A sample barcode scanning system for a fully automatic coagulation tester

Abstract

The present invention discloses a sample barcode scanning system for a fully automatic coagulation tester, specifically relating to the technical field of barcode scanning; by analyzing the optical reflection characteristics of the sample barcode and the thermal imaging data of the conveying track, accurate judgment of the abnormal state of the barcode and the number of test tubes is achieved. According to the determination result, the scanning mode is automatically selected. In the multi-tube scanning mode, a clustering analysis algorithm is used to evaluate the spatial layout of the barcodes, calculate the average minimum distance between barcodes, and evaluate the overall scanning reliability to determine whether the scanning quality meets the standard, effectively improving the barcode recognition accuracy and detection stability, and meeting the requirements of modern coagulation tests for automatic high-precision scanning.

Description

Technical Field

[0001] The present invention relates to the technical field of barcode scanning, and more specifically, to a sample barcode scanning system for a fully automatic coagulation tester. Background Art

[0002] In a fully automatic coagulation tester, the accuracy of sample tube barcode scanning directly determines the accuracy of sample information recognition. In the prior art, there are problems in sample barcode scanning such as inconsistent barcode positions and sizes due to differences among tube manufacturers, and mis-scanning or missed scanning caused by multiple tubes entering the scanning area simultaneously; meanwhile, the steam generated by the coagulation reagent and cleaning liquid may cause abnormal optical properties on the surface of the barcode material, resulting in abnormal optical refraction of the barcode label and reducing the recognition accuracy. These problems seriously affect the stability of the detection process and data accuracy.

[0003] To solve the above problems, a technical solution is provided now. Summary of the Invention

[0004] To overcome the above-mentioned defects of the prior art, an embodiment of the present invention provides a sample barcode scanning system for a fully automatic coagulation tester to solve the problems raised in the above background art.

[0005] To achieve the above object, the present invention provides the following technical solution:

[0006] A sample barcode scanning system for a fully automatic coagulation tester, comprising: a reflection characteristic analysis module, a thermal imaging data analysis module, a scanning mode control module, a spatial independence evaluation module, and a scanning reliability evaluation module:

[0007] The reflection characteristic analysis module analyzes the optical reflection characteristics of the sample barcode to determine whether there is abnormal optical refraction caused by reagent steam on the surface of the sample label;

[0008] The thermal imaging data analysis module analyzes the data of the thermal imaging sensor on the sample conveying track to determine whether there are multiple sample tubes entering the scanning area simultaneously;

[0009] The scanning mode control module determines the sample barcode scanning mode based on the judgment results of whether there is abnormal optical refraction caused by reagent steam on the surface of the sample label and whether there are multiple sample tubes entering the scanning area simultaneously in the scanning area;

[0010] When the sample barcode scanning mode is the multi-tube simultaneous scanning mode, the spatial independence evaluation module evaluates the spatial independence of the sample barcodes in the scanning area by analyzing the spatial layout of the sample barcodes in the scanning area and using a clustering analysis algorithm;

[0011] Based on the spatial independence of the sample barcode, the scanning reliability evaluation module evaluates the overall scanning reliability of the sample barcode in the current scanning environment and determines whether the scanning quality requirements are met.

[0012] In a preferred embodiment, the optical reflection characteristics of the sample barcode are analyzed to determine whether there is an optical refraction anomaly caused by reagent vapor on the surface of the sample label. Specifically:

[0013] Collect the reflected optical image data on the surface of the sample barcode;

[0014] Perform preprocessing of noise filtering and gray-level equalization on the collected image;

[0015] Use the spectrum analysis algorithm to extract the reflected optical characteristic parameters;

[0016] Compare the extracted reflected optical characteristic parameters with the parameters of the standard optical reflection model to determine whether there is a refraction anomaly.

[0017] In a preferred embodiment, the thermal imaging sensor data on the sample transport track is analyzed to determine whether multiple sample tubes enter the scanning area simultaneously. Specifically:

[0018] Collect the thermal image data output by the thermal imaging sensor on the sample transport track;

[0019] Perform temperature range constraint on the thermal image data to filter out the background invalid thermal signals;

[0020] Perform temperature gradient enhancement on the filtered thermal image data to strengthen the thermal characteristics of the sample tube;

[0021] Use the region growing algorithm to separate the candidate sample tube regions;

[0022] Calculate the center coordinates and boundary sizes of the candidate sample tube regions;

[0023] Based on multi-object tracking, determine whether multiple sample tubes enter the scanning area simultaneously.

[0024] In a preferred embodiment, based on the judgment results of whether there is an optical refraction anomaly caused by reagent vapor on the surface of the sample label and whether multiple sample tubes enter the scanning area simultaneously in the scanning area, determine the sample barcode scanning mode. Specifically:

[0025] Perform combined analysis on the optical refraction anomaly judgment result and the sample tube quantity judgment result;

[0026] Construct a scanning mode decision matrix and set the scanning mode corresponding to each determination result combination: When the optical state is normal and only a single sample tube enters the scanning area, it is set to the normal single-tube scanning mode; when the optical state is abnormal and only a single sample tube enters the scanning area, it is set to the optical abnormal single-tube mode; when the number of sample tubes is multiple, it is set to the multi-tube simultaneous scanning mode.

[0027] In a preferred embodiment, when the sample barcode scanning mode is the multi-tube simultaneous scanning mode, by analyzing the spatial layout of the sample barcodes in the scanning area, the density clustering analysis algorithm is used to evaluate the spatial independence of the sample barcodes. Specifically:

[0028] Extract the central coordinates of each sample barcode in the scanning area;

[0029] Based on the coordinate information of each sample barcode, construct a sample barcode spatial coordinate point set;

[0030] Perform density clustering analysis on the sample barcode spatial coordinate point set to identify the sample barcode aggregation areas;

[0031] Calculate the average minimum distance between barcodes within each sample barcode aggregation area;

[0032] Judge whether the average minimum distance meets the set threshold and output the spatial independence evaluation result.

[0033] In a preferred embodiment, based on the spatial independence of the sample barcodes, evaluate the overall scanning reliability of the sample barcodes in the current scanning environment and judge whether it meets the scanning quality requirements. Specifically:

[0034] Receive the average minimum distance and spatial independence evaluation result of the sample barcodes within each clustering cluster;

[0035] Based on the total number of clustering clusters, evaluate the spatial distribution density of the sample barcodes in the scanning area;

[0036] Analyze the deviation degree of the average minimum distance of the sample barcodes within each clustering cluster from the minimum safety spacing threshold;

[0037] Combine the spatial distribution density and the average minimum spacing deviation degree to calculate the environmental complexity index;

[0038] Perform correlation analysis on the environmental complexity index and the barcode spatial independence data;

[0039] Based on the correlation analysis result, output the overall scanning reliability of the sample barcodes in the current scanning environment and judge whether it meets the scanning quality requirements.

[0040] In a preferred embodiment, based on the association analysis results, the overall scanning reliability of the sample barcode in the current scanning environment is output, and it is determined whether the scanning quality requirements are met. Specifically:

[0041] A preset overall scanning reliability threshold is set, and the overall scanning reliability index is compared with the overall scanning reliability threshold:

[0042] When the overall scanning reliability index is greater than or equal to the overall scanning reliability threshold, it indicates that the overall scanning reliability of the sample barcode in the current scanning environment is high and meets the scanning quality requirements;

[0043] When the overall scanning reliability index is less than the overall scanning reliability threshold, it indicates that the overall scanning reliability of the sample barcode in the current scanning environment is low and does not meet the scanning quality requirements.

[0044] The technical effects and advantages of the sample barcode scanning system of a fully automatic coagulation tester of the present invention:

[0045] By comprehensively analyzing the optical reflection characteristics of the sample barcode and the thermal imaging data on the sample transport track, the problems caused by differences among test tube manufacturers resulting in inconsistent barcode positions and sizes, multiple test tubes entering the scanning area simultaneously, and optical refraction anomalies caused by the influence of coagulation reagents or cleaning solution vapor on the barcode material are solved. The sample barcode scanning system can automatically determine the optical state and the number of test tubes, and flexibly switch the scanning mode according to the determination results; in the multi-tube simultaneous scanning mode, the spatial independence of the barcodes is evaluated through a clustering analysis algorithm, and the minimum distance and distribution state between the barcodes are effectively quantified, thereby evaluating the overall scanning reliability. When the overall scanning reliability index reaches the preset threshold, it ensures a high scanning accuracy, significantly reduces the phenomena of mis-scanning and missed scanning, and improves the stability and detection efficiency of barcode recognition. It not only improves the operation accuracy and adaptability of the fully automatic coagulation tester, but also reduces the risks of human intervention and environmental interference. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] Figure 1 It is a schematic structural diagram of a sample barcode scanning system of a fully automatic coagulation tester of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0047] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention. Embodiment

[0048] Figure 1The present invention provides a sample barcode scanning system for a fully automatic coagulation analyzer, including: a reflection characteristic analysis module, a thermal imaging data analysis module, a scanning mode control module, a spatial independence evaluation module, and a scanning reliability evaluation module:

[0049] The reflection characteristic analysis module analyzes the optical reflection characteristics of the sample barcode to determine whether there is an abnormal optical refraction caused by reagent vapor on the surface of the sample label;

[0050] The thermal imaging data analysis module analyzes the thermal imaging sensor data on the sample conveying track to determine whether multiple sample tubes enter the scanning area simultaneously;

[0051] Based on the determination results of whether there is an abnormal optical refraction caused by reagent vapor on the surface of the sample label and whether multiple sample tubes enter the scanning area simultaneously in the scanning area, the scanning mode control module determines the sample barcode scanning mode;

[0052] When the sample barcode scanning mode is the multi-tube simultaneous scanning mode, the spatial independence evaluation module evaluates the spatial independence of the sample barcodes by analyzing the spatial layout of the sample barcodes in the scanning area and using the clustering analysis algorithm;

[0053] Based on the spatial independence of the sample barcodes, the scanning reliability evaluation module evaluates the overall scanning reliability of the sample barcodes in the current scanning environment and determines whether the scanning quality requirements are met.

[0054] Specifically, analyzing the optical reflection characteristics of the sample barcode to determine whether there is an abnormal optical refraction caused by reagent vapor on the surface of the sample label includes:

[0055] Collecting the reflected optical image data on the surface of the sample barcode: The image sensor of the coagulation analyzer irradiates the surface of the sample barcode with a light source in a specific wavelength band, and records the two-dimensional reflected optical image of the area where the sample barcode is located through a high-resolution photosensitive element. The collected original reflected optical image is denoted as ; where represents the gray value at the pixel coordinate ; represents the horizontal pixel coordinate of the image; represents the vertical pixel coordinate of the image.

[0056] Performing noise filtering and gray level equalization preprocessing on the collected image: Since the sample tube may be transported multiple times inside the coagulation analyzer or affected by external ambient light interference, the original reflected optical image usually contains problems such as random noise and uneven light distribution, and noise filtering and gray level equalization preprocessing are required.

[0057] First, the median filtering algorithm is used to perform noise reduction processing on the original reflected optical image to obtain a pre-filtered image: ; among which, represents the image processed by the median filtering algorithm; represents the median filtering algorithm, which replaces the original value by taking the median of all pixel gray values in its neighborhood at each pixel position, thereby eliminating isolated noise.

[0058] Next, perform gray level equalization processing on the image processed by the median filtering algorithm to improve the image contrast and highlight the subtle reflection characteristics. Define the gray level equalization transformation function, then the image after gray level equalization processing is: ; among which, represents the image processed by gray level equalization; represents the gray level equalization transformation function, and its implementation principle is to redistribute the image histogram, making the utilization of each gray level in the image more sufficient and the details more obvious.

[0059] Use the spectral analysis algorithm to extract the reflection optical characteristic parameters: convert the image processed by gray level equalization to the frequency domain through the fast Fourier transform, and extract the characteristic parameters of the reflection optics. Define the transformed frequency domain representation as:

[0060] ; among which, represents the image the frequency domain representation after the fast Fourier transform, including amplitude and phase information; represents the horizontal frequency coordinate in the frequency domain image, describing the frequency components of the image in the horizontal direction; represents the vertical frequency coordinate in the frequency domain image, describing the frequency components of the image in the vertical direction; represents the imaginary unit; represents the total number of horizontal pixels of the image; represents the total number of vertical pixels of the image.

[0061] After the Fourier transform, calculate the spectral amplitude diagram: ; among which, represents the amplitude at the frequency domain coordinate reflecting the energy magnitude of this frequency component.

[0062] Extract the parameters related to the optical reflection characteristics in the image through the local spectral analysis method. Define the set of extracted optical reflection characteristic parameters as: ; among which, represents the set of optical reflection characteristic parameters extracted from the spectral amplitude diagram; represents the total number of the extracted optical reflection characteristic parameters.

[0063] By analyzing the distribution and statistical characteristics of each parameter in the set of optical reflection characteristic parameters, local refraction anomalies caused by the formation of a tiny liquid film or bubble on the surface of the sample label due to reagent vapor can be captured. Its essential manifestation is the shift or abnormal increase in the energy distribution in certain frequency bands of the spectrum, providing a quantitative basis for the judgment of refraction anomalies.

[0064] Compare the extracted reflection optical characteristic parameters with the parameters of the standard optical reflection model to determine whether there is a refraction anomaly: To determine whether there is an optical refraction anomaly in the acquired image, the set of extracted reflection optical characteristic parameters is compared with the set of parameters of the pre-established standard optical reflection model. The standard model parameter set is defined as: ; where represents the set of parameters of the pre-established standard optical reflection model, that is, the set of ideal optical characteristic parameters on the surface of the label barcode without the interference of reagent vapor; represents the total number of parameters of the standard optical reflection model, which is consistent with the number in the set of optical reflection characteristic parameters to ensure a one-to-one correspondence for comparison.

[0065] To quantitatively compare the set of reflection optical characteristic parameters with the set of parameters of the standard optical reflection model, calculate the average deviation value, and its calculation formula is: ; where represents the calculated average deviation value, which is used to quantitatively describe the difference between the set of reflection optical characteristic parameters and the set of parameters of the standard optical reflection model; represents the number of parameters participating in the comparison. Usually , and some key features can be selected for comparison according to actual needs; represents the expected value of the th optical reflection characteristic parameter; represents the

[0066] Set a fluctuation threshold to determine whether the deviation exceeds the normal fluctuation range, that is, when the average deviation value is greater than the fluctuation threshold, it is determined that there is an optical refraction anomaly caused by reagent vapor on the surface of the sample label.

[0067] Specifically, analyze the data of the thermal imaging sensor on the sample transport track to determine whether multiple sample tubes enter the scanning area simultaneously, including:

[0068] Collect the thermal image data output by the thermal imaging sensor on the sample transport track: Collect the original thermal image data through the thermal imaging sensor installed on the sample transport track. Define the collected original thermal image data as ; where represents the original thermal image data; represents the pixel coordinates in the horizontal direction of the original thermal image data, with a value range from 0 to , is the total number of horizontal pixels in the original thermal image data; represents the pixel coordinates in the vertical direction of the original thermal image data, with a value range from 0 to , is the total number of vertical pixels in the original thermal image data.

[0069] Using a highly sensitive thermal imaging sensor, the thermal image data generated by the sample tube on the sample transport track due to temperature distribution is captured in real time to ensure that the thermal image data has sufficient spatial resolution and temperature details.

[0070] Perform temperature range constraint on the thermal image data to filter out background invalid thermal signals: To remove invalid or interfering thermal signals in the background, perform temperature range constraint processing on the original thermal image data. Define the temperature screening function as ; where represents the thermal image data after temperature range constraint processing; represents the temperature screening function, which is used to perform pixel-level judgment on the input thermal image data. When the temperature value of a certain pixel in the thermal image data is lower than the predetermined lower limit or higher than the predetermined upper limit, the pixel is set to the background value (such as 0 or other set values); represents the lower limit threshold in temperature screening, which is determined by experiments to ensure that low-temperature background signals are filtered out; represents the upper limit threshold in temperature screening, which is used to filter out noise signals with too high temperatures to ensure that only the temperature range related to the sample tube is retained in the thermal image data.

[0071] Through temperature range constraint processing, ensure that only the effective temperature interval reflecting the thermal characteristics of the sample tube is retained, and filter out invalid thermal signals caused by the environment, transport track background or other interference factors, thereby improving the accuracy of subsequent image processing.

[0072] Perform temperature gradient enhancement on the screened thermal image data to strengthen the thermal characteristics of the sample tube: To strengthen the thermal characteristics of the sample tube, perform temperature gradient enhancement processing on the thermal image data after temperature range constraint processing. Use the gradient enhancement algorithm to calculate the temperature gradient amplitude at each pixel. Define the enhanced thermal image data as:

[0073] ; where represents the thermal image data after temperature gradient enhancement processing; represents the gradient enhancement processing function, which usually uses an edge detection algorithm (such as the Sobel operator) to process the thermal image data internally. The purpose is to strengthen the edge area with large temperature changes, so as to more clearly present the temperature difference between the sample tube and the background.

[0074] To specifically describe the temperature gradient calculation, the local gradient magnitude is defined as:

[0075] ; where represents the local temperature gradient magnitude calculated at the pixel position and is used to reflect the severity of the local temperature change in the thermal image data, and can identify the contour edge of the sample tube; and represent the pixel coordinates in the gradient magnitude map, and the value range is the same as that of and ; corresponds to the horizontal coordinate, corresponds to the vertical coordinate.

[0076] Through the magnitude calculation, the edge of the sample tube and the thermal anomaly area can be highlighted, facilitating subsequent region segmentation.

[0077] Use the region growing algorithm to separate the candidate sample tube regions: The region growing algorithm is used to segment the gradient-enhanced thermal image data to separate the thermal imaging region contours of each sample tube and obtain the candidate sample tube regions. The segmentation result is defined as: ; where represents the set of all candidate sample tube regions obtained after segmentation by the region growing algorithm; represents the th candidate sample tube region obtained after segmentation; represents the total number of candidate sample tube regions detected after segmentation by the region growing algorithm.

[0078] After the initial seed pixel is selected, the region growing algorithm automatically expands the boundary of the candidate sample tube region according to the pixel temperature gradient and the neighborhood similarity criterion. Each candidate sample tube region is generated to meet the following conditions:

[0079] The temperature value of the seed pixel is higher than the background in the thermal image data after temperature gradient enhancement;

[0080] The difference between the temperature values of adjacent pixels and the seed pixel is within a predetermined tolerance range;

[0081] The region growing termination condition is that the temperature difference between adjacent pixels exceeds the tolerance range or reaches the maximum expansion range.

[0082] Accurate segmentation of the thermal imaging region of the sample tube is achieved, providing a clear region boundary for extracting the candidate sample tube regions.

[0083] Calculate the center coordinates and boundary dimensions of the candidate sample tube regions: For each obtained candidate sample tube region Perform geometric feature calculations to extract its center coordinates and boundary dimensions.

[0084] Define its center coordinates as and the regional boundary dimensions as ; where represents the center coordinates of the th candidate sample tube region, calculated as the average of the horizontal and vertical coordinates of all pixels in the region; represents the horizontal coordinate component of the center of the th candidate sample tube region, which is the average of the horizontal coordinates of all pixels in the region; represents the vertical coordinate component of the center of the th candidate sample tube region, which is the average of the vertical coordinates of all pixels in the region; represents the boundary dimensions of the th candidate sample tube region, including two dimensions of width and height; represents the width of the th candidate sample tube region, which is the difference between the maximum and minimum horizontal coordinates in the region; represents the height of the th candidate sample tube region, which is the difference between the maximum and minimum vertical coordinates in the region.

[0085] After calculating the center coordinates and boundary dimensions for all candidate sample tube regions, a regional feature set is obtained; where represents the set composed of the center coordinates and boundary dimensions of all candidate sample tube regions, with a total of elements.

[0086] By calculating the geometric features of each candidate sample tube region, the center coordinates and boundary dimensions are obtained, and they are formed into a set, providing basic data for target association and tracking, and used to determine whether each hot zone corresponds to an independent sample tube.

[0087] Based on multi-object tracking, determine whether there are multiple sample tubes entering the scanning area simultaneously: Based on the set of center coordinates and boundary dimensions, use multi-object tracking to perform association analysis on each candidate sample tube region to determine whether there are multiple sample tubes in the scanning area. Multi-object tracking is to continuously track and perform data association on candidate sample tube regions in a dynamic scene according to the input set of center coordinates and boundary dimensions, so as to determine the number of independent sample tubes actually present in the scanning area.

[0088] When the fully automatic coagulation analyzer starts running, each candidate sample tube area is initialized as an independent tracking target. An initial state is assigned to each tracking target, including its position in the image (such as the coordinates of the center of the area) and the initial velocity (usually set to zero or estimated based on experimental knowledge).

[0089] For each initialized tracking target, a prediction algorithm (such as the Kalman filter) is used to predict the possible position and motion state of the tracking target in the current frame based on the state of the previous frame. The state of the tracking target is inferred through a mathematical model (state transition matrix).

[0090] In the current frame, new observation data (such as the measured position of the tracking target) is obtained through thermal imaging or region segmentation, and then it is matched with the predicted state of the tracking target. The Euclidean distance is used to construct a cost matrix, and then the optimal matching algorithm (such as the Hungarian algorithm) is used to complete the correct association between the tracking target and the observation data, ensuring that each predicted target is as close as possible to the closest observation correspondence.

[0091] According to the data association result, the observation data in the current frame is combined with the predicted state, and the Kalman filter update formula is used to correct the state of the tracking target. The prediction error can be corrected, and the tracking accuracy can be improved, making the state information (position, velocity, etc.) of the tracking target closer to the real situation.

[0092] According to the matching situation of the tracking target in multiple consecutive frames, the tracking targets that are continuously successfully associated are confirmed as real existing targets; for the tracking targets that cannot match the observation data in multiple consecutive frames, it is considered that they have left the scanning area or are blocked, and they are processed as lost or deleted from the tracking list to prevent incorrect counting.

[0093] After completing the tracking of consecutive frames and target management, the number of tracking targets in the current effective tracking state is counted as the number of sample tubes in the scanning area for the final determination. When the number of tracking targets in the effective tracking state is greater than 1, it can be judged that multiple sample tubes enter the scanning area simultaneously.

[0094] Specifically, based on the judgment results of whether there is an optical refraction anomaly caused by reagent vapor on the surface of the sample label and whether multiple sample tubes enter the scanning area simultaneously in the scanning area, the sample barcode scanning mode is determined, including:

[0095] Combined analysis is performed on the optical refraction anomaly determination result and the sample tube quantity determination result: The optical refraction anomaly determination result includes the presence of optical refraction anomalies caused by reagent vapor, i.e., optical anomalies, and the absence of optical refraction anomalies caused by reagent vapor, i.e., optical normality. The sample tube quantity determination result includes the simultaneous entry of multiple sample tubes into the scanning area, i.e., the multi-tube state, and the absence of the simultaneous entry of multiple sample tubes into the scanning area, i.e., the single-tube state.

[0096] Preset logical relationships, based on the optical refraction anomaly state and the sample tube quantity state, to form clear state combinations. The situations of state combinations include: optical normality and single-tube state; optical anomaly and single-tube state; optical normality and multi-tube state; optical anomaly and multi-tube state. Through the preset logical relationships, the independent determination information of the optical refraction anomaly state and the sample tube quantity state is integrated to obtain clear combined states for the construction of the scanning mode decision matrix.

[0097] Construct a scanning mode decision matrix and set the corresponding scanning modes for each determination result combination: The scanning mode decision matrix takes the combined states obtained from the combined analysis as the horizontal axis and the specific scanning modes as the vertical axis. During the construction of the scanning mode decision matrix, in-depth research and testing are carried out on the scanning scenarios that may be caused by each combined state, and a mapping relationship is established between each combined state and the scanning mode: When the optical state is normal and only a single sample tube enters the scanning area, it is correspondingly set as the normal single-tube scanning mode; when the optical state is abnormal and only a single sample tube enters the scanning area, it is set as the optical anomaly single-tube mode to perform a more targeted compensation scanning; when the number of sample tubes is multiple, regardless of the optical refraction state, it is set as the multi-tube simultaneous scanning mode to ensure the effective identification of bar code information on each sample tube in a multi-target environment. The introduction of the decision matrix avoids complex conditional judgments in the traditional method and effectively improves the execution efficiency and reliability.

[0098] Specifically, when the sample bar code scanning mode is the multi-tube simultaneous scanning mode, by analyzing the spatial layout of the sample bar codes in the scanning area, the spatial independence of the sample bar codes is evaluated using the clustering analysis algorithm, including:

[0099] Extract the central coordinates of each sample bar code in the scanning area: Process the sample bar code images in the scanning area, and use image processing algorithms to accurately extract the geometric features of each bar code. After each sample bar code undergoes image segmentation, edge detection, and morphological processing, its central position is determined. Assume that each sample bar code corresponds to a feature record, and the feature record contains the central coordinates of the bar code, defined as: ; where represents the central coordinate of the th sample bar code; represents the The central position of a sample barcode in the horizontal direction in the sample barcode image, calculated from the barcode pixel distribution; represents the central position of the th sample barcode in the vertical direction in the sample barcode image, calculated from the barcode pixel distribution;

[0100] Based on the coordinate information of each sample barcode, construct a sample barcode spatial coordinate point set: After extracting the geometric features of each sample barcode, use the extracted central coordinate data to construct a sample barcode spatial coordinate point set, which reflects the spatial distribution of all sample barcodes in the scanning area and helps with clustering analysis. The central coordinates of all sample barcodes form a point set ; where represents the sample barcode spatial coordinate point set, and each element represents a two-dimensional coordinate, reflecting the position of the corresponding sample barcode in the scanning area.

[0101] By constructing the sample barcode spatial coordinate point set, organize the discrete sample barcode position data into a format convenient for clustering analysis, providing basic information for identifying whether there is spatial overlap or dense distribution of sample barcodes.

[0102] Perform density clustering analysis on the sample barcode spatial coordinate point set to identify the sample barcode aggregation areas: After obtaining the sample barcode spatial coordinate point set, use the density clustering analysis algorithm for analysis to identify the aggregation areas of the sample barcodes. The main goal of clustering analysis is to divide the sample barcode coordinates that are close to each other and have a high density into the same group. For this purpose, a density-based clustering method is adopted, and its core idea is: within a preset neighborhood radius, if the number of points in the neighborhood of a certain point exceeds a set threshold, then this point is considered to belong to a clustering cluster. Define the clustering cluster set as: ; where represents the clustering cluster set; represents the th clustering cluster, containing several barcode spatial coordinate points; represents the total number of clustering clusters.

[0103] The clustering algorithm uses neighborhood density calculation and groups the sample barcode spatial coordinate point set according to parameter settings (such as neighborhood radius and minimum number of neighborhood points). It not only identifies the local aggregation of barcodes but also provides a basis for calculating the spatial distance between barcodes within each aggregation area.

[0104] Calculate the average minimum distance between barcodes within the aggregation area of each sample barcode: After identifying the aggregation areas of each sample barcode, it is necessary to quantitatively analyze the spatial relationship between the sample barcodes within each aggregation area. To this end, calculate the minimum distance between all sample barcodes within each aggregation area and take the average of this distance to evaluate the distribution independence between the sample barcodes.

[0105] For each cluster, let the number of barcodes contained in the cluster be (where represents the serial number of the cluster), then for each sample barcode within the cluster, the calculation formula for its minimum distance is:

[0106] ; where, represents the minimum distance between the -th sample barcode within the cluster and all other barcodes; and respectively represent the horizontal coordinate and vertical coordinate of the -th sample barcode; is the index of other sample barcodes in the cluster except .

[0107] Then, take the average of the minimum distances of all sample barcodes within the cluster to obtain the average minimum distance: ; where, represents the average minimum distance between all sample barcodes within the cluster ; is the total number of sample barcodes in the cluster .

[0108] Judge whether the average minimum distance meets the set threshold and output the evaluation result of spatial independence: After obtaining the average minimum distance of the sample barcodes within each cluster, evaluate each cluster according to the pre-set minimum safety distance threshold.

[0109] According to the experimental analysis and the standard size of the sample barcode, set the minimum safety distance threshold , which reflects the minimum distance requirement necessary for the sample barcodes to be independently recognized in the scanning environment.

[0110] For each cluster, if its average minimum distance is greater than the minimum safety distance threshold, it is determined that the sample barcodes within the cluster have spatial independence; otherwise, it is determined that the sample barcodes within the cluster do not have spatial independence.

[0111] Specifically, based on the spatial independence of the sample barcodes, evaluate the overall scanning reliability of the sample barcodes in the current scanning environment and judge whether it meets the scanning quality requirements, including:

[0112] Receive the average minimum distance and the results of the spatial independence evaluation of the sample barcodes within each clustering cluster: The average minimum distance between all sample barcodes within the clustering cluster is ; represents the total number of clustering clusters.

[0113] Define the result of the spatial independence evaluation as: ; where represents the result of the spatial independence evaluation.

[0114] Based on the total number of clustering clusters, evaluate the spatial distribution density of the sample barcodes in the scanning area: Define the total area of the scanning area as , then the calculation formula for the spatial distribution density is: ; where represents the spatial distribution density of the sample barcodes; represents the total area of the scanning area.

[0115] The spatial distribution density can reflect the density of the sample barcodes in the scanning area. The greater the spatial distribution density, the closer the layout of the sample barcodes in the scanning area, and the greater the potential risk of occlusion and interference.

[0116] Analyze the deviation degree of the average minimum distance of the sample barcodes within each clustering cluster from the minimum safety spacing threshold: According to the minimum safety spacing threshold , calculate the minimum spacing deviation degree, and its calculation formula is: ; where represents the -th minimum spacing deviation degree of the clustering cluster.

[0117] If the minimum spacing deviation degree is large, it indicates that there is a strong risk of spatial overlap or barcode adhesion within the clustering cluster, which may reduce the scanning accuracy.

[0118] Combine the spatial distribution density and the average minimum spacing deviation degree to calculate the environmental complexity index: Consider the barcode spatial density and the minimum spacing deviation degree within each clustering cluster jointly, and calculate the environmental complexity index to comprehensively reflect the complexity of the current scanning environment.

[0119] Define the environmental complexity index as: ; where represents the environmental complexity index; and are weight coefficients, indicating the relative contributions of the spatial distribution density and the average minimum spacing deviation degree in the environmental complexity evaluation, satisfying .

[0120] The higher the environmental complexity index, the more complex the current scanning environment, and there is a higher risk of interference between sample barcodes.

[0121] Perform a correlation analysis on the environmental complexity index and the barcode spatial independence data: By analyzing the correlation between the environmental complexity index and the evaluation results of the spatial independence of sample barcodes, calculate the correlation index, which is used to evaluate the potential impact of environmental characteristics on the scanning accuracy. The calculation formula is: ; where represents the correlation index; represents the total number of sample barcodes; represents the evaluation results of spatial independence.

[0122] A larger correlation index indicates that the pre-scanning environment is more complex, which may increase the risk of misjudgment or failure in sample barcode scanning.

[0123] Based on the correlation analysis results, output the overall scanning reliability of sample barcodes in the current scanning environment: Define the overall scanning reliability index, and its calculation formula is: ; where represents the overall scanning reliability index; represents the preset maximum risk value, ranging from 0 to 1.

[0124] Preset the overall scanning reliability threshold and compare the overall scanning reliability index with the overall scanning reliability threshold:

[0125] When the overall scanning reliability index is greater than or equal to the overall scanning reliability threshold, it indicates that the overall scanning reliability of sample barcodes in the current scanning environment is high, the overall distribution of sample barcodes is uniform and the independence is good; at this time, the detected barcode spacing and environmental complexity indicators are both within the ideal range, the scanning data accuracy is high, the misjudgment and missed scanning phenomena are few, and the scanning quality requirements are met;

[0126] When the overall scanning reliability index is less than the overall scanning reliability threshold, it indicates that the overall scanning reliability of sample barcodes in the current scanning environment is low, there are large interferences in the scanning environment or problems such as dense or overlapping arrangements of sample barcodes, resulting in a decrease in recognition stability; at this time, the minimum distance between barcodes significantly deviates from the expected value, and the spatial independence is insufficient, which may cause recognition errors or missed scanning and does not meet the scanning quality requirements; it is necessary to trigger an alarm prompt or correction mechanism, and it is recommended to adjust the scanning parameters or optimize the environmental conditions to improve the scanning quality, ensure that the detection results meet the specified scanning quality requirements, and prevent data processing errors.

[0127] The setting of the overall scanning reliability threshold is based on a large amount of experimental data and actual application scenarios, and usually determined by statistical methods. First, by collecting a large amount of scanning data in a stable environment, the mean and variance of the overall scanning reliability index are calculated; secondly, a suitable median or mean is selected as the preliminary threshold in combination with the safety margin; finally, according to the detection requirements and instrument characteristics in actual applications, the threshold is appropriately increased or decreased to ensure high-quality scanning in various environments.

[0128] The above formulas are all calculated by removing the dimension and taking their numerical values. The formulas are obtained by collecting a large amount of data for software simulation to get a formula closest to the actual situation. The preset parameters and threshold selection in the formulas are set by those skilled in the art according to the actual situation. The parts not described in the present invention are applicable to the prior art.

Claims

1. A sample barcode scanning system for a fully automatic coagulation tester, characterized in that, It includes a reflection characteristic analysis module, a thermal imaging data analysis module, a scanning mode control module, a spatial independence evaluation module, and a scanning reliability evaluation module: The reflection characteristic analysis module analyzes the optical reflection characteristics of the sample barcode to determine whether there is an abnormal optical refraction caused by reagent vapor on the surface of the sample label; The thermal imaging data analysis module analyzes the thermal imaging sensor data on the sample transport track to determine whether multiple sample tubes enter the scanning area simultaneously; Based on the judgment results of whether there is an abnormal optical refraction caused by reagent vapor on the surface of the sample label and whether multiple sample tubes enter the scanning area simultaneously in the scanning area, the scanning mode control module determines the sample barcode scanning mode; When the sample barcode scanning mode is the multi-tube simultaneous scanning mode, the spatial independence evaluation module evaluates the spatial independence of the sample barcode by analyzing the spatial layout of the sample barcodes in the scanning area and using the clustering analysis algorithm; Based on the spatial independence of the sample barcode, the scanning reliability evaluation module evaluates the overall scanning reliability of the sample barcode in the current scanning environment and determines whether it meets the scanning quality requirements.

2. The sample barcode scanning system of a fully automatic coagulation tester according to claim 1, wherein Analyze the optical reflection characteristics of the sample barcode to determine whether there is an abnormal optical refraction caused by reagent vapor on the surface of the sample label. Specifically: Collect the reflected optical image data on the surface of the sample barcode; Perform preprocessing of noise filtering and gray level equalization on the collected image; Use the spectrum analysis algorithm to extract the reflected optical characteristic parameters; Compare the extracted reflected optical characteristic parameters with the standard optical reflection model parameters to determine whether there is a refraction abnormality.

3. The sample barcode scanning system of a fully automatic coagulation tester according to claim 2, wherein Analyze the thermal imaging sensor data on the sample transport track to determine whether multiple sample tubes enter the scanning area simultaneously. Specifically: Collect the thermal image data output by the thermal imaging sensor on the sample transport track; Perform temperature range constraint on the thermal image data to filter out background invalid thermal signals; Perform temperature gradient enhancement on the filtered thermal image data to strengthen the thermal characteristics of the sample tube; Use the region growing algorithm to separate the candidate sample tube regions; Calculate the center coordinates and boundary dimensions of the candidate sample tube regions; Based on multi-target tracking, determine whether multiple sample tubes enter the scanning area simultaneously.

4. The sample barcode scanning system of a fully automatic coagulation tester according to claim 3, characterized in that Based on the judgment results of whether there is an abnormal optical refraction caused by reagent vapor on the surface of the sample label and whether multiple sample tubes enter the scanning area simultaneously in the scanning area, determine the sample barcode scanning mode. Specifically: Conduct a combined analysis of the optical refraction abnormality judgment result and the sample tube quantity judgment result; Construct a scanning mode decision matrix and set the scanning mode corresponding to each judgment result combination: when the optical state is normal and only a single sample tube enters the scanning area, it is set as the normal single-tube scanning mode; when the optical state is abnormal and only a single sample tube enters the scanning area, it is set as the optical abnormal single-tube mode; when the number of sample tubes is multiple, it is all set as the multi-tube simultaneous scanning mode.

5. The sample barcode scanning system of a fully automatic coagulation tester according to claim 4, characterized in that, When the sample barcode scanning mode is the multi-tube simultaneous scanning mode, evaluate the spatial independence of the sample barcode by analyzing the spatial layout of the sample barcodes in the scanning area and using the clustering analysis algorithm. Specifically: Extract the center coordinates of each sample barcode in the scanning area; Construct a set of spatial coordinate points of sample barcodes based on the coordinate information of each sample barcode; Perform density clustering analysis on the set of spatial coordinate points of sample barcodes to identify the clustered areas of sample barcodes; Calculate the average minimum distance between barcodes within each clustered area of sample barcodes; Judge whether the average minimum distance meets the set threshold, and output the evaluation result of spatial independence.

6. The sample barcode scanning system of a fully automatic coagulation tester according to claim 5, characterized in that, Based on the spatial independence of sample barcodes, evaluate the overall scanning reliability of sample barcodes in the current scanning environment, and judge whether it meets the scanning quality requirements. Specifically: Receive the average minimum distance and the evaluation result of spatial independence of sample barcodes within each clustering cluster; Evaluate the spatial distribution density of sample barcodes in the scanning area based on the total number of clustering clusters; Analyze the deviation degree of the average minimum distance between sample barcodes within each clustering cluster from the minimum safety distance threshold; Combine the spatial distribution density and the deviation degree of the average minimum distance to calculate the environmental complexity index; Conduct correlation analysis between the environmental complexity index and the barcode spatial independence data; Based on the correlation analysis result, output the overall scanning reliability of sample barcodes in the current scanning environment, and judge whether it meets the scanning quality requirements.

7. The sample barcode scanning system of a fully automatic coagulation tester according to claim 6, characterized in that, Based on the correlation analysis result, output the overall scanning reliability of sample barcodes in the current scanning environment, and judge whether it meets the scanning quality requirements. Specifically: Preset an overall scanning reliability threshold, and compare the overall scanning reliability index with the overall scanning reliability threshold: When the overall scanning reliability index is greater than or equal to the overall scanning reliability threshold, it indicates that the overall scanning reliability of sample barcodes in the current scanning environment is high and meets the scanning quality requirements; When the overall scanning reliability index is less than the overall scanning reliability threshold, it indicates that the overall scanning reliability of sample barcodes in the current scanning environment is low and does not meet the scanning quality requirements.

Images