Diffusion model-oriented high-density fluorescence microscopic image fungal cell population detection method

Abstract

The invention relates to the technical field of computer vision and biomedical image processing, and discloses a diffusion model-oriented high-density fluorescent microscopic image fungal cell population detection method, which comprises the following steps of: performing multi-scale illumination estimation and self-adaptive compensation on an input fluorescent microscopic image, and correcting illumination nonuniformity in the image; local density distribution adjustment is carried out on the image, and density equalization is realized through energy generic function minimization; target detection is carried out based on an improved Diffusion Det framework, and the improved Diffusion Det framework optimizes a conditional diffusion process through a dynamic noise scheduling strategy and extracts cellular morphological features in combination with a multi-scale feature fusion network; and calculating a spatial relationship and feature similarity between cells, and constructing an affinity matrix to optimize a detection result. According to the invention, the detection accuracy of the high-density overlapped cells is improved.

Description

Technical Field

[0001] This application relates to the technical fields of computer vision and biomedical image processing. Specifically, it is a method for detecting fungal cell populations in high-density fluorescence microscopy images for diffusion models, which is particularly suitable for detecting scenarios of high-density, overlapping cells and low signal-to-noise ratio fluorescence images. Background Art

[0002] The existing cell detection technologies mainly have the following problems: 1. Traditional object detection algorithms perform poorly in dealing with high-density and overlapping cells: Region proposal-based methods are prone to generating duplicate detections in overlapping regions; Anchor box-based methods are difficult to adapt to cells with irregular shapes; Single-stage detectors are prone to missing detections in high-density regions. Specifically, traditional object detection algorithms show significant limitations in dealing with high-density and overlapping fungal cell populations: (1) Two-stage detection methods: Algorithms represented by Faster R-CNN have a misdetection rate as high as 35% when dealing with overlapping regions. Especially in regions where the cell density exceeds 200 cells / mm², the problem of duplicate detections is more prominent. (2) Anchor box-based detection methods: When dealing with irregular shapes, the detection accuracy generally drops by 20%-30%. For structures such as hyphae with an aspect ratio exceeding 5:1, the detection recall rate is as low as 40%. (3) Single-stage detectors: Although the inference speed is increased by 3-4 times, the missing detection rate in high-density regions (>300 cells / mm²) reaches 25%, and the recognition accuracy of cell growth states is only 65%.

[0003] 2. The inherent imaging characteristics of fluorescence microscopy images pose challenges: low signal-to-noise ratio, blurred cell boundaries; uneven illumination leads to inconsistent image contrast; background fluorescence interference seriously affects detection accuracy. Specifically, fluorescence microscopy technology faces the following specific problems in fungal observation: The signal-to-noise ratio (SNR) of typical fluorescence images is only 8-12 dB, far lower than 20-25 dB of conventional optical images; The uneven illumination results in a brightness difference of up to 40% between the central and edge regions of the image; The autofluorescence intensity of the fungal cell wall can reach 15-20% of the target signal; During long-term observation, the signal intensity decays by about 8-10% per hour due to the photobleaching effect.

[0004] 3. Existing methods are difficult to effectively model the spatial relationship between cells: lack of modeling of cell population distribution characteristics; underutilization of local structural information; detection results are easily affected by uneven density distribution. Specifically, existing algorithms have obvious defects in processing the distribution characteristics of fungal populations: for typical exponentially growing fungal communities, the density of the central area can be 3-5 times that of the edge; when existing spatial modeling methods process this density gradient, the detection accuracy of the edge area is 25% lower than that of the central area; in multi-scale feature fusion, the computational complexity increases exponentially with the number of feature layers, and it takes 2-3 seconds to process a 1024×1024 resolution image.

[0005] In recent years, the DiffusionDet algorithm based on the diffusion model has shown unique advantages in the field of target detection. By introducing the diffusion process, the algorithm can gradually optimize the position and size of the detection box, effectively solving the performance bottleneck of traditional detectors when dealing with dense targets. The main advantages of DiffusionDet are: breaking through the dependence of traditional detectors on prior boxes and being able to adaptively learn the spatial distribution of targets; improving detection accuracy through an iterative refinement process, especially when dealing with overlapping targets; and having an end-to-end training paradigm that simplifies the model optimization process.

[0006] However, there are still several key challenges when directly applying DiffusionDet to high-density fungal fluorescence microscopy image detection: first, the algorithm is very sensitive to image quality and signal-to-noise ratio, and its performance is unstable when processing low-quality fluorescence images, making it difficult to cope with the imaging characteristics of fungal samples; second, the algorithm is not optimized for the special morphology and growth characteristics of fungal cells, and lacks the ability to model the dynamic changes of cell populations; finally, the iterative calculation of the diffusion process brings a large computational overhead, which is inefficient when processing large-scale fungal samples. Summary of the invention

[0007] The purpose of the present application is to provide a method for detecting fungal cell groups using high-density fluorescence microscopic images oriented to a diffusion model, so as to solve the technical problems raised in the above-mentioned background technology.

[0008] To achieve the above-mentioned purpose, the present application discloses the following technical solution: a method for detecting fungal cell groups using high-density fluorescence microscopic images oriented to a diffusion model, the method comprising the following steps: Step 1: Perform multi-scale illumination estimation and adaptive compensation on the input fluorescence microscopy image through the fluorescence signal enhancement module to correct the illumination inhomogeneity in the image; Step 2: Use the density equalization optimization system to adjust the local density distribution of the image and achieve density equalization by minimizing the energy function; Step 3: Perform object detection based on the improved DiffusionDet framework. The improved DiffusionDet framework optimizes the conditional diffusion process through a dynamic noise scheduling strategy and combines a multi-scale feature fusion network to extract cell morphological features; Step 4: Calculate the spatial relationship and feature similarity between cells through a local affinity module, and construct an affinity matrix to optimize the detection results.

[0009] Preferably, in the step 1, the fluorescence signal enhancement module is specifically configured as: A. Use four Gaussian kernels with different scales for multi-scale illumination estimation to generate illumination components with different receptive fields; B. Based on the adaptive weight fusion of local variance, the weight calculation satisfies the following constraints: Constraint 1: The sum of weights at each scale is 1; Constraint 2: The range of a single weight is ; Constraint 3: The difference between adjacent scale weights is less than or equal to 0.3; C. Dynamically calculate the compensation coefficient according to the local illumination intensity , where: When , ; When , ; When , .

[0010] Preferably, in the fluorescence signal enhancement module, the multi-scale feature map of the input image is generated by four Gaussian kernels with standard deviations of being 3, 7, 15, and 31, and the corresponding receptive fields are 9×9, 21×21, 45×45, and 93×93 pixels respectively; the weight calculation is obtained through the following formula: where, is the local variance of the scale feature map at scale , and is the weight adjustment parameter.

[0011] Preferably, in the step 2, the energy functional adopted by the density equilibrium optimization system is specifically: where, is the regularization parameter, is the target density level, and is the image region a certain point in the local cell density at the location is the density gradient

[0012] Preferably, in the energy functional, the regularization parameter is and the target density level is set to and the maximum number of iterations is 100

[0013] Preferably, in the step 3, the improved DiffusionDet framework specifically includes a. Dynamic noise scheduling strategy: The noise intensity increases linearly with the time step and the calculation formula is where is the total number of diffusion steps b. Multi-scale feature fusion network: Different-level feature fusion is performed through the formula and the feature weights are optimized by combining the channel attention mechanism, where is the fused feature map is the high-level feature map represents the upsampling operation on the high-level features represents the lateral connection feature map of the current layer represents the convolution operation

[0014] Preferably, the backbone network of the multi-scale feature fusion network is ResNet50, and a feature pyramid network is integrated for multi-scale feature extraction

[0015] Preferably, in the step 4, the calculation method of the affinity matrix of the local affinity module is specifically where is the feature vector of cell is the feature vector of cell represents calculating the feature similarity between cell and cell is and the dimension of is the spatial position coordinate of cell is the spatial position coordinate of cell is the spatial decay coefficient

[0016] ​​​​​Advantages: The method for detecting fungal cell populations in high-density fluorescence microscopic images for diffusion models in this application improves the detection accuracy of high-density overlapping cells, effectively overcomes the imaging defects of fluorescence microscopic images, fully utilizes the spatial distribution information of cell populations, and enhances the robustness of the detection model under uneven density distributions. Specifically, by estimating the illumination component using a multi-scale Gaussian kernel, combining an adaptive weight fusion and piecewise compensation strategy, the problem of uneven illumination is effectively corrected; based on the energy functional, the local density distribution is dynamically adjusted to solve the problems of missed detection in high-density regions and false detection in low-density regions; by improving the DiffusionDet framework, the recall rate of hypha detection is increased, the detection rate of tiny cells is improved, and the detection efficiency is enhanced; through the design of the local affinity module, the false detection rate is reduced, and the detection accuracy of the colony edge region is improved. Brief Description of the Drawings

[0017] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only some embodiments of this application. For those skilled in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0018] Figure 1 It is a flowchart of the method for detecting fungal cell populations in high-density fluorescence microscopic images for diffusion models provided by an embodiment of this application; Figure 2 It is a schematic diagram of the operation of the system framework corresponding to the method for detecting fungal cell populations in high-density fluorescence microscopic images for diffusion models provided by an embodiment of this application; Figure 3 It is a schematic diagram of the working principle of the fluorescence signal enhancement module provided by an embodiment of this application; Figure 4 It is a schematic diagram of the detection result obtained using the prior art; Figure 5 It is a schematic diagram of the detection result obtained using the method for detecting fungal cell populations in high-density fluorescence microscopic images for diffusion models provided by this embodiment. Detailed Embodiments

[0019] The following will clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the described embodiments are only some embodiments of this application, rather than all embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of this application.

[0020] In this article, the term "comprising" is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of more restrictions, the elements defined by the sentence "comprising..." do not exclude the presence of other identical elements in the process, method, article or device including the elements.

[0021] This embodiment discloses a Figure 1 The high-density fluorescence microscopy method for detecting fungal cell populations based on the diffusion model shown in the paper is combined with Figures 2 - 5 , this method is introduced in detail. Figure 2 The four core modules of the whole system are demonstrated: Fluorescence Signal Enhancement Module (FSE), Density Adaptive Enhancement Module (DAE), Improved DiffusionDet Detection Module and Local Affinity Module (LAM). These modules process the input fluorescence microscopy images in sequence and finally output the cell detection results. Figure 3 It shows that the fluorescence signal enhancement module adopts a multi-scale illumination estimation strategy, estimates the illumination components through four different Gaussian kernels, integrates the multi-scale results through an adaptive weight fusion method, and finally obtains the enhanced image through illumination compensation.

[0022] Specifically, the method comprises the following steps: Step 1: Perform multi-scale illumination estimation and adaptive compensation on the input fluorescence microscopy image through the fluorescence signal enhancement module to correct the illumination inhomogeneity in the image; Step 2: Use the density equalization optimization system to adjust the local density distribution of the image and achieve density equalization by minimizing the energy function; Step 3: Target detection is performed based on an improved DiffusionDet framework, wherein the improved DiffusionDet framework optimizes the conditional diffusion process through a dynamic noise scheduling strategy and extracts cell morphological features in combination with a multi-scale feature fusion network; Step 4: Calculate the spatial relationship and feature similarity between cells through the local affinity module, and construct an affinity matrix to optimize the detection results.

[0023] In the actual fluorescence microscopy imaging process, due to the influence of various factors such as light source distribution, sample thickness variation, and the characteristics of the optical system itself, the acquired images often exhibit obvious illumination non-uniformity. This non-uniformity not only affects the visual quality of the images but also interferes with the subsequent image analysis and processing. Based on this, this embodiment adopts a collaborative working method of multi-scale illumination estimation, adaptive weight fusion, and compensation coefficient calculation. That is, in the step 1, the fluorescence signal enhancement module is specifically configured as follows: A - Multi-scale illumination estimation: Four Gaussian kernels with different scales are used for multi-scale illumination estimation to generate illumination components with different receptive fields. Specifically, this step includes: Set four reference scales for illumination estimation, and the standard deviation of each Gaussian kernel is 3, 7, 15, and 31 respectively, that is The multi-scale feature maps of the input image are generated by Gaussian kernels with different scales. The specific expression for generating the Gaussian kernel is: The formula for calculating the kernel size is: The receptive fields corresponding to Gaussian kernels with different scales are 9×9, 21×21, 45×45, and 93×93 pixels respectively.

[0024] B - Adaptive weight fusion: Adaptive weight fusion based on local variance. Specifically, the weight calculation is obtained through the following formula: where is the local variance of the scale feature map at scale , is the weight adjustment parameter.

[0025] And the weight calculation satisfies the following constraints: Constraint 1: The sum of weights at each scale is 1; Constraint 2: The range of a single weight is ; Constraint 3: The difference between adjacent scale weights is less than or equal to 0.3.

[0026] The core idea of multi-scale illumination estimation is to comprehensively capture illumination change characteristics through image analysis at different scales. Through the four designed reference scales Gaussian kernels of different sizes cover illumination variations at various scales, from the internal structure of cells to large-scale tissue regions. Using the formula for weight calculation ensures a smooth transition in the spatial domain and avoids artifacts caused by sudden changes. In the specific implementation, the design of calculating the kernel size ensures that each Gaussian kernel can cover 99.7% of the effective information region while maintaining computational efficiency. Through these four-scale settings, receptive fields of 9×9, 21×21, 45×45, and 93×93 pixels are formed respectively, which can correspond to the illumination change characteristics at different spatial scales. Specifically: The first scale (9×9 pixels) is mainly responsible for capturing the most subtle illumination changes, which is particularly important for retaining the fine structure inside cells; the second scale (21×21 pixels) focuses on illumination changes within the size range of a single complete cell and can effectively smooth local illumination fluctuations; the third scale (45×45 pixels) mainly deals with illumination differences within the range of multiple cell groups; the fourth scale (93×93 pixels) is responsible for capturing the gradually changing illumination patterns over a larger range to ensure accurate estimation of the overall illumination trend.

[0027] C Compensation coefficient calculation: The compensation coefficient calculation system adopts a piecewise function strategy to dynamically calculate the compensation coefficient according to the local illumination intensity . Among them, Basic compensation coefficient Piecewise calculation of: Adaptive bandwidth calculation: Among them, the bandwidth constraint is: Minimum bandwidth: h min = 15, Maximum bandwidth: h max = 60.

[0028] The piecewise strategy adopted is: Dark area processing (L < 0.2): That is, when , ; Use the maximum compensation coefficient, aiming to significantly enhance the signal intensity in the dark area and avoid information loss in the dark area; Middle area processing (0.2 ≤ L ≤ 0.8): Adopt an inverse proportion compensation strategy. Feasibly, when , , so that the compensation intensity is dynamically adjusted with the illumination intensity to ensure a smooth transition and avoid sudden changes; Bright area processing (L > 0.8): That is, when , , Use the minimum compensation coefficient to prevent signal saturation caused by over-enhancement and maintain the original signal characteristics.​ In this embodiment, the energy functional adopted by the density balance optimization system in step 2 is specifically as follows: For the energy functional, its parameter settings are specifically as follows: the regularization parameter is defaulted to 0.1, the target density level is set to 0.5, is the local cell density at a certain point in the image region , and is the density gradient, and the maximum number of iterations is 100.

[0029] Furthermore, in the energy functional, the local cell density is normalized to [0, 1], where 0 represents no cells and 1 represents the highest density, and its calculation method is obtained by counting the cell detection results, such as the number of cells detected per unit area. The target density level is used to force the optimized density distribution to tend to be globally uniform, avoiding detection errors caused by local over-density or under-density. The reason for setting its value to 0.5 is: assuming that under ideal conditions, the fungal cell population is uniformly distributed in space, so the intermediate value is set to balance the high-density and low-density regions. The regularization parameter is used to balance the weights of the two terms in the energy function. The first term : drives the density to approach the target value; the second term : constrains the smoothness of the density change. The experimental optimization results show that through cross-validation, when λ = 0.1, the best balance can be achieved between density balance and smoothness. The larger the λ value, the stronger the influence of the density gradient, and the smoother the density distribution, but local details may be sacrificed. The density gradient represents the rate of change of density in space, which can be calculated by discrete difference approximation or Sobel operator in the prior art. Its physical meaning is that the larger the gradient, the more drastic the density mutation. Minimizing the gradient square term can suppress density mutation and ensure a smooth transition of the optimized density distribution. For the maximum number of iterations, it is set to 100 to ensure that the function converges within a finite number of steps and avoid infinite loops. Experiments have shown that in most cases, the energy function has tended to be stable after 100 iterations.

[0030] In order to fully consider the particularity of fluorescence microscopy images and significantly improve the performance of the model in cell detection tasks. In this embodiment, in-depth improvements and optimizations are made to the traditional DiffusionDet object detection framework. Specifically, the improvement process is as follows: (1) Conditional diffusion process During the conditional diffusion process, a diffusion mechanism based on the Gaussian distribution is adopted, and the diffusion process is controlled through a designed probability distribution function In this distribution function, The term controls the degree of retention of the original signal, while The term determines the intensity of the injected noise. This design not only ensures the controllability of the diffusion process but also maintains sufficient randomness, enabling the model to fully explore the possible target state space. In practical applications, this balance is crucial for accurately detecting cells with different brightnesses and morphologies.

[0031] To more precisely control the diffusion process, a dynamic noise scheduling strategy is designed, where the noise intensity increases linearly with the time step The calculation formula is: where, is the total number of diffusion steps.

[0032] This strategy uses a relatively small noise intensity at the initial stage of diffusion and gradually increases the noise level as time progresses. This progressive way of adding noise is particularly suitable for the characteristics of fluorescence microscopy images. At the initial stage of diffusion, the smaller noise helps to retain the fine structure and morphological features of the cells; while in the later stage, the larger noise provides sufficient randomness, enabling the model to better handle different cell morphologies and distributions. By setting a base noise level of 0.1 and a maximum increment of 0.9, it is ensured that the noise intensity remains within a reasonable range throughout the diffusion process.

[0033] (2) Feature extraction network Feasibly, the backbone network of the multi-scale feature fusion network is ResNet50, and a feature pyramid network is integrated for multi-scale feature extraction. The formula for the multi-scale feature fusion network is: The multi-scale feature fusion network is used to fuse features at different levels and combines a channel attention mechanism to optimize the feature weights, where, is the fused feature map, is the high-level feature map, coming from the deeper layers of the backbone network (such as ResNet50), represents the upsampling operation on the high-level features, for example, bilinear interpolation (BilinearUpsampling) is used for upsampling, represents the lateral connection feature map of the current layer, coming from the shallower layers of the backbone network, represents the convolution operation, which is used to fuse the upsampled high-level features and the lateral-connected low-level features. In terms of the feature extraction network, the fusion mechanism of the multi-scale feature fusion network includes an upsampling operation and lateral connections Two key components. The upsampling operation is responsible for passing the high-level semantic features to the current layer, while the lateral connections retain the local detail information of the current layer. Finally, these features are adaptively fused through the convolutional operation Conv() to generate a richer and more effective feature representation.

[0034] Calculation example: Assume the input is a 1024×1024 image and the backbone network is ResNet50: : Feature map from Stage5 (32×32 resolution, 2048 channels); : Bilinearly upsampled to 64×64 resolution, with the number of channels still 2048; : Feature map from Stage3 (64×64 resolution, 512 channels), adjusted to 2048 channels through 1×1 convolution; Fusion and convolution: Element-wise add the upsampled feature map (64×64×2048) and the lateral feature (64×64×2048); Generate the final feature map (64×64×256) through a 3×3 convolution (output channels 256).

[0035] The design of this feature extraction network is particularly suitable for cell detection tasks in fluorescence microscopy images. In actual fluorescence microscopy images, cells often exhibit diverse morphological features: some cells are regular circles or ellipses, while others have irregular shapes; some cell boundaries are clear, while others are relatively blurred; at the same time, there are significant differences in cell sizes, and from tiny organelles to large cell clusters may appear in the same image. Through the designed multi-scale feature fusion mechanism, the network can capture these features of different scales and morphologies simultaneously, thus providing more accurate detection results. In practical applications, the detection recall rate of irregular cells (such as mycelia) has been increased from 40% to 75%; the detection rate of tiny cells (diameter <5μm) has been increased from 55% to 80%, the multi-scale fusion module has increased the feature extraction speed by 50%, and the processing time for a single image has been shortened to 0.8 seconds (1024×1024 resolution).

[0036] In addition, the problem of uneven illumination commonly exists in fluorescence microscopy images, which results in cells with significantly different brightness levels potentially existing in the same image. To address this challenge, an attention mechanism is introduced into the feature extraction network. This mechanism can adaptively adjust the feature weights of different spatial positions and channels, enabling the network to better focus on those image regions that are significant for cell detection, even if the brightness of these regions varies significantly.

[0037] Through these carefully designed improvements, the improved DiffusionDet framework demonstrates excellent performance, especially when dealing with complex fluorescence microscopy images. Experimental results show that the improved framework can not only accurately detect cells of different sizes and morphologies but also exhibits good robustness to problems such as noise and uneven illumination in the images. These improvements provide a more reliable and efficient tool for the field of biomedical image analysis.

[0038] In this embodiment, the local affinity module involves affinity feature extraction and affinity matrix calculation.

[0039] Specifically, affinity feature extraction is achieved through non-local feature calculation, and the calculation formula is: The calculation of the affinity matrix is obtained through an improved affinity metric, and the calculation formula for the corresponding improved affinity metric is: Among them, is the feature vector of cell , which is extracted from the input feature map through a convolutional layer, is the feature vector of cell , which is extracted from the same feature map through another convolutional layer, represents calculating the feature similarity between cell and cell , is and 's dimension, by default , the higher the similarity, the greater the affinity. is the spatial position coordinate of cell , is the spatial position coordinate of cell , is to calculate the squared Euclidean distance between cells, the closer the distance, the stronger the spatial correlation. is the spatial attenuation coefficient, which controls the sensitivity of position correlation, with a default value of 5, the larger it is, the weaker the spatial attenuation, allowing cells at a greater distance to be associated. the smaller it is, only the association of neighboring cells (close distance) is significant.

[0040] Calculation example: Suppose the cells and have the following characteristics and positions: Feature vector: is [0.8, 0.2], is [0.7, 0.3]. For simplicity of calculation, take a simplified example; Position coordinates: is (10, 20), is (12, 22), σ = 5.

[0041] Calculation steps: Feature similarity score: ≈0.438 Spatial decay term: ≈0.852 Softmax normalization: Suppose the sum of similarity scores of other cell pairs is 1.5, then: ≈0.65 Final affinity: ≈0.554 Through the design of the local affinity module, through calculation and experiments, etc., it can be obtained that the continuous modeling of the cell population reduces the false detection rate by 15%, improves the detection accuracy of the colony edge area by 25%, and reduces the computational complexity of the local affinity module by 70% through sparse calculation (only processing adjacent cell pairs).

[0042] From the above, it can be seen that the method for detecting fungal cell populations in high-density fluorescence microscopy images provided in this embodiment has the following technical advantages: Significantly improves the detection accuracy of high-density overlapping cells; effectively overcomes the imaging defects of fluorescence microscopy images; makes full use of the spatial distribution information of the cell population; improves the robustness of the detection model under uneven density distributions.

[0043] In the embodiments provided in the present application, it should be understood that the embodiments described herein can be implemented in hardware, software, firmware, middleware, code, or any suitable combination thereof. For hardware implementation, the processor can be implemented in one or more of the following units: application specific integrated circuit (ASIC), digital signal processor (DSP), digital signal processing device (DSPD), programmable logic device (PLD), field programmable gate array (FPGA), processor, controller, microcontroller, microprocessor, or other electronic units designed to implement the functions described herein, or a combination thereof. For software implementation, part or all of the processes of the embodiments can be completed by instructing the relevant hardware through a computer program. When implemented, the above program can be stored in a computer-readable storage medium or transmitted as one or more instructions or codes on a computer-readable storage medium. The computer-readable storage medium includes computer storage media and communication media, where the communication media includes any medium that facilitates the transfer of a computer program from one place to another. The storage media can be any available medium that can be accessed by a computer. The computer-readable storage medium can include, but is not limited to, RAM, ROM, EEPROM, CD-ROM, or other optical disc storage, magnetic disk storage media, or other magnetic storage devices, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer.

[0044] Finally, it should be noted that the above are only the preferred embodiments of the present application and are not used to limit the present application. Although the present application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A method for detecting fungal cell populations using high-density fluorescence microscopic images based on a diffusion model, characterized in that: The method comprises the following steps: Step 1: Perform multi-scale illumination estimation and adaptive compensation on the input fluorescence microscopy image through the fluorescence signal enhancement module to correct the illumination inhomogeneity in the image; Step 2: Use the density equalization optimization system to adjust the local density distribution of the image and achieve density equalization by minimizing the energy function; Step 3: Target detection is performed based on an improved DiffusionDet framework, wherein the improved DiffusionDet framework optimizes the conditional diffusion process through a dynamic noise scheduling strategy and extracts cell morphological features in combination with a multi-scale feature fusion network; Step 4: Calculate the spatial relationship and feature similarity between cells through the local affinity module, and construct an affinity matrix to optimize the detection results.

2. The method for detecting fungal cell populations using high-density fluorescence microscopic images based on a diffusion model according to claim 1, characterized in that: In step 1, the fluorescence signal enhancement module is specifically configured as follows: A. Use four Gaussian kernels of different scales for multi-scale illumination estimation to generate illumination components of different receptive fields; B. Adaptive weight fusion based on local variance. The weight calculation satisfies the following constraints: Constraint 1: The sum of the weights of each scale is 1; Constraint 2: The range of a single weight is ; Constraint 3: The difference in weights of adjacent scales is less than or equal to 0.3; C. According to the local lighting intensity Dynamic calculation of compensation coefficient ,in: when hour, ; when hour, ; when hour, .

3. The method for detecting fungal cell populations using high-density fluorescence microscopic images based on a diffusion model according to claim 2, characterized in that: In the fluorescence signal enhancement module, the multi-scale feature map of the input image is obtained by The Gaussian kernels of four different scales of 3, 7, 15, and 31 are generated, and the corresponding receptive fields are 9×9, 21×21, 45×45, and 93×93 pixels respectively; the weight calculation is obtained by the following formula: in, The scale feature map is at scale The local variance under is the weight adjustment parameter.

4. The method for detecting fungal cell populations using high-density fluorescence microscopic images based on a diffusion model according to claim 1, characterized in that: In step 2, the energy functional used by the density balance optimization system is specifically: in, is the regularization parameter, is the target density level, For the image area A point in The local cell density at is the density gradient.

5. The method for detecting fungal cell populations using high-density fluorescence microscopic images based on a diffusion model according to claim 4, characterized in that: In the energy functional, the regularization parameter is , the target density level is set to , the maximum number of iterations is 100.

6. The method for detecting fungal cell populations using high-density fluorescence microscopic images based on a diffusion model according to claim 1, characterized in that: In step 3, the improved DiffusionDet framework specifically includes: a. Dynamic noise scheduling strategy: noise intensity With time step Linear increase, the calculation formula is: ,in, is the total number of diffusion steps; b. Multi-scale feature fusion network: through formula Perform feature fusion at different levels and optimize feature weights in combination with the channel attention mechanism, where: is the fused feature map, is a high-level feature map, Indicates upsampling of high-level features. represents the lateral connection feature map of the current layer, Represents a convolution operation.

7. The method for detecting fungal cell populations using high-density fluorescence microscopic images based on a diffusion model according to claim 6, characterized in that: The backbone network of the multi-scale feature fusion network is ResNet50, and a feature pyramid network is integrated to perform multi-scale feature extraction.

8. The method for detecting fungal cell populations using high-density fluorescence microscopic images based on a diffusion model according to claim 1, characterized in that: In step 4, the affinity matrix of the local affinity module is calculated as follows: in, For cells The characteristic vector of For cells The characteristic vector of Represents computational cells and cells The similarity of features between for and The dimension of For cells The spatial position coordinates of For cells The spatial position coordinates of is the spatial attenuation coefficient.

Images