CN-121708021-B - Screen defect detection method and system for pharmaceutical production line pharmaceutical process

CN121708021BCN 121708021 BCN121708021 BCN 121708021BCN-121708021-B

Abstract

The invention belongs to the technical field of pharmaceutical production, and particularly discloses a method and a system for detecting screen defects in a pharmaceutical production line pharmaceutical process. The method comprises the steps of minimum checking area based on a common overlapping area of screen surface images of a normal screen, establishing a slight damage standard and a severe damage standard by constructing a defect type calibration disc and setting a parameter threshold value, classifying the screen after judging the screen as a checked screen, marking the checked screen as a target diagnosis screen or a target replacement screen, re-marking the target diagnosis screen meeting preset replacement conditions as a target replacement screen, and distinguishing the screens with different damage degrees by correlating an accuracy difference value with an economic index. According to the invention, the screen mesh incapable of performing effective visual analysis is removed through the pre-screening mechanism, invalid or wrong defect interpretation is avoided on unqualified images, the defect severity is bound with specific parameter values, and the validity of detection results is ensured.

Inventors

ZHANG BAOYI
ZHANG CHAO
YU FEI
LI CAIBIN
An Haiyue
ZHAO YINGHUI
SUN YUXIAN
JING YANFANG
WANG YINGZHE
WANG YUJUN
ZHANG WEI

Assignees

国药集团工业有限公司
国药集团工业有限公司廊坊分公司

Dates

Publication Date: 20260505
Application Date: 20260213

Claims (6)

1. A screen defect detection method for a pharmaceutical production line pharmaceutical process is characterized by comprising the following steps: Acquiring a screen surface image of a region to be detected of the screen surface, and identifying a target defect from the screen surface image; when the screen where the target defect is located is judged to be a verifiable screen, executing a screen disposal procedure; The step of judging the screen as a screen capable of being checked comprises the steps of judging the screen as the screen capable of being checked when the area of the area to be detected is larger than or equal to a preset minimum check area and the edge characteristics of the screen meet a preset flatness condition; The screen disposing process comprises the steps of classifying the checkable screen based on a preset defect standard to mark the checkable screen as a target diagnosis screen or a target replacement screen, performing screen diagnosis and upgrading on the target diagnosis screen, wherein the screen diagnosis and upgrading comprises accuracy detection on the target diagnosis screen, and re-marking the target diagnosis screen as the target replacement screen when the accuracy detection result meets a preset replacement condition; The step of classifying the screen mesh based on the preset defect standard comprises the steps of setting a first threshold value and a second threshold value which is larger than the first threshold value based on the statistical analysis of parameter values of a large number of standard defect samples and the combination of empirical data of a production process, judging that the severity of one target defect is in accordance with a slight damage standard when the parameter value of the target defect is larger than the first threshold value and smaller than or equal to the second threshold value, marking the screen mesh as a target diagnosis screen mesh, judging that the parameter value is in accordance with a serious damage standard when the parameter value is larger than the second threshold value, and directly marking the screen mesh as a target replacement screen mesh; the step of performing a screen diagnosis and upgrade on the target diagnostic screen includes obtaining a nominal accuracy and a current actual accuracy of the target diagnostic screen, and a product risk weight coefficient related to a value or quality sensitivity of a current production variety, the higher the value of the product, the greater the product risk weight coefficient value, calculating an accuracy difference between the nominal accuracy and the current actual accuracy, multiplying the accuracy difference by the product risk weight coefficient, calculating a diagnostic value, comparing the diagnostic value with a performance benchmark threshold representing a minimum acceptable performance level, and if the calculated diagnostic value is less than the performance benchmark threshold, upgrading the state of the target diagnostic screen, i.e., re-marking as a target replacement screen.
2. The method for inspecting screen defects in a pharmaceutical manufacturing line according to claim 1, wherein determining the minimum check area comprises: The method comprises the steps of carrying out edge denoising processing on screen surface images of a plurality of normal screens to obtain respective contour information, extracting an offset value of the geometric center of each contour information relative to the center of a preset fixed shaft point, determining the normal screen with the maximum offset value as a standard screen, calculating a public overlapping area defined by contour information of all the normal screens by taking the contour information of the standard screen as a standard, and determining the area of the public overlapping area as a minimum check area.
3. The method for detecting defects of a screen mesh in a pharmaceutical production line according to claim 1, wherein the preset defect criteria comprises: The method comprises the steps of constructing a plurality of defect type calibration discs, wherein each defect type calibration disc corresponds to one standard defect image, adding a preset offset to the standard defect image to generate a defect area range diagram, defining a calibration object in the defect area range diagram, acquiring a parameter value of the calibration object, and defining a slight damage standard and a serious damage standard based on a comparison relation between the parameter value and a first threshold and a second threshold.
4. A screen defect detection system for pharmaceutical production line pharmaceutical process, which uses the method for detecting screen defects in pharmaceutical production line pharmaceutical process according to any one of claims 1-3, and is characterized by comprising the following modules: The defect acquisition module is used for acquiring a screen surface image of a region to be detected of the screen surface and identifying a target defect from the screen surface image so as to generate a defect trigger signal; the screen diagnosis classification module is used for responding to the defect trigger signal and judging the screen where the target defect is located as a verifiable screen, classifying the verifiable screen based on a preset defect standard to mark the verifiable screen as a target diagnosis screen or a target replacement screen, detecting the precision of the target diagnosis screen, and re-marking the target diagnosis screen as the target replacement screen when the result of the precision detection meets a preset replacement condition; The step of judging the screen as a screen capable of being checked comprises the steps of judging the screen as the screen capable of being checked when the area of the area to be detected is larger than or equal to a preset minimum check area and the edge characteristics of the screen meet a preset flatness condition; The step of classifying the screen mesh based on the preset defect standard comprises the steps of setting a first threshold value and a second threshold value which is larger than the first threshold value based on the statistical analysis of parameter values of a large number of standard defect samples and the combination of empirical data of a production process, judging that the severity of one target defect is in accordance with a slight damage standard when the parameter value of the target defect is larger than the first threshold value and smaller than or equal to the second threshold value, marking the screen mesh as a target diagnosis screen mesh, judging that the parameter value is in accordance with a serious damage standard when the parameter value is larger than the second threshold value, and directly marking the screen mesh as a target replacement screen mesh; The step of performing screen diagnosis and upgrading on the target diagnosis screen includes obtaining rated accuracy and current actual accuracy of the target diagnosis screen, and a product risk weight coefficient related to the value or quality sensitivity of the current production variety, wherein the higher the value, the higher the product risk weight coefficient value; calculating a precision difference between the rated precision and the current actual precision, multiplying the precision difference by a product risk weight coefficient, calculating a diagnostic value, comparing the diagnostic value with a performance reference threshold representing the lowest acceptable performance level, and if the calculated diagnostic value is smaller than the performance reference threshold, upgrading the state of the target diagnostic screen, namely re-marking the target diagnostic screen as a target replacement screen; and a screen replacement module for performing screen matching and replacement on the screen marked as the target replacement screen by the screen diagnosis classification module.
5. The pharmaceutical production line pharmaceutical process screen defect detection system of claim 4, further comprising: The minimum check area determining module is used for processing the screen surface images of the normal screens to obtain respective contour information, determining a reference screen based on the contour information, calculating a common overlapping area defined by the contour information of all the normal screens, determining the area of the common overlapping area as a minimum check area, and providing the minimum check area to the screen diagnosis classifying module.
6. The pharmaceutical production line pharmaceutical process screen defect detection system of claim 4, wherein the predetermined defect criteria comprises: The method comprises the steps of constructing a plurality of defect type calibration discs, wherein each defect type calibration disc corresponds to one standard defect image, adding a preset offset to the standard defect image to generate a defect area range diagram, defining a calibration object in the defect area range diagram, acquiring a parameter value of the calibration object, and defining a slight damage standard and a serious damage standard based on a comparison relation between the parameter value and a first threshold and a second threshold.

Description

Screen defect detection method and system for pharmaceutical production line pharmaceutical process Technical Field The invention belongs to the technical field of pharmaceutical production, and particularly relates to a method and a system for detecting screen defects in a pharmaceutical production line pharmaceutical process. Background In modern pharmaceutical production lines, screening is used as a basic process of raw material pretreatment and finished product homogenization, screens used in the screening process are key equipment for guaranteeing medicine purity and uniformity of particle size distribution, the integrity of the screens is directly related to whether unqualified particles and foreign matters can be effectively removed, and quality defects of final medicines can be caused by breakage or blockage of the screens. In the prior art, the screen state detection in the pharmaceutical production line is mostly carried out by manual off-line visual inspection, the off-line detection needs to be carried out after stopping the operation of the production line or in the gap of production batch replacement, the production efficiency is reduced while the detection efficiency is low, and if the screen is damaged, real-time response cannot be carried out, so that the medicines in the batch have quality risks, in addition, due to the limitation of manual observation, the screen at the damaged edge is difficult to identify and process due to hidden defects such as micro cracks or initial blockage, and the detection omission or erroneous judgment easily occurs. Based on the technical problems, the application provides a screen defect detection method and system for a pharmaceutical production line pharmaceutical process. Disclosure of Invention In order to solve the technical problems in the prior art, the invention aims to provide a screen defect detection method and system for a pharmaceutical production line pharmaceutical process. In order to achieve the above purpose, the technical scheme adopted by the invention is as follows: a screen defect detection method for pharmaceutical production line pharmaceutical process comprises the following steps: Acquiring a screen surface image of a region to be detected of the screen surface, and identifying a target defect from the screen surface image; when the screen where the target defect is located is judged to be a verifiable screen, executing a screen disposal procedure; The screen disposing process comprises the steps of classifying the checkable screen based on a preset defect standard to mark the checkable screen as a target diagnosis screen or a target replacement screen, performing screen diagnosis and upgrading on the target diagnosis screen, wherein the screen diagnosis and upgrading comprise accuracy detection on the target diagnosis screen, re-marking the target diagnosis screen as the target replacement screen when the accuracy detection result meets a preset replacement condition, and performing screen matching and replacement on the screen marked as the target replacement screen. Preferably, the step of determining the screen as a verifiable screen comprises: And when the area of the area to be detected is larger than or equal to a preset minimum check area and the edge characteristics of the screen meet a preset flatness condition, judging the screen as a check screen. Preferably, determining the minimum check area includes: The method comprises the steps of carrying out edge denoising processing on screen surface images of a plurality of normal screens to obtain respective contour information, extracting an offset value of the geometric center of each contour information relative to the center of a preset fixed shaft point, determining the normal screen with the maximum offset value as a standard screen, calculating a public overlapping area defined by contour information of all the normal screens by taking the contour information of the standard screen as a standard, and determining the area of the public overlapping area as a minimum check area. Preferably, the preset defect criteria include: The method comprises the steps of constructing a plurality of defect type calibration discs, wherein each defect type calibration disc corresponds to one standard defect image, adding a preset offset to the standard defect image to generate a defect area range diagram, defining a calibration object in the defect area range diagram, acquiring a parameter value of the calibration object, and defining a slight damage standard and a serious damage standard based on a comparison relation between the parameter value and a first threshold and a second threshold. Preferably, in the screen diagnosis and upgrade, the step of re-marking the target diagnosis screen as the target replacement screen includes: The method comprises the steps of obtaining rated precision and current actual precision of a target diagnosis screen, calculating a diagnosis value based on