CN-121978105-A - System for detecting and repairing surface defects of gel finished product based on image recognition

CN121978105ACN 121978105 ACN121978105 ACN 121978105ACN-121978105-A

Abstract

The invention relates to the technical field of visual inspection, in particular to an image recognition-based repair gel finished product surface defect quality inspection system which comprises a deformation induction imaging module, a characteristic analysis module, a repair matching module, a detection verification module and a repair result judging module, wherein the deformation induction imaging module is used for applying controllable micro-pressure deformation to a gel sample according to the projection area and the hardness level control pressure of the gel sample, capturing dynamic sequence images of a deformation process, detecting surface microcracks, extracting response characteristics of a defect area from the dynamic image sequence, distinguishing free impurities from fixed impurities, positioning bubble defects through a stress birefringence ring, detecting opening behaviors of the surface microcracks in the deformation process, calculating the relaxation time constant ratio and the stress concentration factor of the defect area and a normal area, evaluating the interface bonding strength, the repair matching module is used for determining repair feasibility level according to the interface bonding strength, performing self-adaption correction on technological parameters, and comparing the change amounts of the stress concentration factor before and after repair.

Inventors

LIU WENJUN
TAN QIHONG
ZHAO FENGMEI

Assignees

江苏一叶兰生物医疗科技有限公司

Dates

Publication Date: 20260505
Application Date: 20251224

Claims (8)

1. The system for detecting the surface defects of the finished gel product based on image recognition is characterized by comprising the following components: the deformation induction imaging module is used for controlling the target pressure to apply controllable micro-pressure deformation to the gel sample according to the projection area and the hardness level of the gel sample, so that the gel is subjected to thickness compression deformation, capturing a dynamic sequence image of the deformation process, and detecting surface microcracks; the characteristic analysis module extracts response characteristics of a defect area from a dynamic image sequence, distinguishes free impurities from fixed impurities by tracking relative displacement tracks of the impurities, positions bubble defects by identifying stress birefringence rings, and detects opening behaviors of surface microcracks in the deformation process; The repair matching module is used for determining a repair feasibility grade according to the interface bonding strength, matching a repair scheme and carrying out self-adaptive correction on the technological parameters according to the interface bonding strength; And the detection and verification module repeatedly executes deformation induction detection on the repaired product to obtain the repaired characteristic parameters, and compares the variation of the stress concentration factors before and after repair to judge the repair result.
2. The system for detecting surface defects of the finished gel product based on image recognition and repair according to claim 1, wherein the detection process of the surface microcracks comprises the following steps: Collecting holographic interference fringe patterns at a plurality of moments in the deformation process, converting the interference fringe patterns into surface phase distribution patterns through a phase reconstruction algorithm, calculating phase differences of the phase distribution patterns before and after deformation, extracting phase mutation areas, carrying out connectivity analysis and geometric feature extraction on the phase mutation areas, judging the phase mutation areas with lengths exceeding a preset threshold and in linear and/or curve forms as surface microcracks, and recording position coordinates, lengths and direction angles of the microcracks.
3. The system for inspecting surface defects of a finished gel product based on image recognition and repair according to claim 1, wherein the process for recognizing the bubble defects positioned by the stress birefringence ring comprises the following steps: The method comprises the steps of continuously collecting a transmitted light image sequence, calculating a time sequence standard deviation of gray values of each pixel point in the image sequence to generate a gray standard differential layout, extracting the pixel points with standard deviation exceeding a preset threshold value from the gray standard differential layout as a change significant region, carrying out connected domain analysis and boundary extraction on the change significant region to calculate geometric feature parameters of a boundary, identifying a region with the circularity feature value meeting a ring judgment condition of the boundary as a feature ring formed by a stress effect, obtaining a center coordinate and a radius of the feature ring, estimating the actual radius of bubbles according to the radius of the feature ring, an applied pressure value and a photoelastic coefficient of a gel material, and determining the position and the size of bubble defects.
4. The system for inspecting surface defects of a finished gel product based on image recognition as set forth in claim 1, wherein the step of detecting the opening behavior of the surface microcracks comprises the steps of: The method comprises the steps of obtaining phase data of a deformation process, extracting phase change curves of an identified surface microcrack area in a pressurizing stage, a pressure maintaining stage and a pressure releasing stage, calculating phase gradients at two sides of the microcrack, calculating crack opening displacement according to the change of the phase gradients, analyzing the change of the crack opening displacement along with the applied pressure to obtain the crack opening displacement, judging the depth of the crack and the stress state of the crack tip according to the opening displacement of the crack under different pressures, and judging the crack with the opening displacement exceeding a preset threshold value as dangerous crack.
5. The system for inspecting surface defects of a finished gel product based on image recognition as set forth in claim 1, wherein the process of evaluating the interfacial bonding strength of the defect and the gel matrix comprises: Establishing an interface bonding strength evaluation matrix, wherein the interface bonding strength evaluation matrix divides a relaxation time constant ratio and a stress concentration factor into a plurality of level intervals, and divides the interface bonding strength into three levels of strong bonding, medium bonding and weak bonding for the defect that the absolute value of the difference value of the relaxation time constant ratio and the unit value is smaller than a first threshold value, and determines weak interface bonding for the defect that the relaxation time constant ratio and the unit value are larger than a second threshold value, and comprehensively determines the level of the interface bonding strength by combining the magnitude of the stress concentration factor.
6. The system for inspecting surface defects of a finished product of a repair gel based on image recognition as set forth in claim 1, wherein the process of determining the repair feasibility level comprises: The method comprises the steps of establishing a repair feasibility judging rule base according to interface bonding strength grades and stress concentration factors, judging good quality repairable grades for defects of weak bonding grades, judging general repairable grades for defects of medium bonding grades if the stress concentration factors are lower than a first factor threshold, judging unrepairable grades for defects of strong bonding grades and/or defects of stress concentration factors exceeding a second factor threshold, comprehensively considering the proportion of defect sizes to product sizes, judging the product directly as a qualified product when the proportion is lower than a preset proportion threshold and the stress concentration factors are lower than the first factor threshold, and outputting the repair feasibility grades.
7. The system for inspecting surface defects of a finished gel product based on image recognition as set forth in claim 1, wherein the process for adaptively correcting the process parameters comprises: the method comprises the steps of obtaining an initial repairing scheme according to a gel type, a defect type and a repairing feasibility grade, obtaining an initial technological parameter combination, adjusting a repairing intensity coefficient according to an interface bonding intensity grade, calculating a depth correction coefficient according to a ratio relation between defect depth and total gel thickness, correcting the initial technological parameter by utilizing the repairing intensity coefficient and the depth correction coefficient, including adjusting at least one parameter of heating temperature, heating time, vibration frequency, vibration acceleration and ultrasonic power, generating a corrected technological parameter combination and outputting the corrected technological parameter combination as an executable repairing instruction.
8. The system for detecting surface defects of a finished product of the repair gel based on image recognition as set forth in claim 1, wherein the process for judging the repair result comprises the following steps: The method comprises the steps of executing deformation induction detection flow which is the same as that before repair on a repaired product, obtaining a relaxation time constant ratio and a stress concentration factor after repair, calculating the change quantity and the change rate of the stress concentration factor before and after repair, judging that the repair is successful when the reduction amplitude of the stress concentration factor exceeds a preset success threshold and the stress concentration factor after repair is lower than a safety threshold, judging that the repair is failed when the reduction amplitude of the stress concentration factor is insufficient and/or the value after repair still exceeds the safety threshold, and recording the results of the repair success and the repair failure, corresponding defect characteristic parameters and process parameters into a database for optimizing and updating a repair strategy knowledge base.

Description

System for detecting and repairing surface defects of gel finished product based on image recognition Technical Field The invention relates to the technical field of visual inspection, in particular to a system for inspecting quality of repairing surface defects of gel finished products based on image recognition. Background The gel products are widely applied to the fields of cosmetics, medical care, industrial materials and the like. In the production process of the gel finished product, due to the factors of uneven mixing of raw materials, fluctuation of filling process parameters, insufficient cleanliness of packaging materials and the like, the appearance of the product is influenced, and the performance and the use safety of the product are possibly influenced. With the development of machine vision technology, some enterprises begin to adopt an automatic quality inspection system based on image recognition, however, the existing image recognition quality inspection system faces special challenges when being applied to gel products, wherein gel materials generally have transparent or semitransparent characteristics, internal bubbles are easily mixed with surface reflection, impurities are not obvious with background contrast, defect recognition accuracy is not high, in addition, the existing system generally only can detect and mark defects, and lacks intelligent analysis on defect types, positions and severity, whether the products can be repaired by reworking or should be scrapped cannot be automatically judged, and secondary judgment still needs to be manually carried out, so that the overall quality inspection efficiency is influenced. Therefore, a quality inspection system which is optimized for the characteristics of the gel finished product, can accurately identify the surface defects and intelligently judge the processing scheme is needed, so that the quality inspection efficiency and accuracy are improved, and the labor cost is reduced. Therefore, a system for repairing the surface defects of the gel finished product based on image recognition is provided. Disclosure of Invention The invention aims to provide a system for detecting and repairing the surface defects of a gel finished product based on image recognition, which realizes automatic detection of defects and self-adaptive adjustment of a repairing process by constructing a gel surface defect quality detecting and repairing system based on image recognition. In order to achieve the above purpose, the present invention provides the following technical solutions: A system for repairing surface defects of a gel finished product based on image recognition, comprising: the deformation induction imaging module is used for controlling the target pressure to apply controllable micro-pressure deformation to the gel sample according to the projection area and the hardness level of the gel sample, so that the gel is subjected to thickness compression deformation, capturing a dynamic sequence image of the deformation process, and detecting surface microcracks; The characteristic analysis module extracts response characteristics of the defect area from the dynamic image sequence, distinguishes free impurities from fixed impurities by tracking relative displacement tracks of the impurities, positions bubble defects by identifying stress birefringence rings, and detects opening behaviors of surface microcracks in the deformation process; The repair matching module is used for determining a repair feasibility grade according to the interface bonding strength, matching a repair scheme and carrying out self-adaptive correction on the technological parameters according to the interface bonding strength; And the detection and verification module repeatedly executes deformation induction detection on the repaired product to obtain the repaired characteristic parameters, and compares the variation of the stress concentration factors before and after repair to judge the repair result. Preferably, the detection process of the surface microcracks comprises the following steps: Collecting holographic interference fringe patterns at a plurality of moments in the deformation process, converting the interference fringe patterns into surface phase distribution patterns through a phase reconstruction algorithm, calculating phase differences of the phase distribution patterns before and after deformation, extracting phase mutation areas, carrying out connectivity analysis and geometric feature extraction on the phase mutation areas, judging the phase mutation areas with lengths exceeding a preset threshold and in linear and/or curve forms as surface microcracks, and recording position coordinates, lengths and direction angles of the microcracks. Preferably, the process of identifying the stress birefringence loop to locate bubble defects includes: The method comprises the steps of continuously collecting a transmitted light image sequence, calculating a time sequenc