Detection method of pre-encapsulated nest box

Abstract

The invention relates to the technical field of aircraft parts, and provides a method for detecting a pre-encapsulated nest box, which comprises the following steps of: S1, acquiring edge lines of Tyvek paper and the inner side of the nest box by using a CCD (Charge Coupled Device) detection camera; s2, a CCD detection camera is used for shooting the weld joint effect of the outer layer paper of the nest box which is preliminarily detected to be qualified, and a weld joint image is obtained; s3, calculating to obtain a weld width expression influence factor W; s4, calculating to obtain a weld joint uniformity expression influence factor U0; s5, calculating to obtain a weld joint area expression influence factor; s6, a welding seam quality factor # imgabs0 # is obtained through calculation; and S7, the weld joint quality factor # imgabs1 # obtained through calculation is compared with a preset weld joint quality factor. The method has the beneficial effects that the optical imaging or laser sensing technology is adopted, an image processing algorithm is combined, the welding quality can be accurately detected, the contradiction between the sealing performance and the easy-to-tear performance can be balanced, detection is achieved on the premise that packaging is not damaged, and traditional manual sampling destructive testing is replaced.

Description

Technical Field

[0001] The invention relates to the technical field of detection methods for pre-filled nest boxes, and in particular to a detection method for pre-filled nest boxes. Background Art

[0002] As an important packaging form in the modern medical field, prefilled syringes are widely used in vaccines, biological preparations, emergency drugs and other scenarios due to their convenience, sterility and high-precision drug delivery characteristics. As the core protective structure of prefilled syringes, nested box packaging must have both sealing, easy opening and physical protection functions. Its quality directly affects the safety and efficiency of drug use. However, the current production process of prefilled nested box packaging still faces technical bottlenecks: For example, the welding process between the outer paper and the nest box has the problem of insufficient stability. Fluctuations in welding temperature, pressure or time parameters can easily lead to uneven weld strength, which manifests as poor sealing effect (such as air leakage, liquid leakage) or excessive welding (such as excessive tearing force). The former may cause the drug to become ineffective during transportation or storage, while the latter affects the rapid access of medical staff, especially in emergency scenarios, which may delay treatment.

[0003] In addition, traditional quality inspection methods mostly rely on manual visual inspection and sampling destructive testing, which are inefficient and lack coverage, making it difficult to achieve real-time monitoring of the entire production process. Existing automated inspection systems have limited recognition accuracy for micron-level deviations and lack the ability to quantitatively evaluate welding strength and tearing performance, resulting in a high risk of defective products flowing into downstream links.

[0004] As the medical industry continues to increase its requirements for packaging safety and compliance, developing a high-precision, non-destructive online detection method to monitor Tyvek paper welding parameters has become a key technical requirement for improving the quality of pre-filled nest box packaging. Summary of the invention

[0005] The present invention aims to solve the problems of weak welding of pre-filled nest box paper, poor sealing effect, or difficulty in tearing after welding; The present invention provides a method for detecting a pre-filled nest box, comprising the following steps: S1. When the nest box reaches the detection position, the light source of the corresponding weld is turned on, and the edge line of the Tyvek paper and the inner edge line of the nest box is acquired by using a CCD detection camera. The directional analysis module converts the image information into digital information, compares the distance L between the edge line of the Tyvek paper and the inner edge line of the nest box and the angle α formed between the corresponding edge lines, and compares them with the preset distance threshold L' and the preset angle threshold α'. If the values ​​of the distance L and the angle α are less than the preset thresholds, it means that the placement position of the inner layer of Tyvek paper is qualified, otherwise it is unqualified; S2. Use a CCD detection camera to shoot the weld effect of the outer paper of the nest box that has passed the preliminary inspection to obtain a weld image, perform semantic segmentation processing on the obtained weld image, obtain the nest box welding surface area image and background information, mark the contours of the edges of the inner paper and the nest box and the weld image of the outer paper, and record the weld brightness, and determine the welding area according to the gray value change of the image area composed of continuous edge points; S3. Calculate the influencing factor W of the weld width; ; Among them, β-γ represents the difference between two adjacent pixels at the edge of the weld in the image, and d represents the width variance of the weld corresponding to the edge image pixel point; S4. Calculate the influencing factors of weld uniformity ; ; in, Indicates the minimum brightness of the weld. Indicates the average brightness of the weld; S5. Calculate the influence factor of weld area performance ; ; Among them, F(x) represents the corresponding fitting relationship curve between gray value and exposure, a, b are the pixel points of the initial point and the end point of the weld, Indicates the preset maximum area threshold; S6. Calculate the weld quality factor ; ; S7. The calculated weld quality factor Compared with the preset minimum weld quality factor and maximum weld quality factor, if the weld quality factor Falling into the minimum weld quality factor and the maximum weld quality factor If the weld quality is between 0.01 and 0.02, the weld quality is qualified, otherwise it is unqualified.

[0006] As a preferred solution, the semantic segmentation processing in S2 is to perform instance segmentation of the image into independent image individuals and enhance marginalization processing, and perform brightness test on the obtained detection image.

[0007] As a preferred solution, the number of CCD detection cameras in S1 is 4, corresponding to the four sides of the nest box respectively.

[0008] As a preferred solution, the CCD detection camera collects and obtains the edge lines of the Tyvek paper and the inner edge line of the nest box, and the directional analysis module converts the image information into digital information, compares the distance L between the four edge lines of the Tyvek paper and the inner edge line of the nest box and the angle α formed between the corresponding edge lines, and compares them with the preset distance threshold L' and the preset angle threshold α'. If the values ​​of the distance L and the angle α are less than the preset thresholds, it means that the placement position of the inner layer of Tyvek paper is qualified, otherwise it is unqualified.

[0009] The beneficial effects of the present invention are: The present invention uses optical imaging or laser sensing technology, combined with image processing algorithms, to accurately detect welding quality, balance the contradiction between sealing and tearability, and achieve detection without damaging the packaging, replacing traditional manual sampling destructive testing. The system uses high-speed data acquisition and processing to output detection results in real time and automatically remove defective products from the production line, significantly improving detection efficiency and defect interception rate, and preventing defective packaging from entering the market. DETAILED DESCRIPTION

[0010] To illustrate the features of the present invention, the present invention will be further described below in conjunction with embodiments.

[0011] Example: The embodiment of the present invention provides a method for detecting a pre-filled nest box, comprising the following steps: S1. When the nest box reaches the detection position, the light source of the corresponding weld is turned on, and the edge lines of the Tyvek paper and the inner edge of the nest box are acquired by using a CCD camera. The directional analysis module converts the image information into digital information, and compares the distance L between the edge line of the Tyvek paper and the inner edge line of the nest box, which is 3.5 mm, and the angle α between the corresponding edge lines, which is 2.7 degrees. The information is also compared with the preset distance threshold L' of 5 mm and the preset angle threshold α' of 3 degrees. If the values ​​of the distance L and the angle α are less than the preset thresholds, it means that the placement position of the inner layer of Tyvek paper is qualified, otherwise it is unqualified. S2. Use a CCD detection camera to shoot the weld effect of the outer paper of the nest box that has passed the preliminary inspection to obtain a weld image, perform semantic segmentation on the obtained weld image, obtain the nest box welding surface area image and background information, perform instance segmentation on the image into independent image individuals and strengthen edge processing, perform brightness test on the obtained detection image, mark the contours of the edges of the inner paper and the nest box and the weld image of the outer paper and record the weld brightness, and determine the welding area according to the gray value change of the image area composed of continuous edge points; S3. Calculate the influencing factor W of the weld width; = ; Among them, β-γ represents the difference between two adjacent pixels at the edge of the weld in the image, and d represents the width variance of the weld corresponding to the edge image pixel point; Among them: β-γ includes {210, 83, 178, 45, 71, 364, 314}, d= ; S4. Calculate the influencing factors of weld uniformity ; ; in, Indicates the minimum brightness of the weld. Indicates the average brightness of the weld; S5. Calculate the influence factor of weld area performance ; ; Among them, F(x) represents the corresponding fitting relationship curve between gray value and exposure, a, b are the pixel points of the initial point and the end point of the weld, Indicates the preset maximum area threshold; S6. Calculate the weld quality factor ; = ; S7. The calculated weld quality factor Compare with the preset minimum weld quality factor and maximum weld quality factor, ; Wherein δ represents the weld quality factor, δmin represents the preset minimum weld quality factor, and δmax represents the preset maximum weld quality factor.

[0012] Weld quality factor If the weld quality falls between the minimum weld quality factor and the maximum weld quality factor, the weld quality is qualified.

[0013] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit the present invention. The present invention is described in detail with reference to the preferred embodiments. It should be understood by those skilled in the art that any changes, modifications, additions or substitutions made by those skilled in the art within the essential scope of the present invention do not deviate from the purpose of the present invention and should also fall within the scope of the claims of the present invention. Other related technical structures not fully disclosed in the present invention are prior art in the art.

Claims

1. A method for detecting a pre-filled nest box, characterized in that: The following steps are involved: S1. When the nest box reaches the detection position, the light source of the corresponding weld is turned on, and the edge line of the Tyvek paper and the inner edge line of the nest box is acquired by using a CCD detection camera. The directional analysis module converts the image information into digital information, compares the distance L between the edge line of the Tyvek paper and the inner edge line of the nest box and the angle α formed between the corresponding edge lines, and compares them with the preset distance threshold L' and the preset angle threshold α'. If the values ​​of the distance L and the angle α are less than the preset thresholds, it means that the placement position of the inner layer of Tyvek paper is qualified, otherwise it is unqualified; S2. Use a CCD detection camera to shoot the weld effect of the outer paper of the nest box that has passed the preliminary inspection to obtain a weld image, perform semantic segmentation processing on the obtained weld image, obtain the nest box welding surface area image and background information, mark the contours of the edges of the inner paper and the nest box and the weld image of the outer paper, and record the weld brightness, and determine the welding area according to the gray value change of the image area composed of continuous edge points; S3. Calculate the influencing factor W of the weld width; ； Among them, β-γ represents the angle difference between adjacent pixels at the edge of the weld in the image, and d represents the variance of the weld width at the corresponding position of the pixel at the edge of the weld image; S4. Calculate the influence factor U0 of weld uniformity performance; ； Among them, E min Indicates the minimum brightness of the weld, E av Indicates the average brightness of the weld; S5. Calculate the influence factor S of the weld area performance; ； Among them, F(x) represents the corresponding fitting relationship curve between gray value and exposure, a and b are the pixel values ​​corresponding to the initial point and the end point of the weld in the two-dimensional coordinate system, respectively, and A max Indicates the preset maximum area threshold; S6. Calculate the weld quality factor ; ； S7. The calculated weld quality factor Compared with the preset minimum weld quality factor and maximum weld quality factor, if the weld quality factor If the weld quality falls between the minimum weld quality factor δmin and the maximum weld quality factor δmax, the weld quality is qualified, otherwise it is unqualified.

2. The method for detecting a prefilled nest box according to claim 1, characterized in that: The semantic segmentation process in S2 is to segment the image into independent image individuals and strengthen the marginalization process, and then perform brightness test on the obtained detection image.

3. The method for detecting the prefilled nest box according to claim 1, characterized in that: The number of CCD detection cameras in S1 is 4, corresponding to the four sides of the nest box.

Images