pixel points according to the initial adaptive window corresponding to the pixel points, including:

[0053] Obtain the initial adaptive window side length corresponding to the pixel point. Take the pixel points in the initial adaptive window of the pixel point whose gray value is greater than the average gray value of all pixel points in the window as the first pixel point set corresponding to the pixel point, and take the average gray value of the pixel points in the first pixel point set corresponding to the pixel point as the first average gray value corresponding to the pixel point; take the pixel points in the initial adaptive window of the pixel point whose gray value is less than the average gray value of all pixel points in the window as the second pixel point set corresponding to the pixel point, and take the average gray value of the pixel points in the second pixel point set corresponding to the pixel point as the second average gray value corresponding to the pixel point; take the calculation result of subtracting the second average gray value corresponding to the pixel point from the first average gray value corresponding to the pixel point as the local contrast corresponding to the pixel point.

[0054] In one embodiment, assume that in the fluorescence microscopic gray image, the coordinate is The pixel point corresponding to the initial adaptive window of the pixel point, and the coordinate is The gray value of the pixel point is , then the calculation expression of the local contrast corresponding to the pixel point is:

[0055]

[0056] Among them, Represents the local contrast corresponding to the pixel point; Represents the coordinate in the fluorescence microscopic gray image as The pixel point corresponding to the initial adaptive window of the pixel point, and the coordinate is The gray value of the pixel point; Represents the initial adaptive window of the pixel point; Represents at The average gray value of all pixels in the window; Represents the coordinate in the fluorescence microscopic gray image as The number of pixel points in the initial adaptive window of the pixel point that are greater than or equal to the gray value mean; Represents the coordinate in the fluorescence microscopic gray image as The number of pixel points in the initial adaptive window of the pixel point that are less than the gray value mean.

[0057] After obtaining the local contrast corresponding to the pixel point, obtaining the second optimization factor of the window corresponding to the pixel point according to the first optimization factor of the window corresponding to the pixel point and the local contrast corresponding to the pixel point, includes:

[0058] Obtain the first window optimization factor corresponding to the pixel and the local contrast corresponding to the pixel, and use the calculation result of dividing the local contrast corresponding to the pixel by the first window optimization factor corresponding to the pixel as the second window optimization factor corresponding to the pixel.

[0059] In one embodiment, the calculation expression of the second window optimization factor corresponding to the pixel is:

[0060]

[0061] Wherein, represents the second window optimization factor corresponding to the pixel; represents the local contrast corresponding to the pixel with coordinates in the fluorescence microscopy grayscale image; represents the first window optimization factor of the pixel with coordinates in the fluorescence microscopy grayscale image.

[0062] It should be noted that, and The calculation of avoids the problem of noise sensitivity caused by directly using the maximum and minimum grayscale values in the window to calculate the contrast. The local contrast reflects the gray level difference between the bright area and the dark area within the window. The optimization factor combines the local contrast information with the noise level information obtained in the above steps by the ratio of . Specifically, in areas with higher contrast or larger noise, has a larger value, which will cause a further increase in the final window size, so as to better estimate the background; while in areas with lower contrast and smaller noise, has a smaller value, thus avoiding excessive increase in the window size and helping to retain weak signals.

[0063] After obtaining the second window optimization factor of the pixel, the adaptive local window side length corresponding to the pixel can be obtained according to the first window optimization factor corresponding to the pixel and the second window optimization factor corresponding to the pixel, including:

[0064] Obtain the set window limit constant, the first window optimization factor corresponding to the pixel and the second window optimization factor corresponding to the pixel; use the calculation result of multiplying the window limit constant, the first window optimization factor corresponding to the pixel and the second window optimization factor corresponding to the pixel and rounding down as the second window side length, and use the calculation result of adding 3 to 2 times the second window side length as the adaptive local window side length corresponding to the pixel.

[0065] In one embodiment, the calculation expression of the side length of the adaptive local window corresponding to the pixel is:

[0066]

[0067] Wherein, represents the side length of the adaptive local window of the pixel with coordinates in the fluorescence microscopic grayscale image; represents the window limit constant; represents the first window optimization factor of the pixel with coordinates in the fluorescence microscopic grayscale image; represents the second window optimization factor of the pixel with coordinates in the fluorescence microscopic grayscale image.

[0068] Step S104: Obtain the local adaptive contrast threshold of the pixel according to the first window optimization factor corresponding to the pixel, the second window optimization factor corresponding to the pixel, and the local contrast corresponding to the pixel; perform fluorescence microscopic grayscale image segmentation according to the adaptive local window corresponding to the pixel and the local adaptive contrast threshold of the pixel to obtain a binary segmentation image corresponding to the fluorescence microscopic grayscale image.

[0069] After obtaining the side length of the adaptive local window corresponding to the pixel, the local adaptive contrast threshold of the pixel can be obtained according to the first window optimization factor corresponding to the pixel, the second window optimization factor corresponding to the pixel, and the local contrast corresponding to the pixel, including:

[0070] Obtain the first window optimization factor corresponding to the pixel, the second window optimization factor corresponding to the pixel, and the local contrast corresponding to the pixel; use the calculation result of dividing the first window optimization factor corresponding to the pixel by the second window optimization factor corresponding to the pixel as the contrast threshold evaluation factor, and use the calculation result of adding a constant one to the contrast threshold evaluation factor and multiplying by the local contrast corresponding to the pixel as the local adaptive contrast threshold of the pixel.

[0071] In one embodiment, the calculation expression of the local adaptive contrast threshold of the pixel is:

[0072]

[0073] Wherein, represents the local adaptive contrast threshold of the pixel with coordinates in the fluorescence microscopic grayscale image; represents the local contrast of the pixel with coordinates in the fluorescence microscopic grayscale image; represents the first optimization factor of the window for the pixel point with coordinates in the fluorescence microscopic grayscale image; represents the second optimization factor of the window for the pixel point with coordinates in the fluorescence microscopic grayscale image.

[0074] After obtaining the local adaptive contrast threshold of the pixel point and the adaptive local window of the pixel point, the fluorescence microscopic grayscale image can be segmented according to the adaptive local window corresponding to the pixel point and the local adaptive contrast threshold of the pixel point to obtain a binary segmentation image corresponding to the fluorescence microscopic grayscale image, including:

[0075] Obtain the side length of the adaptive local window corresponding to the pixel point and the local adaptive contrast threshold of the pixel point. Determine the local adaptive window of the pixel point in the fluorescence microscopic grayscale image according to the side length of the adaptive local window corresponding to the pixel point. Take the calculation result of adding the average gray value of all pixel points in the local adaptive window corresponding to the pixel point and the local adaptive contrast threshold of the pixel point as the segmentation threshold of the pixel point. Compare the gray value of the pixel point with the segmentation threshold of the pixel point to obtain the binary segmentation label of the pixel point, complete the binary segmentation of the fluorescence microscopic image, and obtain the binary segmentation image corresponding to the fluorescence microscopic grayscale image.

[0076] Among them, comparing the gray value of the pixel point with the segmentation threshold of the pixel point to obtain the binary segmentation label of the pixel point includes:

[0077] Obtain the segmentation threshold of the pixel point. If the gray value of the pixel point is greater than or equal to the segmentation threshold of the pixel point, mark the pixel point as a target pixel point; if the gray value of the pixel point is less than the segmentation threshold of the pixel point, mark the pixel point as a background pixel point. Thus, the image binary segmentation process of the fluorescence probe detection image is completed.

[0078] In summary, the embodiments of the present invention collect the fluorescence microscopic images of cells to be detected; preprocess and perform gray-scale conversion on the collected fluorescence microscopic images of cells to obtain fluorescence microscopic gray-scale images, analyze the local gray-scale changes within the initial window of the pixel points in the fluorescence microscopic gray-scale images to obtain the first window optimization factor of the pixel points; adjust the initial window of the pixel points through the first window optimization factor to obtain the preliminary adaptive window side length of the pixel points in the fluorescence microscopic gray-scale images; analyze the contrast of the preliminary adaptive window of the pixel points in the fluorescence microscopic gray-scale images to obtain the local contrast of the pixel points; analyze the local contrast of the pixel points to obtain the second window optimization factor of the pixel points in the fluorescence microscopic gray-scale images; evaluate the window of the pixel points in the fluorescence microscopic gray-scale images through the first window optimization factor and the second window optimization factor to obtain the adaptive local window side length of the pixel points; analyze the window content of the adaptive local window of the pixel points to obtain the local adaptive contrast threshold of the pixel points; segment the fluorescence microscopic gray-scale images through the adaptive local window and the local adaptive contrast threshold of the pixel points to obtain the binary segmentation image corresponding to the fluorescence microscopic gray-scale images. Among them, the first window optimization factor is obtained through the standard deviation of the gray-scale values of all pixel points within the initial window of the pixel points in the fluorescence microscopic images, the preliminary adaptive window is obtained according to the first window optimization factor, the second window optimization factor is evaluated through the preliminary adaptive window, and then the local adaptive window for Bernsen binary segmentation of the fluorescence microscopic images is obtained through the first window optimization factor and the second window optimization factor, and the adaptive image segmentation threshold is obtained according to the first window optimization factor, the second window optimization factor and the local contrast in the local adaptive window, thereby improving the accuracy of image segmentation and enhancing the detection ability for weak signals.

[0079] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications 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 invention, and should all be included in the protection scope of the present invention.

Claims

1. A fluorescent probe detection image processing method based on noise analysis, comprising: Collecting fluorescent microscopic images of cells to be detected; The collected cell fluorescence microscopic image is preprocessed and grayscale converted to obtain a fluorescence microscopic grayscale image, which is characterized in that a first window optimization factor of the pixel is obtained by analyzing the local grayscale changes of the pixel points in the fluorescence microscopic grayscale image within the initial window; the initial window of the pixel points is adjusted by the first window optimization factor to obtain the preliminary adaptive window side length of the pixel points in the fluorescence microscopic grayscale image; the local contrast of the pixel points is obtained by performing contrast analysis on the preliminary adaptive window of the pixel points in the fluorescence microscopic grayscale image; the second window optimization factor of the pixel points in the fluorescence microscopic grayscale image is obtained by analyzing the local contrast of the pixel points; the pixel points in the fluorescence microscopic grayscale image are window evaluated by the first window optimization factor and the second window optimization factor to obtain the adaptive local window side length of the pixel points; the local adaptive contrast threshold of the pixel points is obtained by performing window content analysis on the adaptive local window of the pixel points; the fluorescence microscopic grayscale image is segmented by the adaptive local window of the pixel points and the local adaptive contrast threshold to obtain a binary segmented image corresponding to the fluorescence microscopic grayscale image.

2. The method for processing fluorescent probe detection images based on noise analysis according to claim 1, characterized in that: According to the method of analyzing the local grayscale changes of the pixels in the fluorescence microscopic grayscale image within the initial window, the first window optimization factor of the pixels is obtained, and the specific steps include: The set window adjustment factor, the window optimization regularization factor and the initial window size are obtained, the pixel points in the fluorescence microscopy image are divided into local windows according to the initial window size, and the first local window corresponding to the pixel point in the fluorescence microscopy image is obtained. The window adjustment factor is divided by the sum of the standard deviation of the grayscale value of the pixel point in the first local window corresponding to the pixel point in the fluorescence microscopy image and the window optimization regularization factor as the first optimization factor of the window corresponding to the pixel point.

3. The method for processing fluorescent probe detection images based on noise analysis according to claim 1, characterized in that: According to the adjustment of the initial window of the pixel point by the first window optimization factor, the preliminary adaptive window side length of the pixel point in the fluorescence microscopic grayscale image is obtained, and the specific steps include: Get the set window limit constant, multiply the window limit constant by the first window optimization factor and round down the result as the first window side length, multiply constant 2 by the first window side length and add the result to constant 3 as the preliminary adaptive window side length corresponding to the pixel point.

4. The method for processing fluorescent probe detection images based on noise analysis according to claim 1, characterized in that: According to the method, the local contrast of the pixel points is obtained by performing contrast analysis on the preliminary adaptive window of the pixel points in the fluorescence microscopic grayscale image, and the specific steps include: Obtain the side length of the preliminary adaptive window corresponding to the pixel point, take the pixel point whose grayscale value in the preliminary adaptive window of the pixel point is greater than the average grayscale value of all the pixels in the window as the first pixel point set corresponding to the pixel point, and take the average grayscale value of the pixels in the first pixel point set corresponding to the pixel point as the first average grayscale value corresponding to the pixel point; take the pixel point whose grayscale value in the preliminary adaptive window of the pixel point is less than the average grayscale value of all the pixels in the window as the second pixel point set corresponding to the pixel point, and take the average grayscale value of the pixels in the second pixel point set corresponding to the pixel point as the second average grayscale value corresponding to the pixel point; and take the result of subtracting the first average grayscale value corresponding to the pixel point from the second average grayscale value corresponding to the pixel point as the local contrast corresponding to the pixel point.

5. The method for processing fluorescent probe detection images based on noise analysis according to claim 1, characterized in that: According to the analysis of the local contrast of the pixel points, the second window optimization factor of the pixel points in the fluorescence microscopic grayscale image is obtained, and the specific steps include: A first window optimization factor corresponding to the pixel and a local contrast corresponding to the pixel are obtained, and a calculation result of dividing the local contrast corresponding to the pixel by the first window optimization factor corresponding to the pixel is used as a second window optimization factor corresponding to the pixel.

6. The method for processing fluorescent probe detection images based on noise analysis according to claim 1, characterized in that: According to the pixel points in the fluorescence microscopy grayscale image are evaluated by the first window optimization factor and the second window optimization factor, an adaptive local window side length of the pixel points is obtained, and the specific steps include: Get the set window limit constant, the first window optimization factor corresponding to the pixel point and the second window optimization factor corresponding to the pixel point; multiply the window limit constant, the first window optimization factor corresponding to the pixel point and the second window optimization factor corresponding to the pixel point and round down the result as the second window side length, multiply the constant 2 by the second window side length and add the result to the constant 3 as the adaptive local window side length corresponding to the pixel point.

7. The method for processing fluorescent probe detection images based on noise analysis according to claim 1, characterized in that: According to the method of performing window content analysis on the adaptive local window of the pixel point to obtain the local adaptive contrast threshold of the pixel point, the specific steps include: Obtain a first window optimization factor corresponding to the pixel, a second window optimization factor corresponding to the pixel, and a local contrast corresponding to the pixel; divide the first window optimization factor corresponding to the pixel by the second window optimization factor corresponding to the pixel as a contrast threshold evaluation factor, add a constant one to the contrast threshold evaluation factor and multiply the result by the local contrast corresponding to the pixel as a local adaptive contrast threshold of the pixel.

8. The method for processing fluorescent probe detection images based on noise analysis according to claim 1, characterized in that: The fluorescence microscopic grayscale image is segmented according to the adaptive local window through the pixels and the local adaptive contrast threshold to obtain a binary segmented image corresponding to the fluorescence microscopic grayscale image, wherein the specific steps include: The side length of the adaptive local window corresponding to the pixel point and the local adaptive contrast threshold of the pixel point are obtained, and the local adaptive window of the pixel point in the fluorescence microscopy grayscale image is determined according to the side length of the adaptive local window corresponding to the pixel point. The average grayscale value of all pixels in the local adaptive window corresponding to the pixel point and the local adaptive contrast threshold corresponding to the pixel point are added together as the segmentation threshold of the pixel point, and the grayscale value of the pixel point is compared with the segmentation threshold of the pixel point to obtain a binary segmentation label of the pixel point, complete the binary segmentation of the fluorescence microscopy image, and obtain a binary segmented image corresponding to the fluorescence microscopy grayscale image.

9. The method for processing fluorescent probe detection images based on noise analysis according to claim 8, characterized in that: According to the step of comparing the gray value of the pixel with the segmentation threshold of the pixel, a binary segmentation label of the pixel is obtained, and the specific steps include: Obtain the segmentation threshold of the pixel point. If the grayscale value of the pixel point is greater than or equal to the segmentation threshold of the pixel point, mark the pixel point as a target pixel point; if the grayscale value of the pixel point is less than the segmentation threshold of the pixel point, mark the pixel point as a background pixel point.

Images