Quality control method for production and preparation of Xiangdan injection

Abstract

The invention discloses a quality control method for production and preparation of a Xiangdan injection. The method comprises the following steps: step 1, shooting the Xiangdan injection to obtain an image A, and preprocessing the image A by using an improved CLAHE algorithm; step 2, carrying out anomaly detection on the preprocessed image by using a trained improved YOLOv5s network model; step 3, if the detection result is not abnormal, indicating that the Xiangdan injection is qualified; if the detection result is abnormal, reminding is fed back to the user. According to the method, the accuracy of abnormity of the Xiangdan injection can be improved, the accuracy of foreign matter detection can be further improved by removing bottle body scratches, fault reasons can be accurately fed back to a production and manufacturing end, and the production and manufacturing efficiency and the purity of the Xiangdan injection are improved.

Description

Technical Field

[0001] The invention relates to the technical field of quality control of Xiangdan injection production and preparation, and specifically to a quality control method of Xiangdan injection production and preparation. Background Art

[0002] Xiangdan injection is a pure Chinese medicine preparation that can be used to treat vasodilation, increase coronary blood flow, and can be used for angina pectoris and myocardial infarction. Impurities are inevitable in the injection of this pure Chinese medicine preparation, and the impurities in the liquid can cause hypersensitivity, diarrhea, muscle tremor, sinus tachycardia, central nervous system symptoms, hemolytic uremic syndrome, shock, acute laryngeal edema, liver damage, asthma, oral sores, gastrointestinal bleeding, exfoliative dermatitis, high blood pressure and other adverse reactions. Therefore, Xiangdan injection is prohibited from use if there are precipitation or impurities to prevent harm to patients. Therefore, how to control the quality of the production and preparation of Xiangdan injection and improve the purity of Xiangdan injection is a problem that needs to be solved urgently. Summary of the invention

[0003] The purpose of the present invention is to provide a quality control method for the production and preparation of Xiangdan injection, which can detect impurities in Xiangdan injection and thus improve the purity of Xiangdan injection.

[0004] To achieve the above-mentioned purpose, the first embodiment of the present application proposes a quality control method for the production of Xiangdan injection, comprising the following steps:

[0005] Step 1, photographing Xiangdan injection to obtain image A, and using the improved CLAHE algorithm to preprocess image A;

[0006] Step 2: Use the trained improved YOLOv5s network model to perform anomaly detection on the above preprocessed images;

[0007] Step 3: If the test result is normal, it indicates that Xiangdan injection is qualified; if the test result is abnormal, a feedback reminder is given to the user.

[0008] Preferably, the preprocessing of image A using the improved CLAHE algorithm comprises the following steps:

[0009] Step 10, dividing the image A into N regions of the same size;

[0010] Step 20, count the grayscale histogram of each region, and then clip the histogram of each region according to the initial clipping threshold.

[0011] Step 30, counting the number of pixels clipped from each area, and setting a new clipping threshold for each block according to the number of pixels clipped from each area;

[0012] Step 40, cropping the grayscale histogram of each region again according to the new cropping threshold, and evenly distributing the cropped parts to the entire interval.

[0013] Step 50, obtaining a corresponding histogram cumulative distribution function according to the clipped grayscale histogram, and mapping it to a specified grayscale space;

[0014] Step 60: Perform bilinear interpolation operation according to the pixel position and the grayscale mapping function of each region to obtain the enhanced image A.

[0015] Preferably, the use of the trained improved YOLOv5s network model to perform foreign body detection on the preprocessed image A comprises the following steps:

[0016] Step 210, improve the Yolov5s model: add SE modules after the second CSP1 module, the third CSP1 module and the SPP module respectively; use GSConv convolution to replace the ordinary feature layer in the neck network, and the specific implementation method is: first generate high-dimensional features through ordinary convolution, then use depthwise separable convolution to transform the high-dimensional features, then concatenate the two features obtained in the above two steps, and shuffle the feature maps of different groups together through shuffling operations; use the improved Kmeans algorithm to cluster the initial anchor boxes.

[0017] Step 220, training the improved Yolov5s model using the training sample set to obtain a trained improved Yolov5s model;

[0018] Step 230: input the preprocessed image A into the trained improved Yolov5s model for foreign body detection.

[0019] Preferably, clustering the initial anchor boxes using the improved Kmeans algorithm comprises the following steps:

[0020] Step 00, randomly determine the first cluster center point;

[0021] Step 01, calculate the shortest distance from each sample point to the cluster center point;

[0022] Step 02: Calculate the probability of each sample point becoming the next cluster center based on the shortest distance from the sample point to the cluster center, and select the sample point with the maximum probability as the new cluster center;

[0023] Step 03, repeat steps 01 and 02 until K cluster centers are determined.

[0024] Preferably, if the test result is abnormal, execute step 4;

[0025] Step 4, execute the operation steps for removing scratches on the bottle body of Xiangdan Injection bottle; if it is determined that the abnormality is a scratch on the bottle body of Xiangdan Injection bottle, then the Xiangdan Injection is determined to be qualified; if it is determined that the abnormality is not a scratch on the bottle body of Xiangdan Injection bottle, then the Xiangdan Injection is determined to be unqualified, that is, there is foreign matter in the Xiangdan Injection, and feedback is given to the user.

[0026] By adopting the above technical solution, the present invention achieves the following beneficial effects:

[0027] Firstly, the improved CLAHE algorithm is used to preprocess the image to be detected, which can improve the accuracy of image detection;

[0028] Secondly, by using the improved YOLOv5s network model for foreign body detection, the accuracy of foreign body detection can be greatly improved;

[0029] Again, the above-mentioned quality control method for the production and preparation of Xiangdan Injection can distinguish between real foreign matter and scratches on the bottle body, thereby accurately achieving quality control of Xiangdan Injection.

[0030] Finally, the quality control method for the production and preparation of Xiangdan injection of the present application can provide accurate feedback on faults, abnormalities or errors in the production process through foreign matter, improve the efficiency of production and improve the purity of Xiangdan injection. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 This is a schematic diagram of the quality control method for the production of Xiangdan injection of the present invention. DETAILED DESCRIPTION

[0032] In order to more clearly understand the above-mentioned purposes, features and advantages of the present application, the present application is further described in detail below in conjunction with the accompanying drawings and specific implementation methods. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0033] In the following description, many specific details are set forth to facilitate a full understanding of the present application. However, the present application may also be implemented in other ways different from those described herein. Therefore, the scope of protection of the present application is not limited by the specific embodiments disclosed below. In order to make the purpose, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0034] The terms used in the embodiments of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The singular forms "a", "said" and "the" used in the embodiments of the present invention and the appended claims are also intended to include plural forms, unless the context clearly indicates other meanings, and "multiple" generally includes at least two.

[0035] It should be understood that the term "and / or" used in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship.

[0036] As used herein, the words "if" and "if" may be interpreted as "at the time of" or "when" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrases "if it is determined" or "if (stated condition or event) is detected" may be interpreted as "when it is determined" or "in response to determining" or "when detecting (stated condition or event)" or "in response to detecting (stated condition or event)", depending on the context.

[0037] It should also be noted that the term "includes", "comprising" or any other variation thereof is intended to cover non-exclusive inclusion, so that a commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprising a ..." do not exclude the existence of other identical elements in the commodity or device including the elements.

[0038] The optional embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0039] Reference Figure 1 This embodiment provides a quality control method for the production of Xiangdan injection, comprising the following steps:

[0040] Step 1, photographing Xiangdan injection to obtain image A, and using the improved CLAHE algorithm to preprocess image A;

[0041] Step 2: Use the trained improved YOLOv5s network model to perform anomaly detection on the preprocessed images;

[0042] It should be noted that the sample set used for training the network model includes positive samples and negative samples. Positive samples refer to samples containing anomalies, where anomalies include scratches on the bottle body and foreign matter in the bottle. Negative samples refer to samples that do not contain anomalies.

[0043] Step 3: If the test result is normal, it indicates that Xiangdan injection is qualified; if the test result is abnormal, a feedback reminder is given to the user.

[0044] Therefore, the present invention improves a quality control method for the production and preparation of Xiangdan injection, which first preprocesses image A through an improved CLAHE algorithm, and then uses a trained improved YOLOv5s network model to judge the abnormality of Xiangdan injection. When the abnormality of Xiangdan injection is determined, in order to improve the accuracy of judgment, the scratches on the bottle body are further eliminated to improve the accuracy of monitoring, thereby improving the quality control of production and preparation and improving the purity of Xiangdan injection.

[0045] According to an embodiment of the present application, the preprocessing of image A using the improved CLAHE algorithm comprises the following steps:

[0046] Step 10, dividing the image A into N regions of the same size;

[0047] Step 20, counting the grayscale histogram of each region, and then clipping the histogram of each region according to the initial clipping threshold;

[0048] Step 30, counting the number of pixels clipped from each area, and setting a new clipping threshold for each block according to the number of pixels clipped from each area;

[0049] Step 40, cropping the grayscale histogram of each region again according to the new cropping threshold, and evenly distributing the cropped parts to the entire interval;

[0050] Step 50, obtaining a corresponding histogram cumulative distribution function according to the clipped grayscale histogram, and mapping it to a specified grayscale space;

[0051] Step 60: Perform bilinear interpolation operation according to the pixel position and the grayscale mapping function of each region to obtain the enhanced image A.

[0052] According to an embodiment of the present application, the use of the trained improved YOLOv5s network model to perform foreign body detection on the preprocessed image A comprises the following steps:

[0053] Step 210, improving the Yolov5s model: adding SE modules after the second CSP1 module, the third CSP1 module and the SPP module respectively; using GSConv convolution to replace the common feature layer in the neck network, specifically implementing the following: firstly generating high-dimensional features through common convolution, then transforming the high-dimensional features through depthwise separable convolution, then concatenating the two features obtained in the above two steps, and shuffling the feature maps of different groups together through shuffling operation; clustering the initial anchor boxes using the improved Kmeans algorithm;

[0054] Step 220, training the improved Yolov5s model using the training sample set to obtain a trained improved Yolov5s model;

[0055] Step 230: input the preprocessed image A into the trained improved Yolov5s model for foreign body detection.

[0056] According to an embodiment of the present application, clustering the initial anchor boxes using the improved Kmeans algorithm includes the following steps:

[0057] Step 00, randomly determine the first cluster center point;

[0058] Step 01, calculate the shortest distance from each sample point to the cluster center point;

[0059] Step 02: Calculate the probability of each sample point becoming the next cluster center based on the shortest distance from the sample point to the cluster center, and select the sample point with the maximum probability as the new cluster center;

[0060] Step 03, repeat steps 01 and 02 until K cluster centers are determined.

[0061] Further, if the test result is abnormal, execute step 4;

[0062] Step 4, execute the operation steps for removing scratches on the bottle body of Xiangdan Injection bottle; if it is determined that the abnormality is a scratch on the bottle body of Xiangdan Injection bottle, then the Xiangdan Injection is determined to be qualified; if it is determined that the abnormality is not a scratch on the bottle body of Xiangdan Injection bottle, then the Xiangdan Injection is determined to be unqualified, that is, there is foreign matter in the Xiangdan Injection, and feedback is given to the user.

[0063] Abnormalities in Xiangdan injection include the presence of foreign matter in the Xiangdan injection itself and scratches on the bottle of Xiangdan injection. In order to improve the accuracy of Xiangdan injection detection, it is necessary to further eliminate detection errors caused by bottle body division. Therefore, the present application also performs the operation steps of eliminating scratches on the bottle of Xiangdan injection; if the abnormality is determined to be a scratch on the bottle of Xiangdan injection, the Xiangdan injection is determined to be qualified; if the abnormality is determined to be a scratch on the bottle of non-Xiangdan injection, the Xiangdan injection is determined to be unqualified and foreign matter exists in the Xiangdan injection.

[0064] Furthermore, the operation steps for removing scratches on the bottle body of Xiangdan injection bottle are performed, specifically including: controlling the Xiangdan injection to rotate in a clockwise direction at a rotation speed V1 for a first preset time t1, and after stopping for a second preset time t2, rotating the Xiangdan injection in a counterclockwise direction at a rotation speed V2; when rotating the Xiangdan injection in a counterclockwise direction at the rotation speed V2 for a third preset time t3, using visible light to irradiate the Xiangdan injection and capture a transmission image B, scanning the transmission image B row by row, if the difference in pixel values ​​between two points on the scanning line is greater than a certain threshold value T, it indicates that the abnormality in the image is an impurity in the liquid, otherwise the abnormality in the image is a scratch on the bottle body.

[0065] It should be noted that the method here to distinguish between scratches on the bottle and real impurities is based on the following principle: by rotating the Xiangdan injection in a clockwise direction, stopping for a predetermined time and then rotating in the opposite direction, on the one hand, to gather foreign matter in the center of the bottle for easy detection; on the other hand, to make the scratches on the bottle blurred, so as to determine whether the abnormality is caused by foreign matter.

[0066] Specifically, the clockwise rotation speed V1 is the minimum speed at which the foreign matter in the Xiangdan injection bottle can rotate in the Xiangdan injection. Preferably, V1 is 450 rpm. The first preset time t1 is the time for the foreign matter in the Xiangdan injection bottle to rotate. Preferably, t1 is 1 s.

[0067] The second preset time t2 is greater than the shortest stop time for collecting foreign matter in the Xiangdan injection at the center of the Xiangdan injection bottle. Preferably, t2 is 0.7s.

[0068] The counterclockwise rotation speed V2 is the minimum rotation speed at which the scratch on the bottle of Xiangdan injection bottle rotates horizontally within one frame and is difficult to detect. Preferably, V2 is 710 rpm. T3 is 0.6 s, which is the best time for the scratch on the bottle of Xiangdan injection bottle to be blurred.

[0069] Furthermore, the feedback to the user may be feedback to the user that "there are foreign objects in Xiangdan injection".

[0070] Further, it also includes that if there is a foreign object in the Xiangdan injection, the height of the foreign object from the bottom of the bottle is judged according to the position of the foreign object obtained in step 2 and the preset heights H1 and H2, where H1 < H2. If the height of the foreign object from the bottom of the bottle < H1, a reminder of abnormal pH value adjustment is issued; if the height of the foreign object from the bottom of the bottle > H2, a reminder of abnormal filtration is issued.

[0071] The situations where there are foreign objects in the Xiangdan injection mainly include precipitation and floccules. If there is abnormal filtration during the production and manufacturing of the Xiangdan injection, such as the filtration device malfunctions and the filtration device needs to be replaced, etc., it will cause the floccules in materials such as salvia miltiorrhiza not to be completely filtered, and then the floccules float in the Xiangdan injection as foreign objects; if the pH value is not adjusted accurately during the production and manufacturing of the Xiangdan injection, precipitation will occur in the Xiangdan injection and then accumulate at the bottom of the bottle. Therefore, by determining the position of the foreign object, this application gives a reminder to the front end of production and manufacturing, which facilitates the production manufacturer to accurately know the abnormal links in production and manufacturing, and then facilitates timely adjustment.

[0072] Specifically, the way of reminder can be to send reminder information to the monitor or mobile phone of the production manufacturer, such as "Abnormal pH value adjustment, improvement is needed", or "Abnormal filtration, the filtration device needs to be replaced or repaired".

[0073] In the description of this specification, the descriptions with reference to terms such as "one embodiment", "some embodiments", "example", "specific example", or "some examples" mean that the specific features, structures, materials, or characteristics described in connection with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials, or characteristics described can be combined in a suitable manner in any one or more embodiments or examples. In addition, without contradiction, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of different embodiments or examples.

[0074] In addition, the terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" can explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "a plurality" is at least two, such as two, three, etc., unless otherwise specifically defined.

[0075] Any process or method description in a flowchart or otherwise described herein may be understood to represent a control module, segment or portion of a code comprising one or more executable instructions for implementing the steps of a custom logical function or process, and the scope of the preferred embodiments of the present invention includes alternative implementations in which functions may not be performed in the order shown or discussed, including performing functions in a substantially simultaneous manner or in reverse order depending on the functions involved, which should be understood by technicians in the technical field to which the embodiments of the present invention belong.

[0076] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute the instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purpose of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or otherwise processing in a suitable manner if necessary, and then stored in a computer memory.

[0077] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, a plurality of steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0078] A person skilled in the art may understand that all or part of the steps in the above-mentioned embodiment method may be completed by instructing related hardware through a program, and the program may be stored in a computer-readable storage medium, which, when executed, includes one or a combination of the steps of the method embodiment.

[0079] In addition, each functional unit in each embodiment of the present invention may be integrated into a processing control module, or each unit may exist physically separately, or two or more units may be integrated into a control module. The above-mentioned integrated control module may be implemented in the form of hardware or in the form of a software function control module. If the integrated control module is implemented in the form of a software function control module and sold or used as an independent product, it may also be stored in a computer-readable storage medium.

[0080] The storage medium mentioned above may be a read-only memory, a magnetic disk or an optical disk, etc. Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and cannot be understood as limiting the present invention. A person of ordinary skill in the art may change, modify, replace and modify the above embodiments within the scope of the present invention.

Claims

1. A quality control method for the production of Xiangdan injection, comprising the following steps: Step 1, photographing Xiangdan injection to obtain image A, and using the improved CLAHE algorithm to preprocess image A; Step 2: Use the trained improved YOLOv5s network model to perform anomaly detection on the preprocessed images; Step 3: If the test result is normal, it indicates that Xiangdan injection is qualified; if the test result is abnormal, a feedback reminder is given to the user.

2. The quality control method according to claim 1, characterized in that: The method of preprocessing the image A using the improved CLAHE algorithm comprises the following steps: Step 10, dividing the image A into N regions of the same size; Step 20, counting the grayscale histogram of each region, and then clipping the histogram of each region according to the initial clipping threshold.

3. The quality control method according to claim 2, characterized in that: The method of preprocessing the image A using the improved CLAHE algorithm further includes the following steps: Step 30, counting the number of pixels clipped from each area, and setting a new clipping threshold for each block according to the number of pixels clipped from each area; Step 40, cropping the grayscale histogram of each region again according to the new cropping threshold, and evenly distributing the cropped parts to the entire interval.

4. The quality control method according to claim 3, characterized in that: The method of preprocessing the image A using the improved CLAHE algorithm further includes the following steps: Step 50, obtaining a corresponding histogram cumulative distribution function according to the clipped grayscale histogram, and mapping it to a specified grayscale space; Step 60: Perform bilinear interpolation operation according to the pixel position and the grayscale mapping function of each region to obtain the enhanced image A.

5. The quality control method according to claim 1, characterized in that: The method of using the trained improved YOLOv5s network model to detect foreign objects in the preprocessed image A includes the following steps: Step 210, improve the Yolov5s model: add SE modules after the second CSP1 module, the third CSP1 module and the SPP module respectively; use GSConv convolution to replace the ordinary feature layer in the neck network, and the specific implementation method is: first generate high-dimensional features through ordinary convolution, then use depthwise separable convolution to transform the high-dimensional features, then concatenate the two features obtained in the above two steps, and shuffle the feature maps of different groups together through shuffling operations; use the improved Kmeans algorithm to cluster the initial anchor boxes.

6. The quality control method according to claim 5, characterized in that: The method of using the trained improved YOLOv5s network model to detect foreign objects in the preprocessed image A includes the following steps: Step 220, training the improved Yolov5s model using the training sample set to obtain a trained improved Yolov5s model; Step 230: input the preprocessed image A into the trained improved Yolov5s model for foreign body detection.

7. The quality control method according to claim 6, characterized in that: The improved Kmeans algorithm is used to cluster the initial anchor frames, comprising the following steps: Step 00, randomly determine the first cluster center point; Step 01, calculate the shortest distance from each sample point to the cluster center point; Step 02: Calculate the probability of each sample point becoming the next cluster center based on the shortest distance from the sample point to the cluster center, and select the sample point with the maximum probability as the new cluster center; Step 03, repeat steps 01 and 02 until K cluster centers are determined.

8. The quality control method according to claim 1, characterized in that: If the test result is abnormal, go to step 4; Step 4, execute the operation steps for removing scratches on the bottle body of Xiangdan Injection bottle; if it is determined that the abnormality is a scratch on the bottle body of Xiangdan Injection bottle, then the Xiangdan Injection is determined to be qualified; if it is determined that the abnormality is not a scratch on the bottle body of Xiangdan Injection bottle, then the Xiangdan Injection is determined to be unqualified, that is, there is foreign matter in the Xiangdan Injection, and feedback is given to the user.

Images