Automatic inspection method and device

Abstract

The invention relates to the technical field of clinical laboratory inspection, discloses a clinical laboratory equipment management method, and particularly relates to an automatic inspection method and device. The quality of the target image is improved by performing quality evaluation on the collected target image and performing updating according to the deviation of the image in brightness and contrast, and the updated target image is automatically classified through an automatic detection method, so that the efficiency and accuracy of medical examination image processing are improved.

Description

Technical Field

[0001] The present invention relates to the field of laboratory testing technology, and is a method for managing laboratory equipment, and in particular, to an automated testing method and device. Background Art

[0002] Medical images are technologies and analytical tools that use different medical imaging devices to obtain visual representations of the internal structures and organs of the human body. Their purpose is to provide strong support for medical research, diagnosis and treatment. Microscopes, as an important medical imaging instrument, can magnify the observed objects, helping doctors and researchers to gain in-depth understanding and analysis of the characteristics of objects at the microscopic level. Medical microscopic images, that is, histological or cytological images of the inside of an organism captured by a microscope, play an important role in many aspects such as pathology, genetics, immunology and biomedicine, and help clinical personnel to explore the microscopic structure and function of the organism in more depth.

[0003] However, the diagnosis and analysis of tissue and cytology microscopic images are usually time-consuming, resource-intensive, and very challenging. In actual clinical practice, on the one hand, the analysis of tissue and cytology microscopic images is mainly based on manual reading, which requires researchers to review them for a long time and consumes a lot of energy, resulting in high labor costs; on the other hand, the results of clinical diagnosis and analysis are often subjective, and the diversity of target morphology and the differences in researchers' experience often lead to increased diagnostic rates. Summary of the invention

[0004] In order to solve the above technical problems and improve the speed and accuracy of the detection results of microbial detection, an automated inspection method and device are provided to realize the automated classification of the acquired microscopic images, thereby achieving high efficiency and high accuracy of detection.

[0005] In order to achieve the above purpose, the technical solution adopted in the embodiment of the present application is as follows:

[0006] In a first aspect, an automated inspection method is provided, which is applied to an automated microbiological inspection system, the system comprising a microscope objective, an image acquisition device and a server, the image acquisition device being used to obtain a target image in a microscope objective window and transmitting the target image to the server, the method being applied to the server, comprising: determining a brightness evaluation value and a contrast evaluation value of the target image, and determining whether the target image meets requirements in terms of brightness and contrast, and adjusting the brightness and / or contrast of the target image according to the difference; performing feature extraction on the target image that meets the requirements to obtain a feature map, and classifying the microbial tissue in the microbial image according to the feature map to obtain a classification result.

[0007] Further, determining the brightness evaluation value of the target image includes: obtaining the average brightness and the required average brightness of the target image, and determining the brightness evaluation value of the target image based on the following formula: where μ x is the average brightness of the target image, μ y To require average brightness, C 1 The constant is taken as 0.1.

[0008] Further, determining the contrast evaluation value of the target image includes: obtaining a standard deviation and a required standard deviation of the target image, and determining the contrast evaluation value of the target image based on the following formula: where σ x is the standard deviation of the target image, σ y To require the standard deviation, C 2 The constant is taken as 0.04.

[0009] Further, adjusting the brightness and / or contrast and / or structure of the target image according to the difference includes: obtaining a brightness adjustment value according to a brightness difference-compensation mapping relationship, obtaining a contrast adjustment value according to a contrast difference-compensation mapping relationship, and adjusting the brightness and / or contrast of the target image according to the brightness adjustment and / or the contrast adjustment value.

[0010] Furthermore, extracting features from the target image that meets the requirements to obtain a feature map includes: extracting features of the target image, and generating feature maps of different scales based on a special pyramid.

[0011] Furthermore, the classifying the microbial tissue in the microbial image according to the feature map includes: generating candidate regions with different classification results, and performing segmentation based on the candidate regions based on sparse coding to obtain segmentation results.

[0012] Furthermore, the generating of candidate regions of different classification results includes: generating an initial candidate region, and performing preliminary category prediction and bounding box regression to update the candidate region to obtain the candidate region.

[0013] Furthermore, the segmentation based on the candidate area is performed based on sparse coding to obtain a segmentation result, including: performing region of interest alignment processing on the candidate area based on a multi-stage cascade to achieve prediction of masks, categories and bounding boxes, and in the final stage, the sparse coding guided by the sparse coding of the base shape dictionary learning realizes the prediction of sparse coding and final mask to obtain a segmentation result.

[0014] Furthermore, generating the initial candidate region includes: setting corresponding multiple different candidate region sizes and corresponding aspect ratios on feature maps of different scales, so that each pixel point of each feature map generates multiple candidate regions with different aspect ratios, and finally generating multiple initial candidate regions of different sizes and aspect ratios.

[0015] In a second aspect, an automated inspection device is provided, which is applied to an automated microbiological inspection system. The system includes a microscope objective, an image acquisition device and a server. The image acquisition device is used to obtain a target image in a microscope objective window and transmit the target image to the server. The device is applied to the server and includes: an image correction module, which is used to determine a brightness evaluation value and a contrast evaluation value of the target image, and determine whether the target image meets the requirements in terms of brightness and contrast, and adjust the brightness and / or contrast of the target image according to the difference; a classification module, which is used to extract features of the target image that meets the requirements to obtain a feature map, and classify the microbial tissue in the microbial image according to the feature map to obtain a classification result.

[0016] In the technical solution provided in the embodiment of the present application, the quality of the target image collected is improved by performing quality evaluation on the target image and updating the image according to the deviation in brightness and contrast, and the updated target image is automatically classified through an automated detection method, thereby improving the efficiency and accuracy of medical examination image processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0018] The methods, systems and / or programs in the accompanying drawings will be further described according to exemplary embodiments. These exemplary embodiments will be described in detail with reference to the drawings. These exemplary embodiments are non-limiting exemplary embodiments, wherein example numbers represent similar mechanisms in the various views of the accompanying drawings.

[0019] Figure 1 It is a schematic diagram of the structure of the automated microbiological testing system provided in the embodiment of the present application.

[0020] Figure 2 It is a flow chart of the automated inspection method provided in the embodiment of the present application.

[0021] Figure 3 It is a schematic diagram of the structure of the automated inspection device provided in an embodiment of the present application.

[0022] Figure 4 It is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0023] In order to better understand the above technical scheme, the technical scheme of the present application is described in detail below through the accompanying drawings and specific embodiments. It should be understood that the embodiments of the present application and the specific features in the embodiments are detailed descriptions of the technical scheme of the present application, rather than limitations on the technical scheme of the present application. In the absence of conflict, the embodiments of the present application and the technical features in the embodiments can be combined with each other.

[0024] In the following detailed description, numerous specific details are set forth by way of example in order to provide a comprehensive understanding of the relevant guidance. However, it will be apparent to those skilled in the art that the present application may be practiced without these details. In other cases, well-known methods, procedures, systems, compositions and / or circuits have been described at a relatively high level, without detail, in order to avoid unnecessarily obscuring aspects of the present application.

[0025] Flowcharts are used in the present application to illustrate the execution process performed by the system according to the embodiment of the present application. It should be clearly understood that the execution process of the flowchart may not be performed in order. On the contrary, these execution processes may be performed in reverse order or simultaneously. In addition, at least one other execution process may be added to the flowchart. One or more execution processes may be deleted from the flowchart.

[0026] Before further describing the embodiments of the present invention in detail, the nouns and terms involved in the embodiments of the present invention are described. The nouns and terms involved in the embodiments of the present invention are subject to the following interpretations.

[0027] (1) In response, it is used to indicate the conditions or states on which the executed operations depend. When the dependent conditions or states are met, one or more operations executed may be in real time or with a set delay. Unless otherwise specified, there is no restriction on the order in which the multiple operations executed are executed.

[0028] (2) Based on is used to indicate the conditions or states on which the executed operations depend. When the dependent conditions or states are met, one or more operations executed may be in real time or have a set delay. Unless otherwise specified, there is no restriction on the order in which the multiple operations executed are executed.

[0029] Microbiological testing devices are essential tools for hospital laboratories and laboratories. They obtain the target tissues required for the test items by obtaining the morphological information of the microbial tissues, and analyze the target tissues to obtain physiological information about the microbial tissue objects. For example, the concentration of white blood cell tissues in urine can be used to determine whether a patient has the possibility of inflammation, and the concentration of red blood cell tissues in urine can be used to determine whether a patient has the possibility of hematuria. Therefore, microbiological testing devices are tools for biochemical testing based on bioimaging. In actual situations, hospitals, especially large hospitals, have many demands for testing patients. Traditional testing departments use equipment that mainly uses chemical staining and analytical instruments to analyze microbial tissues in body fluids. This process includes multiple processing steps, mainly including shaking, staining, retention and analysis, and generally takes about 10 minutes. With the application of machine vision technology in medical products, some hospitals have now configured new testing equipment to obtain biochemical information by acquiring microbial tissue images and segmenting the target tissues in the images based on image processing technology, and finally obtain test results. This solution is implemented based on machine vision technology, which is currently a new type of artificial intelligence technology. The automated microbial testing device in the embodiment of the present application is a new testing device based on this technology. This device uses any device from any manufacturer in the existing technology, and this device can achieve fully automated operation without the need for human intervention.

[0030] Because this solution analyzes and judges by acquiring images of microbial tissues, the quality of the images of microbial tissues is the basis for judgment and analysis. Based on the applicant's daily use of this device, the applicant found that because this device obtains images by taking pictures of microbial tissues, and different inspection items have different target microbial tissues, their morphological characteristics are distributed differently, so when this device is performing image acquisition, the image quality is different due to the morphological characteristics of different microbial tissues, which leads to the need to continuously perform manual image acquisition and adjustment of images of different types of microbial tissues, which is also a technical problem that the current device needs to solve. However, this method has the following defects: (1) The labor cost is high, so that the overall analysis process is not significantly shortened; (2) The manual re-judgment is only based on the overall expression of the image, such as the overall clarity of the image, but the required clarity of different microbial tissue morphological characteristics for different inspection items is different, and manual re-judgment requires a large cost. Therefore, based on the above two defects, this solution cannot optimally solve the existing problems.

[0031] Therefore, based on the above technical background, the embodiment of the present application provides an automated microbial testing system 100, which automatically re-evaluates the images collected by the microbial testing device to ensure that the clarity of the collected images meets the processing requirements, and automatically processes the images, thereby achieving the effect of automated testing throughout the entire process, and solving the problems of relying on manual secondary re-evaluation and inaccurate image analysis results in the prior art.

[0032] See also Figure 1 The system includes a microscope objective lens 110, an image acquisition device 120 and a server 130, wherein the image acquisition device acquires microorganism images based on the transmitted light information.

[0033] In addition to the above-mentioned devices, the main improvement in the embodiment of the present application is that a server is provided, and the server communicates with the image detection device to obtain the target image captured by the image acquisition device, and execute the automated inspection method based on the target image. For the automated inspection method provided in this embodiment, please refer to Figure 2 , including the following steps:

[0034] Step S21. Determine the brightness evaluation value and contrast evaluation value of the target image, determine whether the target image meets the requirements in terms of brightness and contrast, and adjust the brightness and / or contrast of the target image according to the difference.

[0035] In this embodiment, brightness and contrast are used to evaluate whether the target image meets the requirements. Specifically, the brightness evaluation value and contrast evaluation value corresponding to the target image are obtained respectively, and whether the brightness and contrast need to be adjusted is determined according to the brightness evaluation value and the contrast evaluation value.

[0036] Specifically, the brightness evaluation value of the target image is determined by obtaining the average brightness of the target image and the required average brightness, and based on the following formula:

[0037] where μ x is the average brightness of the target image, μ y To require average brightness, C 1 The constant is taken as 0.1.

[0038] The contrast evaluation value of the target image is determined by obtaining the standard deviation and the required standard deviation of the target image and based on the following formula: where σ x is the standard deviation of the target image, σ y To require the standard deviation, C 2 The constant is taken as 0.04.

[0039] In this embodiment, the above processing method can be used to obtain evaluation values ​​corresponding to the brightness and contrast of the target image, respectively. The evaluation values ​​are used to evaluate whether the brightness and contrast of the target image meet the requirements.

[0040] The difference between the corresponding evaluation value and the standard evaluation value is used as the brightness difference and contrast difference corresponding to the brightness and contrast respectively, and the brightness adjustment value is determined according to the brightness difference-compensation mapping relationship, and the contrast adjustment value is determined through the contrast difference-compensation mapping relationship, and the brightness and / or contrast of the target image is adjusted according to the brightness adjustment value and the contrast adjustment value so that the target image meets the requirements.

[0041] Step S22: extracting features from the target image that meets the requirements to obtain a feature map, and classifying the microbial tissues in the microbial image according to the feature map to obtain a classification result.

[0042] In the present embodiment, this step is implemented through a segmentation classification model. The segmentation classification model includes a backbone network, a special pyramid network, an RPN network, and a fusion cascade network, wherein the backbone network is used to extract features from the target image, and the special pyramid network is used to generate feature maps of different scales to enhance the multi-scale representation capability, thereby better obtaining the input feature maps of the RPN network and the fusion cascade network. The RPN network follows the design of the Mask RCNN network, and is used to generate pre-selected candidate regions and perform preliminary category prediction and bounding box regression to obtain refined higher-quality candidate regions, which are input into the fusion cascade network. The fusion cascade network introduces a multi-stage cascade and fusion architecture and a sparse coding guided branch based on shape dictionary learning, which can improve the detection and segmentation accuracy of the overall network to generate more accurate instance segmentation results.

[0043] Specifically, the backbone network uses a ResNet network, which contains 5 levels C1-C5. The feature maps of the last four levels C2-C5 are selected as the input of the FPN network, and the number of channels is (256, 512, 1024, 2048). The FPN network first performs a 1*1 convolution operation on the four output feature maps of C2-C5 to change the number of channels to 256, and then adds them to the feature maps of the corresponding scale in the upsampling path, thereby enhancing the multi-scale information of the feature map. In addition, the FPN network also adds a downsampling path to further enhance the global information of a larger scale. Finally, the FPN network uses a 3*3 convolution on the feature maps of the five levels to obtain the five output feature maps of P2-P6, and the downsampling rates are {4, 8, 16, 32, 64} respectively.

[0044] The RPN network is used to generate a priori candidate regions and perform corresponding category prediction and bounding box regression, and finally outputs candidate regions with higher quality after refinement. First, the RPN network sets 5 different candidate region sizes (32, 64, 128, 256, 512) and corresponding 3 aspect ratios (0.5, 1.0, 2.0) on the five output feature maps of P2-P6, so that each pixel of each feature map can generate 3 candidate regions with different aspect ratios. At the same time, due to the different size settings of different feature maps, a series of initial candidate regions with different sizes and aspect ratios can be generated, and the category prediction and bounding box regression of the initial candidate regions are performed. First, 3x3 convolution is performed on the five feature maps and input into the classification branch and bounding box regression branch. The classification branch uses 1x1 convolution with 3 output channels, indicating the probability that the candidate region corresponding to the pixel in each feature map is the foreground. The bounding box regression branch uses 1x1 convolution with 12 output channels, indicating the 4 predicted position offsets of the 3 candidate regions. Finally, the RPN network refines the candidate regions and performs alignment operations to obtain higher quality candidate regions as the input of the fusion cascade network.

[0045] The overall prediction process of the fusion cascade network is as follows: (1) In the first stage, the network performs region of interest alignment operations on the feature maps of different scales output by the FPN network according to the initial candidate regions generated by the RPN network, and obtains fixed-size feature maps as the input of the fusion cascade network to achieve mask, category and bounding box prediction; (2) In the second and third stages, the network adopts a similar structure and prediction process to realize multi-stage cascading to improve detection accuracy; (3) At the same time, feedforward fusion operations are introduced in multiple stages, that is, the input feature map in the current stage is fused with the feedforward output feature map of the previous stage to improve segmentation accuracy; (4) In the fourth stage, the network realizes the prediction of sparse coding and final instance mask respectively, and uses the instance shape attention weight map reconstructed by the predicted sparse coding to perform explicit shape guidance on the input feature map, thereby further improving the accuracy and shape integrity of the instance mask.

[0046] In this embodiment, sparse coding is to represent the instance mask as a linear combination of a set of coefficient weight vectors and a set of pre-learned super-complete shape dictionaries. The shape dictionary is obtained by pre-setting and an online dictionary learning method is performed in each actual prediction process. The shape dictionary in this embodiment can be constructed by using the dictionary construction method in the prior art, and its data can be obtained by training with a medical data set, which will not be described in detail in this embodiment.

[0047] An automated inspection method is provided for step S21-step S22, which improves the quality of the target image by evaluating the quality of the acquired target image and updating it according to the deviations in brightness and contrast of the image, and automatically classifies the updated target image through an automated detection method, thereby improving the efficiency and accuracy of medical inspection image processing.

[0048] See also Figure 3 Based on step S21-step S22, a virtual device, namely an automated inspection device 30, is provided. The device is configured in the server and is used to execute the processing of step S21-step S22. The device includes the following units:

[0049] An image correction module 31 is used to determine a brightness evaluation value and a contrast evaluation value of the target image, determine whether the target image meets the requirements in terms of brightness and contrast, and adjust the brightness and / or contrast of the target image according to the difference;

[0050] The classification module 32 is used to extract features from the target image that meets the requirements to obtain a feature map, and classify the microbial tissues in the microbial image according to the feature map to obtain a classification result.

[0051] See also Figure 4 , the above method can also be integrated into the provided electronic device 400. In view of the fact that the device may have relatively large differences due to different configurations or performances, it can include one or more processors 401 and memory 402, and the memory 402 can store one or more storage applications or data. Among them, the memory 402 can be a temporary storage or a permanent storage. The application stored in the memory 402 may include one or more modules (not shown in the figure), and each module may include a series of computer executable instructions in the electronic device. Furthermore, the processor 401 can be configured to communicate with the memory 402, and the electronic device executes a series of computer executable instructions in the memory 402. The electronic device may also include one or more power supplies 403, one or more wired / wireless network interfaces 404, one or more input / output interfaces 405, one or more keyboards 406, etc.

[0052] In a specific embodiment, the electronic device includes a memory and one or more programs, wherein the one or more programs are stored in the memory, and the one or more programs may include one or more modules, and each module may include a series of computer executable instructions for the electronic device, and the one or more programs configured to be executed by one or more processors include the following computer executable instructions:

[0053] Determine a brightness evaluation value and a contrast evaluation value of the target image, determine whether the target image meets the requirements in terms of brightness and contrast, and adjust the brightness and / or contrast of the target image according to the difference;

[0054] Feature extraction is performed on the target image that meets the requirements to obtain a feature map, and the microbial tissues in the microbial image are classified according to the feature map to obtain a classification result.

[0055] The following is a detailed introduction to the various components of the processor:

[0056] In this embodiment, the processor is an application specific integrated circuit (ASIC), or is configured to implement one or more integrated circuits of the embodiments of the present application, such as one or more microprocessors (digital signal processor, DSP), or one or more field programmable gate arrays (field programmable gate array, FPGA).

[0057] Optionally, the processor may execute various functions by running or executing a software program stored in the memory and calling data stored in the memory, such as executing the above-mentioned Figure 3 The method shown.

[0058] In a specific implementation, as an embodiment, the processor may include one or more microprocessors.

[0059] Among them, the memory is used to store the software program that executes the solution of the present application, and the execution is controlled by the processor. The specific implementation method can refer to the above method embodiment and will not be repeated here.

[0060] Optionally, the memory may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, 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, but is not limited thereto. The memory may be integrated with the processor, or may exist independently and be coupled to the processing unit through the interface circuit of the processor, and the embodiments of the present application do not specifically limit this.

[0061] It should be noted that the structure of the processor shown in this embodiment does not constitute a limitation on the device, and the actual device may include more or fewer components than shown in the figure, or combine certain components, or arrange the components differently.

[0062] In addition, the technical effects of the processor can refer to the technical effects of the method described in the above method embodiment, which will not be repeated here.

[0063] It should be understood that the processor in the embodiments of the present application may be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), field programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0064] It should also be understood that the memory in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example and not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic random access memory (DRAM), synchronous DRAM (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0065] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (for example, a floppy disk, a hard disk, a tape), an optical medium (for example, a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0066] In this application, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can be represented by: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.

[0067] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0068] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0069] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0070] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0071] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0072] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0073] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical disks.

[0074] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. An automated inspection method, characterized in that: The method is applied to an automated microbiological inspection system, the system comprising a microscope objective lens, an image acquisition device and a server, the image acquisition device is used to acquire a target image in a microscope objective lens window and transmit the target image to the server, and the method is applied to the server, comprising: Determine a brightness evaluation value and a contrast evaluation value of the target image, determine whether the target image meets the requirements in terms of brightness and contrast, and adjust the brightness and / or contrast of the target image according to the difference; Feature extraction is performed on the target image that meets the requirements to obtain a feature map, and the microbial tissues in the microbial image are classified according to the feature map to obtain a classification result.

2. The automated inspection method according to claim 1, characterized in that: Determining the brightness evaluation value of the target image includes: obtaining the average brightness and the required average brightness of the target image, and determining the brightness evaluation value of the target image based on the following formula: where μ x is the average brightness of the target image, μ y To require average brightness, C1 is a constant and is taken as 0.

1.

3. The automated inspection method according to claim 1, characterized in that: Determining the contrast evaluation value of the target image includes: obtaining a standard deviation and a required standard deviation of the target image, and determining the contrast evaluation value of the target image based on the following formula: where σ x is the standard deviation of the target image, σ y To require the standard deviation, C2 is taken as a constant and is set to 0.

04.

4. The automated inspection method according to claim 1, characterized in that: The step of adjusting the brightness and / or contrast and / or structure of the target image according to the difference includes: obtaining a brightness adjustment value according to a brightness difference-compensation mapping relationship, obtaining a contrast adjustment value according to a contrast difference-compensation mapping relationship, and adjusting the brightness and / or contrast of the target image according to the brightness adjustment value and / or the contrast adjustment value.

5. The automated inspection method according to claim 1, characterized in that: The step of extracting features from the target image that meets the requirements to obtain a feature map includes: extracting features of the target image and generating feature maps of different scales based on a special pyramid.

6. The automated inspection method according to claim 5, characterized in that: The classifying the microbial tissue in the microbial image according to the feature map includes: generating candidate regions with different classification results, and performing segmentation based on the candidate regions based on sparse coding to obtain segmentation results.

7. The automated inspection method according to claim 6, characterized in that: The generating of candidate regions of different classification results includes: generating an initial candidate region, and performing preliminary category prediction and bounding box regression to update the candidate region to obtain the candidate region.

8. The automated inspection method according to claim 6, characterized in that: The method of segmenting the candidate region based on sparse coding to obtain a segmentation result includes: performing region of interest alignment processing on the candidate region based on a multi-stage cascade to achieve prediction of masks, categories and bounding boxes, and in the final stage, the sparse coding guided by the learning of a base shape dictionary to achieve prediction of sparse coding and final mask to obtain a segmentation result.

9. The automated inspection method according to claim 6, characterized in that: The generating of the initial candidate region includes: setting a corresponding plurality of different candidate region sizes and corresponding aspect ratios on feature maps of different scales, so that each pixel point of each feature map generates a plurality of candidate regions with different aspect ratios, and finally generating a plurality of initial candidate regions with different sizes and aspect ratios.

10. An automated inspection device, characterized in that: Applied to an automated microbiological inspection system, the system includes a microscope objective lens, an image acquisition device and a server, the image acquisition device is used to obtain a target image in the microscope objective lens window and transmit the target image to the server, the device is applied to the server, and includes: An image correction module, used to determine a brightness evaluation value and a contrast evaluation value of the target image, determine whether the target image meets the requirements in terms of brightness and contrast, and adjust the brightness and / or contrast of the target image according to the difference; The classification module is used to extract features from the target image that meets the requirements to obtain a feature map, and classify the microbial tissues in the microbial image according to the feature map to obtain a classification result.

Images