Visual inspection method, system and device for disinfection package of medical instrument

Abstract

The invention discloses a visual detection method, system and device for disinfection packaging of medical instruments, and relates to the technical field of machine visual process.The method includes the steps that an original image is corrected to a standard coordinate system to eliminate perspective distortion, a reference template is used for determining an effective packaging detection area, and multi-dimensional channel expansion is conducted; and comparing with a reference feature to obtain surface optical abnormal data. Deviation is calculated according to a preset sealing line position, a sealing line consistency index is obtained through nonlinear aggregation, a comprehensive judgment index is obtained through calculation, a package qualification or disqualification signal is output after the comprehensive judgment index is compared with a threshold value, and automatic detection of the appearance and the sealing performance of the medical instrument disinfection package is achieved. An effective detection area is obtained by correcting an image and matching a reference template, and a judgment signal is output by calculating a comprehensive judgment index and comparing the comprehensive judgment index with a preset threshold value, so that the appearance and sealing detection precision, robustness and detection efficiency of the disinfection package of the medical instrument are comprehensively improved on the premise of multiple dimensions and standardization.

Description

Technical Field

[0001] The present invention relates to the technical field of machine vision processing, and particularly to a visual inspection method, system and device for medical device disinfection packaging. Background Art

[0002] Before leaving the factory, medical devices often need to undergo strict disinfection and aseptic packaging processes to reduce the risk of infection during clinical use. Therefore, the integrity, sealing performance, cleanliness of medical device packaging and the accuracy of identification information play important roles in quality control. Traditional quality inspection methods usually rely on manual visual inspection or simple machine vision methods, and there are many bottlenecks in the face of high-speed and high-volume production lines.

[0003] On the one hand, manual inspection is easily affected by factors such as the experience of operators, fatigue level and subjective judgment, and it is difficult to ensure the consistency and reliability of inspection results; on the other hand, simple machine vision inspection means are often limited to the basic analysis of two-dimensional grayscale or color images, and are insufficient in multi-dimensional optical feature, micro-difference and complex perspective distortion correction capabilities, and cannot comprehensively quantify and accurately determine the subtle abnormalities (such as micro-scratches, stains) on the packaging surface, local breaks or deviations of the sealing line, and structural abnormalities.

[0004] In addition, in the prior art, although some devices already have basic image correction and region recognition capabilities, the analysis of the packaging surface and the state of the sealing line still relies too much on simple statistical quantities and linear measurement methods. In the face of complex materials, different batches and diverse packaging types, simple linear determination is often prone to missed detection or misjudgment, and cannot meet the strict quality standards of medical devices. Summary of the Invention

[0005] Based on the above-mentioned disadvantages of the prior art, the purpose of the present invention is to provide a visual inspection method, system and device for medical device disinfection packaging to solve the above technical problems.

[0006] To achieve the above purpose, the present invention provides the following technical solution: A visual inspection method for medical device disinfection packaging, including:

[0007] S1: Correct the acquired original packaging image, map the image to a standard coordinate system to eliminate the shooting angle and perspective distortion, and use a predetermined reference template to perform region recognition and mask extraction on the corrected image to determine the effective detection region of the packaging;

[0008] S2: Within the determined effective detection region, perform multi-dimensional channel expansion on the optical characteristics of the image, and perform difference measurement between the corrected image and the reference features of the predetermined reference template to obtain data describing the optical abnormal features of the packaging surface;

[0009] S3: According to the defined position of the sealing line, calculate the deviation between the pixel points in the same effective detection area and the sealing line reference feature, and obtain the sealing line consistency index through weighted and non-linear deviation aggregation to reflect the deviation of the sealing line in terms of structure and strength distribution;

[0010] S4: Calculate the comprehensive judgment index based on the surface optical anomaly feature data and the sealing line consistency index for comprehensively evaluating the appearance and sealing state of the package. Compare the comprehensive judgment index with a predetermined threshold. When the index exceeds the threshold, output a non-conforming signal; otherwise, output a conforming signal to achieve automatic visual detection and judgment of the appearance and sealing performance of the medical device disinfection package.

[0011] The present invention is further configured such that step S1 includes:

[0012] Define the mapping relationship from the original coordinate system (x, y) to the standard coordinate system (u, v) to align the corrected image with the standard reference coordinates;

[0013] Based on the defined mapping relationship, map the gray value of each pixel (x, y) in the original image to the corresponding position of the corrected coordinates (u, v) to generate a corrected image;

[0014] Perform feature separation processing on the corrected image and the predetermined reference template. Extract the differential features through the feature separation function and output the corresponding mask;

[0015] Perform conditional judgment on the mask. Mark the pixels that meet the conditions as valid pixels, and define the area composed of all valid pixels as the effective packaging area.

[0016] The present invention is further configured such that the feature separation function is: M(u, v) = exp|(X ′ (u, v) - R(u, v))|, where M(u, v) is the mask of the coordinate (u, v), X ′ (u, v) is the gray value of the pixel (u, v) in the corrected image, and R(u, v) is the gray value of the pixel (u, v) in the predetermined reference template;

[0017] Effective packaging area: D(u, v) = f(M(u, v) ≤ η), where D(u, v) is the effective packaging area, f(·) is the conditional judgment function, which outputs 1 when the condition is satisfied and 0 otherwise, and η is the condition threshold.

[0018] The present invention is further configured such that step S2 includes:

[0019] Non-linearly map the single-channel images of the corrected image and the predetermined reference template image to multi-dimensional channels;

[0020] By calculating the difference values of the multi-dimensional channels between the calibrated image and a predetermined reference template, a difference metric is generated through product operations between channels.

[0021] The present invention is further configured such that the calculation logic for non-linearly mapping the single-channel images of the calibrated image and the predetermined reference template image to multi-dimensional channels is: Z(u, v, ρ) = (X ′ (u, v) + Δ(ρ)) τ(ρ) , Y(u, v, ρ) = (R(u, v) + Δ(ρ)) τ(ρ) , where Z(u, v, ρ) is the feature representation of the calibrated image non-linearly mapped to the ρ channel, Δ(ρ) is the channel offset parameter for introducing displacement in the channel, τ(ρ) is the channel non-linear exponent, and Y(u, v, ρ) is the feature representation of the predetermined reference template image non-linearly mapped to the ρ channel;

[0022] The calculation logic for the difference metric is: where H(u, v) is the difference metric, Ω is the total number of channels, and ξ is the correction parameter.

[0023] The present invention is further configured such that step S3 includes:

[0024] For each pixel (u, v) at the predetermined seal line position, calculate the intensity deviation Δ(u, v) between the pixel and the seal line reference feature;

[0025] Obtain the seal line consistency index by aggregating the weighted and non-linear deviations of the predetermined seal line based on the intensity deviation Δ(u, v).

[0026] The present invention is further configured such that the calculation logic for the intensity deviation Δ(u, v) is: Δ(u, v) = X ′ (u, v) - W(u, v), where W(u, v) is the gray value of the seal line reference feature;

[0027] The calculation logic for the seal line consistency index is: S is the seal line consistency index, (u, v) ∈ C indicates that the pixel (u, v) belongs to the seal line set C, Π is the weighting function, θ and are non-integer positive real power exponents for non-linearly amplifying the deviation, σ is the channel amplification parameter for non-linearly suppressing the deviation in the denominator term, and κ is the correction parameter.

[0028] The present invention is further configured such that step S4 includes:

[0029] Perform regional aggregation on the difference metric to generate the overall surface anomaly metric parameter, SH = ∑ u,v H(u, v), where SH is the overall surface anomaly metric parameter;

[0030] The comprehensive judgment index is calculated according to the overall surface anomaly measurement parameter and the sealing line consistency index, Q = α·SH + β·S, where α and β are weight coefficients.

[0031] The present invention also provides a visual inspection system for medical device disinfection packaging, and the system includes:

[0032] A calibration module: calibrates the acquired original packaging image, maps the image to a standard coordinate system to eliminate the shooting angle and perspective distortion, and performs region recognition and mask extraction on the calibrated image using a predetermined reference template to determine the effective detection area of the packaging;

[0033] A measurement module: within the determined effective detection area, performs multi-dimensional channel expansion on the optical characteristics of the image, and measures the difference between the calibrated image and the reference features of the predetermined reference template to obtain data describing the optical anomaly characteristics of the packaging surface;

[0034] A calculation module: according to the defined position of the sealing line, calculates the deviation of the pixel points in the same effective detection area from the reference features of the sealing line, and obtains the sealing line consistency index through weighted and non-linear deviation aggregation, which is used to reflect the deviation of the sealing line in terms of structure and strength distribution;

[0035] A judgment module: calculates the comprehensive judgment index according to the surface optical anomaly characteristic data and the sealing line consistency index, which is used to comprehensively evaluate the appearance and sealing state of the packaging, compares the comprehensive judgment index with a predetermined threshold, and outputs a non-conforming signal when the index exceeds the threshold, otherwise outputs a conforming signal, so as to realize the automatic visual inspection and judgment of the appearance and sealing performance of medical device disinfection packaging.

[0036] The present invention also provides a visual inspection device for medical device disinfection packaging, and the device includes:

[0037] One or more processors;

[0038] A storage medium for storing one or more programs, and when the one or more programs are executed by the one or more processors, the device realizes a visual inspection method for medical device disinfection packaging as described in any one of the above.

[0039] The present invention provides a visual inspection method, system and device for the disinfection packaging of medical devices. The method corrects the acquired original packaging image, maps the image to a standard coordinate system to eliminate the shooting angle and perspective distortion, and uses a predetermined reference template to perform region recognition and mask extraction on the corrected image to determine the effective detection region of the packaging; within the determined effective detection region, the optical characteristics of the image are extended in multiple dimensions, and the corrected image is compared with the reference features of the predetermined reference template for difference measurement to obtain data describing the optical anomaly characteristics on the packaging surface; according to the defined position of the sealing line, the pixel points in the same effective detection region are calculated for deviation from the reference features of the sealing line, and a sealing line consistency index is obtained through weighted and non-linear deviation aggregation to reflect the deviation of the sealing line in terms of structure and strength distribution; a comprehensive judgment index is calculated based on the surface optical anomaly characteristic data and the sealing line consistency index for comprehensively evaluating the appearance and sealing state of the packaging. The comprehensive judgment index is compared with a predetermined threshold value. When the index exceeds the threshold value, an unqualified signal is output, otherwise a qualified signal is output, realizing the automatic visual inspection and judgment of the appearance and sealing performance of the medical device disinfection packaging. The beneficial effects generated include:

[0040] 1. Improve detection accuracy and robustness: By defining the mapping relationship from the original image to the standard coordinate system and correcting the image, the interference of the shooting angle and perspective distortion on the detection result can be eliminated; using a predetermined reference template to perform region recognition and mask extraction on the corrected image ensures that subsequent optical feature analysis and sealing performance detection focus on the interior of the effective region, reducing misjudgment caused by background stray information;

[0041] 2. Achieve multi-dimensional optical feature enhancement and precise difference measurement: Extend the optical characteristics of the image in multiple dimensions within the determined effective detection region, breaking through the limitations of traditional two-dimensional grayscale or simple color features. Through non-linear mapping and offset adjustment between channels, a more expressive feature space is obtained. The corrected image is subjected to difference measurement with the reference features, making it easier to amplify and identify surface optical anomalies in multiple dimensions, effectively improving the detection sensitivity and reliability for weak defects;

[0042] 3. Fine-grained sealing line analysis and structural consistency measurement: According to the defined position of the sealing line, the pixel points in the effective detection region are calculated for deviation from the reference features of the sealing line, and a sealing line consistency index is obtained through weighted and non-linear deviation aggregation. This index can quantify the local anomalies of the sealing line at the structural and brightness distribution levels, making potential defects such as poor sealing and line offset more intuitively obvious, ensuring the sealing quality and sterility of the packaging;

[0043] 4. The comprehensive judgment index improves the overall detection efficiency: The data of surface optical anomaly features and the sealing line consistency index are non-linearly fused to form a comprehensive judgment index. This index realizes a higher-level comprehensive evaluation based on the coupling of multi-source information. It can not only judge the appearance quality and sealing state of the package simultaneously, but also maintain a high discrimination ability under different types and degrees of abnormal conditions. By comparing the comprehensive judgment index with a predetermined threshold, a qualified or unqualified signal is output, thus achieving a fast, automated, and standardized quality control effect.

[0044] The above description is only an overview of the technical solution of this application. In order to be able to understand the technical means of this application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features, and advantages of this application more obvious and understandable, the following specifically illustrates the specific implementation manners of this application. Brief Description of the Drawings

[0045] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings. In the drawings:

[0046] Figure 1 It is a flowchart of a visual inspection method for a medical device disinfection package shown in an exemplary embodiment of the present invention;

[0047] Figure 2 It is a schematic structural diagram of a visual inspection system for a medical device disinfection package shown in an exemplary embodiment of the present invention. Detailed Description of the Preferred Embodiments

[0048] The following will illustrate the embodiments of the present invention with reference to the drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific implementation manners. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention, rather than for limiting the protection scope of the present invention.

[0049] It should be noted that the drawings provided in the following embodiments only illustrate the basic concept of the present invention in a schematic manner. Therefore, only the components related to the present invention are shown in the drawings, rather than being drawn according to the number, shape, and size of the components in actual implementation. The type, quantity, and ratio of each component in actual implementation can be arbitrarily changed, and the component layout type may also be more complex.

[0050] In the following description, a large number of details are explored to provide a more thorough explanation of the embodiments of the present invention. However, it is obvious to those skilled in the art that the embodiments of the present invention can be implemented without these specific details. In other embodiments, well-known structures and devices are shown in the form of block diagrams rather than in detail to avoid making the embodiments of the present invention difficult to understand.

[0051] Embodiment 1

[0052] A visual inspection method for the disinfection packaging of medical devices, as Figure 1 shown, includes:

[0053] S1: Correct the acquired original packaging image, map the image to a standard coordinate system to eliminate the shooting angle and perspective distortion, and use a predetermined reference template to perform region recognition and mask extraction on the corrected image to determine the effective detection area of the packaging;

[0054] S2: Within the determined effective detection area, perform multi-dimensional channel expansion on the optical characteristics of the image, and perform difference measurement between the corrected image and the reference features of the predetermined reference template to obtain data describing the optical anomaly features on the packaging surface;

[0055] S3: According to the defined position of the sealing line, calculate the deviation of the pixel points in the same effective detection area from the reference features of the sealing line, and obtain the sealing line consistency index through weighted and non-linear deviation aggregation to reflect the deviation of the sealing line in terms of structure and strength distribution;

[0056] S4: Calculate a comprehensive judgment index based on the surface optical anomaly feature data and the sealing line consistency index for comprehensively evaluating the appearance and sealing state of the packaging. Compare the comprehensive judgment index with a predetermined threshold. When the index exceeds the threshold, output a non-conforming signal; otherwise, output a conforming signal to achieve automatic visual detection and judgment of the appearance and sealing performance of the medical device disinfection packaging.

[0057] Specifically, step S1 aims to provide a stable and standardized coordinate system and effective area basis for subsequent feature analysis and judgment. By mapping the originally acquired packaging image to a standard reference coordinate system and extracting the effective detection area according to the reference template, it can ensure that the subsequent analysis of optical features and sealing line consistency is carried out under a standardized, repeatable, and highly confident benchmark. The present invention is further configured such that step S1 includes:

[0058] Define the mapping relationship from the original coordinate system (x, y) to the standard coordinate system (u, v) to align the corrected image with the standard reference coordinates. Specifically, the pixel distribution of the original image is affected by the pose during shooting (including camera tilt, distance change, perspective distortion, etc.). Through camera calibration means (usually using calibration boards, known-size markers, etc. in actual operations), mapping parameters can be obtained, enabling the mapping of the original image coordinates (x, y) to the standard reference coordinates (u, v). The standard coordinate system is an idealized, distortion-free planar coordinate system. In this coordinate system, the boundaries, markings, and feature positions of the package are strictly aligned with the predetermined template, so that regardless of how the original shooting perspective changes, after mapping and correction, a consistent image reference relationship can be obtained in the standard coordinate system;

[0059] Based on the defined mapping relationship, map the gray value of each pixel (x, y) in the original image to the corresponding position of the corrected coordinates (u, v) to generate a corrected image. Specifically, after having the coordinate mapping relationship, interpolate and map the gray value at the position (x, y) in the original image to the position (u, v). Each pixel of the corrected image generated through this process can be considered to directly correspond to the reference template in the same standard coordinate framework;

[0060] Perform feature separation processing on the corrected image and the predetermined reference template, and extract the differential features through the feature separation function and output the corresponding mask. The present invention is further set such that the feature separation function is: M(u, v) = exp|(X ′ (u, v) - R(u, v))|, where M(u, v) is the mask at the coordinates (u, v), X ′ (u, v) is the gray value of the pixel (u, v) in the corrected image, and R(u, v) is the gray value of the pixel (u, v) in the predetermined reference template; the feature separation function utilizes the monotonically increasing and non-linear characteristics of the exponential function to map the brightness difference between the corrected image and the reference template, so that the larger the difference value, the larger the corresponding M(u, v), thereby achieving the amplification and differentiation of the difference;

[0061] Perform condition judgment on the mask. Those meeting the conditions are marked as valid pixels, and the area composed of all valid pixels is defined as the valid packaging area. Valid packaging area: D(u, v) = f(M(u, v) ≤ η), where D(u, v) is the valid packaging area, f(·) is the condition judgment function, which outputs 1 when the condition holds and 0 otherwise, and η is the condition threshold. Specifically, the mask M(u, v) has been calculated in the previous steps to describe the feature differences between the corrected image and the reference template. To convert this continuous or non-linear metric into a practically usable valid area indication, a condition judgment needs to be imposed on M(u, v). By setting a condition threshold η, it is judged whether M(u, v) meets the specific condition, M(u, v) ≤ η, so as to determine the pixels meeting the conditions as "valid pixels". The set of all pixels meeting the conditions is defined as the valid packaging area D(u, v). Through threshold screening, only the pixels close to the reference features are retained as the valid area. This can significantly reduce the interference of background pixels on subsequent detection and analysis, making the subsequent evaluation of optical anomaly features and seal line status more accurate and efficient.

[0062] Specifically, in step S2, the image is lifted from the single-channel feature to the multi-channel feature space, and the difference between the corrected image and the reference template is measured in this high-dimensional space. The core idea of this logic is that under the condition of a single channel (such as grayscale value), the ability to distinguish tiny anomaly features is relatively limited and is easily affected by changes in illumination, color, or material. When the single-channel image is extended to multi-dimensional channels through non-linear mapping, each channel can highlight certain specific optical characteristics or brightness distribution features. Thus, when measuring the difference between the corrected image and the reference template in the multi-channel space, surface anomalies and structural deviations can be captured more comprehensively and meticulously, thereby improving the overall detection accuracy. The present invention is further configured such that step S2 includes:

[0063] Non-linearly map the single-channel images of the corrected image and the predetermined reference template image to multi-dimensional channels; the present invention is further configured such that the calculation logic for non-linearly mapping the single-channel images of the corrected image and the predetermined reference template image to multi-dimensional channels is: Z(u, v, ρ) = (X ′ (u, v) + Δ(ρ)) τ(ρ) , Y(u, v, ρ) = (R(u, v) + Δ(ρ)) τ(ρ), where \(Z(u, v, \rho)\) is the feature representation of the corrected image non-linearly mapped to the \(\rho\) channel, \(\Delta(\rho)\) is the channel offset parameter used to introduce a displacement in the channel, \(\tau(\rho)\) is the channel non-linear exponent, and \(Y(u, v, \rho)\) is the feature representation of the predetermined reference template image non-linearly mapped to the \(\rho\) channel; specifically, the original single-channel grayscale image is extended to a multi-dimensional channel feature representation by introducing channel offset and non-linear power mapping. In this way, each pixel point \((u, v)\) is no longer described by a single grayscale value after mapping, but is represented by a set of feature values \(Z(u, v, 1), Z(u, v, 2), \ldots, Z(u, v, \rho)\) and \(Y(u, v, 1), Y(u, v, 2), \ldots, Y(u, v, \rho)\). Each channel performs a specific transformation on the original brightness through the offset parameter \(\Delta(\rho)\) and the non-linear exponent \(\tau(\rho)\), enhancing the sensitivity to different brightness intervals and specific optical features. Through this multi-dimensional non-linear mapping, small abnormal features can be more effectively highlighted in the subsequent difference measurement between the corrected image and the reference template; the value of the offset parameter \(\Delta(\rho)\) is determined according to the brightness distribution of the target image and specific detection targets, and the value range is \([-0.5, 0.5]\); the non-linear exponent \(\tau(\rho)\) is a non-integer positive real number, and the value range is \((0.5, 2)\); the single-channel grayscale value cannot be sensitive to different brightness regions simultaneously. Through the offset and non-linear exponent, multiple channels can respectively emphasize the subtle changes in different brightness intervals, effectively improving the detection ability for surface micro-abnormalities;

[0064] By calculating the difference values of the multi-dimensional channels of the corrected image and the predetermined reference template, fusion is performed through the product operation between channels to generate a difference metric. The calculation logic of the difference metric is: where \(H(u, v)\) is the difference metric, \(\Omega\) is the total number of channels, and \(\xi\) is a correction parameter. Specifically, after performing multi-dimensional channel feature mapping on the corrected image and the reference template image, the difference values between the two are calculated for each channel. To avoid the linearization problem caused by simply adding the difference values, this logic selects to fuse the differences through the product operation between channels, and finally obtains the difference degree through non-linear correction and normalization processing. The multi-channel product fusion combined with non-linear correction enables small differences to be amplified in specific channels. When there is an abnormality, at least one channel will generate a large difference value, which will then spread to the overall difference metric through the product, improving the detectability of the abnormality. If some channels are affected by noise or local illumination changes, the stable performance of the remaining channels can still enhance the comprehensive effect through the product, preventing the system from being greatly disturbed by a single-point abnormality and improving the stability and repeatability of the detection.

[0065] The present invention is further set such that step S3 includes:

[0066] For each pixel (u, v) at a predetermined seal line position, calculate the intensity deviation Δ(u, v) between the pixel and the seal line reference feature; the present invention is further configured such that the calculation logic of the intensity deviation Δ(u, v) is: Δ(u, v) = X ′ (u, v) - W(u, v), where W(u, v) is the gray value of the seal line reference feature; specifically, in the detection of medical device disinfection packaging, whether the seal line is complete, uniform, and abnormal is an important quality determination criterion. For this reason, it is necessary to quantify and analyze the gray scale features of each pixel at the predetermined seal line position in the image, and compare the pixel value X ′ (u, v) corresponding to the seal line position in the corrected image with the standard gray value W(u, v) of the seal line reference feature, and describe the deviation degree of the pixel point from the ideal seal line state in actual production by calculating the intensity deviation Δ(u, v). This deviation value is the basis for subsequent non-linear weighted aggregation and consistency index calculation;

[0067] Obtain a seal line consistency index by weighted and non-linear deviation aggregation for the predetermined seal line according to the intensity deviation Δ(u, v). The calculation logic of the seal line consistency index is: S is the seal line consistency index, (u, v) ∈ C indicates that the pixel (u, v) belongs to the seal line set C, Π is the weighting function, θ and are non-integer positive real power exponents for non-linearly amplifying the deviation, σ is the channel amplification parameter for non-linearly suppressing the deviation in the denominator term, and κ is the correction parameter. Specifically, for each pixel (u, v) in the seal line pixel set C in the above calculation logic, the corresponding intensity deviation Δ(u, v) has been obtained in the previous step. Using this deviation, a seal line consistency index S is constructed through weighted and non-linear deviation aggregation to quantify the overall consistency degree of the seal line in terms of structure and brightness distribution; the non-integer positive real power exponents θ and are determined according to the sensitivity requirement for the deviation size, and the value range is (0.5, 2), and the channel amplification parameter σ is a positive number for adjusting the amplification effect of the denominator term;

[0068] The present invention is further configured such that step S4 includes:

[0069] Perform regional aggregation on the difference metric to generate an overall surface anomaly metric parameter, SH = ∑ u,v H(u, v), where SH is the overall surface anomaly metric parameter; specifically, the surface optical anomaly metric matrix H(u, v) and the seal line consistency index S have been obtained in the previous step. Step S4 sums H(u, v) within the effective region to obtain the overall surface anomaly metric parameter SH, which describes the aggregation of the overall anomaly degree of the packaging surface;

[0070] The comprehensive judgment index is calculated based on the overall surface anomaly measurement parameter and the seal line consistency index, Q = α·SH + β·S, where α and β are weight coefficients. Specifically, compared with only looking at a single index, by fusing SH and S, the comprehensive judgment index Qs incorporates both surface anomalies and seal line quality into the decision-making process, making the final judgment more comprehensive. It neither ignores small but widely distributed surface problems nor misses the requirements for the sensitivity to local anomalies of the seal line; it realizes the fusion and simplification of multi-dimensional characteristics of packaging quality, provides a highly adaptable and easily adjustable judgment mechanism, and helps to efficiently and accurately screen qualified and unqualified medical device packages on the industrial production line.

[0071] Embodiment 2

[0072] Please refer to Figure 2 , the exemplary visual inspection system for medical device disinfection packaging includes:

[0073] Calibration module: Calibrate the acquired original packaging image, map the image to the standard coordinate system to eliminate the shooting angle and perspective distortion, and use a predetermined reference template to perform region recognition and mask extraction on the calibrated image to determine the effective detection area of the packaging;

[0074] Measurement module: In the determined effective detection area, perform multi-dimensional channel expansion on the optical characteristics of the image, and measure the difference between the calibrated image and the reference features of the predetermined reference template to obtain data describing the optical anomaly characteristics of the packaging surface;

[0075] Calculation module: According to the defined position of the seal line, calculate the deviation between the pixel points in the same effective detection area and the reference features of the seal line, and obtain the seal line consistency index through weighted and non-linear deviation aggregation to reflect the deviation of the seal line in terms of structure and strength distribution;

[0076] Judgment module: Calculate the comprehensive judgment index based on the surface optical anomaly characteristic data and the seal line consistency index, which is used to comprehensively evaluate the appearance and seal state of the packaging. Compare the comprehensive judgment index with a predetermined threshold. When the index exceeds the threshold, an unqualified signal is output, otherwise a qualified signal is output, realizing the automatic visual inspection and judgment of the appearance and sealing performance of medical device disinfection packaging.

[0077] It should be noted that a visual inspection system for medical device disinfection packaging provided in the above embodiment and a visual inspection method for medical device disinfection packaging provided in the above embodiment belong to the same concept. The specific ways in which each module and unit perform operations have been described in detail in the method embodiment, and will not be repeated here. In practical applications, the visual inspection system for medical device disinfection packaging provided in the above embodiment can allocate the above functions to different functional modules as needed, that is, divide the internal structure of the system into different functional modules to complete all or part of the functions described above, and this will not be limited here either.

[0078] An embodiment of the present application further provides a visual inspection device for medical device disinfection packaging, including: one or more processors; a storage medium for storing one or more programs, and when the one or more programs are executed by the one or more processors, the device implements a visual inspection method for medical device disinfection packaging provided in each of the above embodiments.

[0079] The above embodiments can be implemented in whole or in part by software, hardware, 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 processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. 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 a website, computer, server, or data center to another website, computer, server, or data center by wired (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that the computer can access or a data storage device such as a server or data center that includes a collection of one or more available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0080] It should be understood that the term "and / or" in this text is merely a description of the association relationship between associated objects, indicating that there can be three relationships. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone. Here, A and B can be singular or plural. Additionally, the character " / " in this text generally represents an "or" relationship between the associated objects before and after, but it may also represent an "and / or" relationship. The specific meaning can be understood by referring to the context before and after.

[0081] In this application, "at least one" means one or more, and "a plurality" means two or more. "At least one of the following items (or elements)" or its similar expressions refer to any combination of these items, including any combination of single items (or elements) or plural items (or elements). For example, at least one of a, b, or c can represent: a, b, c, a - b, a - c, b - c, or a - b - c, where a, b, and c can be single or multiple.

[0082] It should be understood that in various embodiments of this application, the magnitudes of the serial numbers of the above - mentioned processes do not imply the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of this application.

[0083] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or by a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of this application.

[0084] 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 foregoing method embodiments, and will not be elaborated herein.

[0085] In several embodiments provided in this application, it should be understood that the disclosed system can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there can be other division methods. For example, 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 displayed or discussed coupling or direct coupling or communication connection between each other can be through some interfaces. The indirect coupling or communication connection of devices or units can be in an electrical, mechanical, or other form.

[0086] The unit described as a separation component may or may not be physically separated. The component shown as a unit may or may not be a physical unit, that is, it may be located in one place or distributed over multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0087] In addition, each functional unit in various embodiments of the present application may be integrated into a processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit.

[0088] If the above-mentioned function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art, or part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to enable a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0089] The above is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered by the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A visual inspection method for sterilized packaging of medical devices, characterized in that: include: S1: Correct the acquired original packaging image and map the image to a standard coordinate system to eliminate the shooting angle and perspective distortion, and use a predetermined reference template to perform region recognition and mask extraction on the corrected image to determine the effective inspection area of ​​the packaging; S2: within the determined effective detection area, perform multi-dimensional channel expansion on the optical characteristics of the image, and measure the difference between the corrected image and the reference characteristics of the predetermined reference template to obtain data describing the optical abnormality characteristics of the packaging surface; S3: According to the predetermined seal line position definition, the deviation between the pixel points in the same effective detection area and the seal line reference feature is calculated, and the seal line consistency index is obtained by weighted and nonlinear deviation aggregation to reflect the deviation of the seal line in structure and intensity distribution; S4: A comprehensive judgment index is calculated based on the surface optical abnormality feature data and the sealing line consistency index, which is used to comprehensively evaluate the appearance and sealing status of the package. The comprehensive judgment index is compared with a predetermined threshold value. When the index exceeds the threshold value, an unqualified signal is output, otherwise a qualified signal is output, thereby realizing automated visual inspection and judgment of the appearance and sealing of the sterilized packaging of medical devices.

2. A visual inspection method for sterilization packaging of medical devices according to claim 1, characterized in that: Step S1 includes: Define the mapping relationship from the original coordinate system (x, y) to the standard coordinate system (u, v) so that the rectified image is aligned with the standard reference coordinates; Based on the defined mapping relationship, the grayscale value of each pixel (x, y) in the original image is mapped to the corresponding position of the corrected coordinates (u, v) to generate a corrected image; Perform feature separation processing on the corrected image and the predetermined reference template, extract the difference features through the feature separation function and output the corresponding mask; The mask is subjected to conditional judgment, pixels meeting the condition are marked as valid pixels, and an area consisting of all valid pixels is defined as a valid packaging area.

3. A visual inspection method for sterilization packaging of medical devices according to claim 2, characterized in that: The feature separation function is: M(u,v)=exp|(X ′ (u,v)-R(u,v))|, where M(u,v) is the mask of coordinate (u,v), X ′ (u, v) is the gray value of the corrected image pixel (u, v), and R(u, v) is the gray value of the predetermined reference template pixel (u, v); Valid packaging area: D(u,v)=f(M(u,v)≤η), where D(u,v) is the valid packaging area, f(·) is the conditional judgment function, which outputs 1 when the condition is met, otherwise it outputs 0, and η is the conditional threshold.

4. A visual inspection method for sterilization packaging of medical devices according to claim 3, characterized in that: Step S2 includes: Nonlinearly mapping the single-channel images of the correction image and the predetermined reference template image to multi-dimensional channels; The difference value of the multi-dimensional channel between the corrected image and the predetermined reference template is calculated, and the difference metric is generated by fusing them through the product operation between the channels.

5. A visual inspection method for sterilization packaging of medical devices according to claim 4, characterized in that: The calculation logic of nonlinearly mapping the single-channel image of the correction image and the predetermined reference template image to the multi-dimensional channel is: Z(u,v,ρ)=(X ′ (u,v)+Δ(ρ)) τ(ρ) , Y(u,v,ρ)=(R(u,v)+Δ(ρ)) τ(ρ) , where Z(u,v,ρ) is the feature representation of the nonlinear mapping of the corrected image to the ρ channel, Δ(ρ) is the channel offset parameter used to introduce displacement in the channel, τ(ρ) is the channel nonlinear index, and Y(u,v,ρ) is the feature representation of the nonlinear mapping of the predetermined reference template image to the ρ channel; The calculation logic of the difference metric is: Among them, H(u,v) is the difference metric, Ω is the total number of channels, and ξ is the correction parameter.

6. A visual inspection method for sterilization packaging of medical devices according to claim 3, characterized in that: Step S3 includes: For each pixel (u, v) at the predetermined sealing line position, the intensity deviation Δ(u, v) between the pixel and the sealing line reference feature is calculated; According to the intensity deviation Δ(u,v), the sealing line consistency index is obtained by weighting and nonlinear deviation aggregation for the predetermined sealing line.

7. A visual inspection method for sterilization packaging of medical devices according to claim 6, characterized in that: The calculation logic of intensity deviation Δ(u,v) is: Δ(u,v)=X ′ (u,v)-W(u,v), where W(u,v) is the grayscale value of the seal line reference feature; The calculation logic of the sealing line consistency index is: S is the sealing line consistency index, (u,v)∈C means that the pixel (u,v) belongs to the sealing line set C, Π is the weighting function, θ and is a non-integer positive real power exponent, used to perform nonlinear amplification on the deviation, σ is the channel amplification parameter, used to perform nonlinear suppression on the deviation in the denominator, and κ is the correction parameter.

8. A visual inspection method for sterilization packaging of medical devices according to claim 7, characterized in that: Step S4 includes: The difference metric is regionally aggregated to generate the overall surface anomaly metric parameter, SH = ∑ u,v H(u,v), where SH is the overall surface anomaly measurement parameter; The comprehensive judgment index is calculated based on the overall surface anomaly measurement parameters and the sealing line consistency index, Q = α·SH + β·S, where α and β are weight coefficients.

9. A medical device sterilization packaging visual inspection system, used to implement a medical device sterilization packaging visual inspection method according to any one of claims 1 to 8, characterized in that: include: Correction module: Correct the original packaging image obtained, map the image to the standard coordinate system to eliminate the shooting angle and perspective distortion, and use the predetermined reference template to perform area recognition and mask extraction on the corrected image to determine the effective detection area of ​​the packaging; Measuring module: within the determined effective detection area, the optical characteristics of the image are expanded in multiple dimensions, and the difference between the corrected image and the reference characteristics of the predetermined reference template is measured to obtain data describing the optical abnormality characteristics of the packaging surface; Calculation module: According to the predetermined seal line position definition, the deviation between the pixel points in the same effective detection area and the seal line reference features is calculated, and the seal line consistency index is obtained through weighted and nonlinear deviation aggregation to reflect the deviation of the seal line in structure and intensity distribution; Judgment module: A comprehensive judgment index is calculated based on the surface optical abnormality feature data and the sealing line consistency index, which is used to comprehensively evaluate the appearance and sealing status of the package. The comprehensive judgment index is compared with the predetermined threshold. When the index exceeds the threshold, an unqualified signal is output, otherwise a qualified signal is output, thereby realizing automated visual inspection and judgment of the appearance and sealing of the sterilized packaging of medical devices.

10. A visual inspection device for sterilized packaging of medical devices, characterized in that: The device comprises: one or more processors; A storage medium for storing one or more programs, which, when executed by the one or more processors, enables the device to implement a visual inspection method for sterilization packaging of medical devices as described in any one of claims 1 to 8.

Images