A Standardized Evaluation Method for Ampoule Neck Defects Detection

Abstract

The present invention discloses a standardized evaluation method for detecting the scorched heads of ampoules. The brightness of the light source and the exposure of the industrial camera are adjusted to appropriate parameters, and a blank background image is collected. After the rotating base makes the scorched head at the bottle head of the ampoule face the industrial camera, an image is collected. The collected image is subjected to mean filtering and adaptive thresholding to accurately obtain the scorched head, and the pixel size of the area of the scorched head is collected. The pixel size of the area of the scorched head is subjected to morphological processing, and the pixel size of the area after dilation is collected. The difference between the gray mean value of the scorched head and the area after its dilation is calculated. The foreign object detection device for ampoules verifies whether there are foreign objects in the liquid of this ampoule due to the shedding of the scorched head. The present invention can quantify the scorched head detection standard of QA, save the waste of production and labor costs caused by the randomness of QA, and the detection of verifying the quality of the scorched head by using the method of collecting images by the computer 3 can track the QA data, with wide applicability and high economy, and is suitable for popularization and use.

Description

Technical Field

[0001] The present invention relates to the technical field of detecting burnt heads of pharmaceutical ampoules, and more specifically, to a standardized evaluation method for detecting burnt heads of ampoules. Background Art

[0002] During the wire drawing and sealing process of ampoules, due to factors such as incorrect needle position, when the needle exits the ampoule, the remaining drops of the needle stick to the neck wall, the position of the needle entering the ampoule is too deep or too shallow, resulting in the liquid medicine splashing or backspraying onto the neck wall during medicine filling, and incorrect travel of the needle during medicine filling, etc., it is very easy for the bottle head to produce burnt heads. The burnt heads are manifested as black substances, and the depth of the black substances has light and dark colors. The fallen burnt heads into the ampoule injection become visible foreign matters, which affect human life and health. Therefore, pharmaceutical factories producing ampoule injections are very strict about the detection of burnt heads and conduct multiple detections during the production process. At present, the evaluation criteria for detecting burnt heads are arbitrary. As long as the QA faintly sees light-colored dots on the bottle head, it is judged as unqualified. Some even require that if more than two burnt heads are missed in a batch (generally about 600,000 pieces), the whole batch has to be reworked. Due to the lack of standardized evaluation, it will cause waste of production costs during the detection of burnt heads on the ampoule bottle heads.

[0003] Therefore, the present invention proposes a standardized evaluation method for detecting burnt heads, which objectively evaluates the detection of burnt heads through scientific, standardized and quantified criteria. Through this method, not only the detection of burnt heads can be objectively evaluated, but also the waste of production costs caused by the arbitrariness of QA can be avoided. Summary of the Invention

[0004] To solve the above technical problems, the present invention provides a standardized evaluation method for detecting burnt heads of ampoules, which is characterized by including the following steps:

[0005] S1. The lamp inspection machine runs without load first, adjusts the brightness of the light source and the exposure of the industrial camera to appropriate parameters, collects an empty background image, and the computer calculates the average gray value of the empty background image;

[0006] S2. Based on S1, fix the ampoule on the bottle rotating seat of the lamp inspection machine. After the rotating seat makes the burnt head at the bottle head of the ampoule face the industrial camera, collect an image;

[0007] S3. Based on S2, perform mean filtering and adaptive thresholding on the collected image to accurately obtain the burnt head, and collect the pixel size of the area of the burnt head;

[0008] S4. Based on S3, perform morphological processing on the pixel size of the area of the burnt head, expand it with a circular structure by D (D>3), and continue to collect the pixel size of the expanded area;

[0009] S5. Based on S4, calculate the difference between the gray - scale mean value of the focal head and the region after its expansion;

[0010] S6. Based on S5, shake the ampoule bottle so that the focal head falls off into the ampoule injection liquid until the industrial camera can no longer capture this focal head;

[0011] S7. Based on S6, manually or with an ampoule bottle foreign - object detection device, verify whether there are foreign objects in the liquid of this ampoule bottle due to the falling - off of the focal head;

[0012] S8. Based on S7, according to the verification results of manual or the ampoule bottle foreign - object detection device, repeat the experimental comparison to give an evaluation method for the quality detection of the focal head.

[0013] Preferably, in the step S5, the calculation method steps for the difference between the gray - scale mean value of the focal head and the region after its expansion are as follows:

[0014] (1). Denote the distance between the ampoule bottle head and the light source as M, and the distance between the ampoule bottle head and the industrial camera as m;

[0015] (2). Denote the regional pixel size of the focal head obtained in step S3 as Region1, and calculate the gray - scale mean value of Region1 as X;

[0016] (3). Perform morphological processing on Region1, apply step S4, denote the regional pixel size after expansion as Region2, and find the gray - scale mean value of the region Region2 - Region1, denoted as Y;

[0017] (4). Denote the difference between the gray - scale mean value of the focal head and the region after its expansion as Z, and Z = Y - X.

[0018] Preferably, in the step S8, if manual or the ampoule bottle foreign - object detection device verifies that there are foreign objects, the value of Z can be further reduced until when the value of Z is reduced to a certain value and no foreign objects can be seen by manual or the ampoule bottle foreign - object detection device, the value of Z at this time is the standardized test value for the focal head detection, and the quality detection of the focal head is controlled by this value of Z.

[0019] Preferably, in the step S8, if manual or the ampoule bottle foreign - object detection device does not detect any foreign objects in the liquid of the ampoule bottle, then the liquid in this ampoule bottle is qualified.

[0020] Compared with the prior art, the present invention has the following beneficial effects:

[0021] (1) The detection method of the present invention can quantify the focal head detection standard of QA, save the waste of production and labor costs caused by the randomness of QA. Verifying the quality detection of the focal head by means of computer image acquisition can track the QA data, has wide applicability and high economy, and is suitable for popularization and use. Description of the Drawings

[0022] Figure 1 This is a schematic diagram of the overall structure of the present invention.

[0023] Figure 2 This is one of the schematic diagrams of the imaging for the capping detection and evaluation experiment of the present invention.

[0024] Figure 3 This is another schematic diagram of the imaging for the capping detection and evaluation experiment of the present invention.

[0025] Figure 4 This is the third schematic diagram of the imaging for the capping detection and evaluation experiment of the present invention.

[0026] Figure 5 This is the fourth schematic diagram of the imaging for the capping detection and evaluation experiment of the present invention.

[0027] Figure 6 This is the fifth schematic diagram of the imaging for the capping detection and evaluation experiment of the present invention. Detailed implementation manners

[0028] The present invention will be further described below in conjunction with the accompanying drawings and specific implementation manners.

[0029] As Figures 1 to 6 shown, a standardized evaluation method for ampoule capping detection includes a light source 1, an industrial camera 2, a computer 3, an ampoule 4, a vial rotating base 5, a capping 6, and an ampoule bottle head detection range 7.

[0030] In the description of the present invention, unless otherwise specified, "a plurality of" means two or more; the orientation or positional relationships indicated by the terms "upper", "lower", "left", "right", "inner", "outer", "front end", "rear end", "head", "tail", etc. are based on the orientation or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as a limitation to the present invention. In addition, the terms "first", "second", "third", etc. are only used for descriptive purposes and cannot be construed as indicating or implying relative importance.

[0031] In the description of the present invention, it should be noted that unless otherwise clearly specified and defined, the terms "connected" and "coupled" should be understood in a broad sense. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0032] As Figure 1As shown, the standardized evaluation method for ampoule head detection includes the following steps:

[0033] S1. The lamp inspection machine runs without load first. Adjust the brightness of light source 1 and the exposure of industrial camera 2 to appropriate parameters, collect an empty background image, and computer 3 calculates the average gray value of the empty background image;

[0034] S2. Based on S1, fix the ampoule 4 on the bottle rotating seat 5 of the lamp inspection machine. After the rotating seat makes the head 6 at the bottle head of the ampoule 4 face the industrial camera 2, collect an image;

[0035] S3. Based on S2, perform mean filtering and adaptive thresholding on the collected image to accurately obtain the head 6, collect the pixel size of the area of the head 6. In the present invention, the image algorithm used does not limit the area size of the head 6 obtained by mean filtering and adaptive thresholding, and any existing technology that can obtain the size of the head 6 through image segmentation technology can be used;

[0036] S4. Based on S3, perform morphological processing on the pixel size of the area of the head 6, expand it with a circular structure D (D>3), and continue to collect the pixel size of the expanded area. In the present invention, the expansion with a circular structure is a preferred embodiment. In the actual use process, the expansion morphology is limited to a circular structure, and the expansion of other structural morphologies can also achieve the calculation and processing effect;

[0037] S5. Based on S4, calculate the difference between the gray mean value of the head 6 and its expanded area. In the present invention, as a preferred embodiment, calculate the mean gray difference between the head 6 and its domain area. In the actual use process, the standard such as variance difference and sum difference can also be used to compare its parameters;

[0038] S6. Based on S5, shake the ampoule 4 so that the head 6 falls off into the ampoule injection solution until the industrial camera 2 can no longer capture this head 6;

[0039] S7. Based on S6, manually or with an ampoule foreign object detection device, verify whether there are foreign objects in the liquid of this ampoule 4 due to the fall of the head 6;

[0040] S8. Based on S7, repeat the experiment and comparison according to the verification result of the manual or ampoule foreign object detection device to give an evaluation method for head quality detection. Specifically, if the manual or ampoule foreign object detection device verifies that there are foreign objects, the Z value can continue to be reduced until when the Z value is reduced to a certain value and the manual or ampoule foreign object detection device can no longer see foreign objects, the Z value at this time is the standardized test value for head detection, and the head quality detection is controlled by this Z value; if the manual or ampoule foreign object detection device does not detect any foreign objects in the liquid of the ampoule 4, then the liquid of this ampoule 4 is qualified.

[0041] In step S5, the steps for calculating the difference between the gray-scale mean of the ampoule head and its expanded area are as follows:

[0042] (1) Denote the distance between the head of the ampoule bottle 4 and the light source 1 as M, and the distance between the head of the ampoule bottle 4 and the industrial camera 2 as m;

[0043] (2) Denote the pixel size of the area of the ampoule head 6 obtained in step S3 as Region1, and calculate the gray-scale mean of Region1 as X;

[0044] (3) Perform morphological processing on Region1. Apply step S4. Denote the pixel size of the expanded area as Region2 after expansion, and calculate the gray-scale mean of the area Region2 - Region1 as Y;

[0045] (4) Denote the difference between the gray-scale mean of the ampoule head 6 and its expanded area as Z, and Z = Y - X.

[0046] For the embodiment of applying the standardized evaluation method for detecting the ampoule head of the present invention, the experimental imaging schematic diagram is as Figures 2 to 6 shown; related experimental equipment: industrial camera Canadian GreyPoint: Flea2 FL2G-13S2M, Shanghai Jiali light source: JL-BRL-101X76-GL-C, Shanghai Jiali light source controller JL-APH2-6024-4, computer Advantech AIMB-784 I7 4790K / 8g / 500g.

[0047] Take M = 50mm and m = 80mm, and detect the detection range 7 of the ampoule bottle head;

[0048] Start running step S1. As Figure 2 shown, the mean value of the blank background image is 254;

[0049] Run step S2. As Figure 3 shown, it can be obtained that there is a marked ampoule head within the black circle;

[0050] Run step S3. Mark the image of Region1 after image processing with a black circle as Figure 4 shown, and its gray-scale mean X = 209.378;

[0051] Run step S4. Perform morphological processing on Region1. Expand it with a circular structure D = 10. After expansion, obtain Region2 as the large black outer circle as Figure 5 shown, and calculate the gray-scale mean of the area Region2 - Region1 as Y = 225.26;

[0052] Run step S5. Calculate the difference Z between the gray-scale mean of the ampoule head and its expanded area. Z = Y - X, and Z = 15.882;

[0053] Run step S6 to shake the ampoule bottle 4 so that the burnt heads fall off into the ampoule injection solution until the industrial camera 2 cannot capture the burnt heads as shown in Figure 6 the figure;

[0054] Run step S7. Through a large number of similar repeated experiments, it is obtained that Z≤16. Even if the burnt heads fall into the liquid, no foreign objects will be formed, which meets the GMP standard for the inspection of ampoule bottle liquids. Manual inspection can distinguish the difference of Z = 10 under a white background of light. Therefore, through this method, the inspection standard for burnt heads can be effectively and objectively quantified, the production efficiency of pharmaceutical factories can be improved, and the production cost can be saved.

[0055] The inspection method of the present invention can not only be applied to the focus inspection of the ampoule bottle head, but also be applicable to the inspection standards of other light-transmitting or semi-light-transmitting materials, such as black objects like glass, plastic, and film products.

[0056] By using the inspection method of the present invention, the inspection standard for burnt heads of QA is quantified, and the waste of production and labor costs caused by the randomness of QA is saved. The inspection of the quality of burnt heads verified by the method of collecting images by the computer 3 can track the QA data. It has a wide range of applicability and high economy, and is suitable for popularization and use.

[0057] The above embodiments are only the preferred embodiments of the present invention and cannot be used to limit the scope of the rights of the present invention. Therefore, modifications, equivalent changes, improvements, etc. made according to the scope of the patent application of the present invention still fall within the scope covered by the present invention.

Claims

1. A standardized evaluation method for detecting the burnt heads of ampoules, characterized in that, It includes the following steps: S1. The lamp inspection machine runs without load first, adjusts the brightness of the light source (1) and the exposure of the industrial camera (2) to appropriate parameters, acquires an empty background image, and the computer (3) calculates the average gray value of the empty background image; S2. Based on S1, fix the ampoule bottle (4) on the bottle rotating base (5) of the lamp inspection machine. After the rotating base makes the burnt head (6) at the bottle head of the ampoule bottle (4) face the industrial camera (2), acquire an image; S3. Based on S2, perform mean filtering and adaptive thresholding on the acquired image to accurately obtain the burnt head (6), and acquire the pixel size of the area of the burnt head (6); S4. Based on S3, perform morphological processing on the pixel size of the area of the burnt head (6), expand it with a circular structure by D, where D > 3, and continue to acquire the pixel size of the expanded area; S5. Based on S4, calculate the difference between the gray mean value of the burnt head (6) and its expanded area; S6. Based on S5, shake the ampoule bottle (4) so that the burnt head (6) falls off into the ampoule injection liquid until the industrial camera (2) can no longer capture this burnt head (6); S7. Based on S6, manually or with an ampoule bottle foreign object detection device verify whether there are foreign objects in the liquid of this ampoule bottle (4) due to the falling off of the burnt head; S8. Based on S7, repeat the experiment and comparison according to the verification result of the manual or the ampoule bottle foreign object detection device, and give an evaluation method for the quality detection of the burnt head; In the step S5, the calculation method steps for the difference between the gray mean value of the burnt head (6) and its expanded area are as follows: (1). Denote the distance between the bottle head of the ampoule bottle (4) and the light source (1) as M, and the distance between the bottle head of the ampoule bottle (4) and the industrial camera (2) as m; (2). Denote the pixel size of the area of the burnt head (6) obtained in step S3 as Region1, and calculate the gray mean value of Region1 as X; (3). Perform morphological processing on Region1, apply step S4, denote the pixel size of the expanded area as Region2, and obtain the gray mean value of the area of Region2 - Region1, denoted as Y; (4). Denote the difference between the gray mean value of the burnt head (6) and its expanded area as Z, and Z = Y - X; In the step S8, if the manual or the ampoule bottle foreign object detection device verifies that there are foreign objects, the value of Z continues to decrease until when the value of Z decreases to a certain value and the manual or the ampoule bottle foreign object detection device can no longer see foreign objects, the value of Z at this time is the standardized test value for the burnt head detection, and the quality detection of the burnt head is controlled by this value of Z.

2. The standardized evaluation method for ampoule vial head defect detection according to claim 1, characterized in that: In the step S8, if the manual or the ampoule bottle foreign object detection device does not detect any foreign objects in the liquid of the ampoule bottle (4), then the liquid of this ampoule bottle (4) is qualified.

Images