Medical refrigerator body manufacturing detection system and method

Abstract

The invention relates to the technical field of quality detection, in particular to a medical refrigerator body manufacturing detection system and method, and the system comprises a coating reflection detection module, a sound wave response measurement module, an optical signal fluctuation analysis module, a surface defect recognition module and a structural feature state evaluation module. According to the method, the reflection amplitude gradient change of the adjacent scanning areas is compared with the spectral curve continuity, so that the sensing precision of the surface reflection distribution continuity is improved, the recognition capability of weak anomalies is enhanced, and the internal potential defects are accurately captured in combination with the sound wave echo time and structural stability analysis of a specific part; a light intensity gradient model is constructed according to a sound wave analysis result, a splicing seam and a light intensity jump area of a sunken area are identified, abnormal forms such as scratches and bubbles are further extracted, coordinates are marked, the whole process depends on dynamic feedback and multi-signal linkage calibration, the identification precision and the coverage rate are improved, and medical box body quality global efficient detection is achieved.

Description

Technical Field

[0001] The present invention relates to the technical field of quality inspection, and in particular to a medical refrigerator box manufacturing inspection system and method. Background Art

[0002] The field of quality inspection technology includes a series of methods used to ensure that products or systems meet the specified standards and requirements. The core content of this technical field includes measuring, evaluating and verifying the performance, size, shape, quality and other attributes of objects or equipment. Quality inspection involves the application of various measuring tools and testing equipment, such as dimensional measuring instruments, surface inspection, functional testing, etc., to ensure that products in the production process can meet quality standards. Quality inspection technology is not only widely used in the manufacturing industry, but also plays a vital role in many industries such as medical, aviation, automobile, food, etc., to ensure the reliability, safety and effectiveness of each product or system.

[0003] Among them, the medical refrigerator box manufacturing inspection system refers to a system designed for quality inspection during the production process of medical refrigerator boxes. The system mainly solves the inspection problems of the size, structural integrity and other key quality indicators of medical refrigerator boxes during the production process. By combining measurement technology and automation means, the system can accurately detect various indicators of the box and ensure that it meets the medical standard requirements. The system covers technical matters such as box size measurement and material surface inspection, and by setting specific inspection standards and inspection equipment, it ensures that the manufacturing quality of the box meets the requirements of safe use and standardization.

[0004] Existing technologies mainly rely on dimensional measuring instruments and surface imaging tools for parameter detection and structural evaluation. When faced with multi-structure, multi-material complexes such as medical refrigerator bodies, it is difficult to fully reflect the stability and integrity of the internal structure. Conventional optical detection methods have insufficient recognition capabilities when faced with uneven spectral distribution or slight surface reflection differences, and are prone to image misjudgment or abnormal missed detection, affecting the accuracy and credibility of the detection results. Acoustic wave testing methods are mostly fixed-point collection, and the sound wave propagation path is significantly affected by changes in material density, and lacks a linkage comparison mechanism with other detection signals. It is difficult to accurately judge the integrity of the internal structure, and potential cracks and looseness hazards in key parts are missed. Optical recognition is often limited to static image comparison, ignoring abnormal behavior under dynamic changes in light intensity. In complex morphological structures such as joints and corner transition areas, the recognition ability is significantly reduced, affecting the detection performance of complex curved surfaces and detailed areas. For example, reflective dead spots are easily formed at corners. Traditional imaging technology is limited to single-angle shooting and cannot effectively extract complete image information, resulting in detection blind spots and the risk of missing defects. Technical limitations directly affect the precise control of key quality indicators of medical refrigerator cabinets, increasing safety hazards and performance degradation risks during subsequent use. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings in the prior art and to propose a medical refrigerator box manufacturing detection system and method.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions: Medical refrigerator box manufacturing inspection system, including: The coating reflection detection module extracts multi-band reflection spectrum data of the front door panel and side panel area based on the external data of the medical refrigerator body, compares the reflection amplitude gradient difference and the continuity of the spectrum curve change of adjacent scanning areas, determines the reflection distribution continuity of the material surface in the scanning path, and obtains the reflection continuous distribution characteristics; The acoustic wave response measurement module collects the ultrasonic emission echo time values ​​of the bottom corner reinforcement position and the side wall embedded layer of the medical refrigerator body based on the reflection continuous distribution characteristics, compares the time interval difference of the emission points with the structural echo stability, identifies the number of abnormal points in the conduction process, and obtains the acoustic wave abnormal response distribution; The optical signal fluctuation analysis module records the light intensity gradient change in the optical scanning path of the joint seam and the recessed area on the top of the medical refrigerator according to the abnormal response distribution of the acoustic wave, selects the jump point and analyzes the distribution interval span to obtain the light intensity jump distribution segment; The surface defect recognition module calls the light intensity jump distribution segment to identify the positions of scratches, bubbles, and detached areas on the panel and corners of the medical refrigerator body, determines the boundary morphological similarity of multiple defects in the scanned image, and marks the position coordinate group to obtain surface anomaly positioning data.

[0007] As a further solution of the present invention, the reflection continuous distribution characteristics include reflection consistency index, spectral change trend characteristics, and local reflection difference distribution; the acoustic wave abnormal response distribution includes emission echo time abnormal value, structural interference intensity level, and abnormal echo concentration area; the light intensity jump distribution segment includes light intensity mutation amplitude value, jump point spacing interval, and light intensity discontinuous segment; the surface abnormality positioning data includes the center coordinates of the defect area, boundary contour morphological parameters, and position distribution density.

[0008] As a further solution of the present invention, the coating reflection detection module includes: The reflection spectrum acquisition submodule collects the multi-band reflectivity of each point in the scanning path of the front door panel and the side panel based on the external data of the medical refrigerator, records the band position and the corresponding reflection value, and obtains the multi-band reflection value sequence; The amplitude gradient analysis submodule calls the multi-band reflection value sequence, identifies the band reflection amplitude difference between adjacent scanning point positions, analyzes the positive and negative changes of the difference and the amplitude fluctuation, and obtains the reflection gradient change trend; The curve continuity judgment submodule compares the smoothness and jump difference of the reflection values ​​of the continuous points according to the reflection gradient change trend, and quantifies the difference between the reflection value and the gradient value and the accumulated amount of trend change using the formula: ; Calculate the reflection consistency value between continuous segments, make judgments based on the reflection fluctuation level, and obtain the reflection continuous distribution characteristics; in, Representative The band reflectance value of each point, Representative The band reflectance value of each point, Representative The reflection gradient value corresponding to the point, Representative The reflection gradient value corresponding to the point, Indicates the reflection consistency value between consecutive segments, is the total number of points in the scan path.

[0009] As a further solution of the present invention, the acoustic wave response measurement module includes: The acoustic wave time analysis submodule collects the ultrasonic emission and echo time values ​​of the bottom corner reinforcement and the side wall embedded layer of the medical refrigerator body based on the reflection continuous distribution characteristics, records the propagation speed and path changes, and obtains the acoustic wave propagation time data; The structural echo stability detection submodule compares the acoustic wave echo time value with the original stability data and analyzes the volatility of the echo time using the formula: ; Get echo stability index; in, is the echo stability index, For the The echo time of the measurement, is the original average echo time, is the total number of measurements; The abnormal point identification submodule compares the echo stability index with a preset threshold, identifies the number of abnormal points, locates the problem area in the sound wave propagation, and obtains the abnormal response distribution of the sound wave.

[0010] As a further solution of the present invention, the optical signal fluctuation analysis module includes: The light intensity gradient acquisition submodule monitors and records the change of light intensity according to the abnormal response distribution of the acoustic wave and the optical scanning path of the joint seam and the recessed area on the top of the medical refrigerator, and obtains the light intensity gradient change data; A jumping point screening submodule screens the light intensity gradient change data, detects and marks jumping points, screens abnormal points that meet a set threshold, and obtains a jumping point data set; The light intensity jump distribution analysis submodule analyzes the distribution interval span of the jump point according to the jump point data set, identifies the associated distribution segment, and uses the formula: ; Calculate the light intensity change value to obtain the light intensity jump distribution segment; in, represents the light intensity value of the jth sampling point, Representative The light intensity value at the sampling point, represents the average light intensity, Represents the total number of sampling points, is the light intensity change value.

[0011] As a further solution of the present invention, the surface defect recognition module includes: The light intensity jump data application submodule calls the light intensity jump distribution segment, identifies the jump point data in each interval and determines the existing defect area, and screens the potential defect area by comparing the similarity between the light intensity change and the scratch, bubble, and detachment area, and obtains the defect area data set; The defect morphology recognition submodule determines the boundary morphology similarity of defects in the region according to the defect region data set, identifies the graphic features of scratches, bubbles and detached regions, and compares them with the preset defect template using the formula: ; Calculate the defect boundary similarity and obtain defect morphological feature data; in, Representative The defect boundary coordinate value of the sampling point, Representative The preset template boundary coordinate value of the sampling point, Representative The mean of the defect boundary coordinate values ​​at the sampling points, Represents the total number of sampling points, is the defect boundary similarity; The defect position marking submodule marks the position coordinates of the defect in the scanned image based on the defect morphological feature data and the defect type, identifies the defect position coordinate group, and obtains surface anomaly positioning data.

[0012] As a further solution of the present invention, the system also includes a structural feature state evaluation module: The structural feature status assessment module determines the association between the structural connection point defect and the acoustic wave interference based on the surface abnormality positioning data and the acoustic wave path change and reflection difference between the connection part of the medical refrigerator and the evaporator area, identifies the number of offset path intervals, and obtains the refrigerator body structural defect interference block; The refrigerator body structural defect interference block includes an abnormal path section of the connection part, an abnormal reflection wave offset area, and a structural response imbalance area.

[0013] As a further solution of the present invention, the structural feature state evaluation module includes: The abnormality positioning submodule extracts the coordinates of multiple points of the cold storage connection part according to the surface abnormality positioning data, selects the regional point group whose change rate exceeds the abnormality detection threshold, counts the number of abnormal points and continuous distribution trend under each unit area, and obtains the abnormality density interval value; The acoustic wave path judgment submodule calls the abnormal density interval value, compares the acoustic wave reflection data from the connection part to the evaporator area, analyzes the reflection angle deviation rate and screens the path segment exceeding the interference threshold, and combines the abnormal area difference with the reflection attenuation to obtain the reflection path deviation coefficient; The path interval identification submodule calls the reflection path deviation coefficient, identifies the segment whose deviation amplitude exceeds the stability threshold according to the path segment length and the propagation speed, extracts the number of deviation segments between the start and end nodes, and obtains the path deviation interval amount; The interference block extraction submodule calls the path offset interval quantity, compares the number of offset segments with the overlap of abnormal dense areas, screens the spatial range whose overlap exceeds the structural overlap threshold, and obtains the interference block of the refrigerator body structure defect.

[0014] The present invention also provides a medical refrigerator case manufacturing and testing method, which is performed based on the above-mentioned medical refrigerator case manufacturing and testing system, and includes the following steps: S1: Based on the external data of the medical refrigerator, the reflection spectrum data of the front door panel and the side panel area are extracted, the reflection amplitude gradient difference of adjacent scanning areas is analyzed, the continuity of the band change is judged, the change trend of adjacent areas is compared, the curve jump value is identified and the reflection gradient deviation range is determined, the discontinuous change segment is screened, and the reflection distribution continuity index set is established; S2: Based on the reflection distribution continuity index set, the ultrasonic emission echo time values ​​of the bottom corner reinforcement position and the side wall embedded layer are collected, the echo time interval difference and the amplitude stability value are compared, the abnormal point position is identified, and the sound wave abnormal response distribution area is obtained; S3: Based on the abnormal acoustic wave response distribution area, record the light intensity gradient change value in the optical scanning path of the top joint seam and the concave area, determine the position and span of the light intensity jump point, and establish a light intensity jump distribution segment sequence; S4: based on the light intensity jump distribution segment sequence, extracting the boundary morphological feature values ​​in the panel and corner area image scanning data, identifying the similarity of the defect boundary, marking the spatial coordinates, and obtaining the surface anomaly positioning data set; S5: Based on the surface anomaly positioning data set, the acoustic wave path data of the cold storage compartment connection part and the evaporator area are extracted, the defect coordinates are matched with the path change position, the number of path offset segments is counted, and the refrigerator body structure defect interference block is obtained.

[0015] Compared with the prior art, the present invention has the following beneficial effects: In the present invention, by comparing the reflection amplitude gradient change and spectral curve continuity of adjacent scanning areas, the dynamic perception ability of the continuity of the reflection distribution of the material surface is enhanced, and the recognition accuracy of weak surface anomalies is significantly improved. On the basis of the reflection continuity results, the ultrasonic emission echo time of specific parts of the structure is collected, and the time interval difference between the emission points is compared with the structural echo stability to form an accurate discrimination mechanism for the propagation consistency in the sound wave path, effectively capturing potential defect areas inside the structure. With the support of the acoustic wave response analysis results, an optical scanning light intensity gradient model of the top joint seam and the recessed area is constructed, and with the help of the jump point and distribution interval span extraction strategy, the rapid positioning of the surface light intensity drastic change area is achieved. Based on the jump light intensity segment, the morphological analysis and position marking of surface abnormal features such as scratches, bubbles, and shedding are further carried out, and the coordinate output of the defect area is realized in combination with the boundary similarity comparison algorithm. In this detection process, each step is based on the data of the previous stage, forming a dynamic feedback and linkage calibration mechanism, which enhances the detection accuracy and local anomaly recognition capability, breaks through the limitation of traditional reliance on a single signal source for judgment, and enhances the comprehensiveness and stability of complex structural defect recognition, thereby ensuring full coverage of quality indicators in all areas of the box, and establishing a highly repeatable and automated data link for box production quality control under medical standards. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 is a system flow chart of the present invention; Figure 2 This is a flow chart of the coating reflection detection module in the present invention; Figure 3 This is a flow chart of the acoustic wave response measurement module in the present invention; Figure 4 This is a flow chart of the optical signal fluctuation analysis module in the present invention; Figure 5 This is a flow chart of the surface defect recognition module in the present invention; Figure 6 This is a flow chart of the structural feature state evaluation module in the present invention. DETAILED DESCRIPTION

[0017] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0018] See also Figure 1 The present invention provides a technical solution: a medical refrigerator box manufacturing and detection system comprises: The coating reflection detection module extracts multi-band reflection spectrum data of the front door panel and side panel area based on the external data of the medical refrigerator body, compares the reflection amplitude gradient difference and the continuity of the spectrum curve change of adjacent scanning areas, determines the reflection distribution continuity of the material surface in the scanning path, and obtains the reflection continuous distribution characteristics; The spectral data of the front door panel and side panel area of ​​the medical refrigerator are analyzed, and the reflectance spectrum of the area is collected. The spectrum analyzer is used to scan light of multiple wavelengths and record the reflectivity of different wavelengths. Assuming that the scanning is performed in the infrared band, the collected wavelength is 700nm to 1000nm, and the reflectivity data is between 0.5 and 0.75. Then the reflection amplitude gradient of the adjacent scanning area is calculated, which requires the difference of the reflectivity of two consecutive scanning points to obtain the gradient difference value. For example, the reflectivity of adjacent scanning points is 0.7 and 0.72, and the gradient difference is 0.02. Then the gradient difference is compared to determine whether there is a significant reflectivity change. The criterion for significant change is that the gradient difference value exceeds 0.05. Finally, the continuity of the spectral curve is analyzed to check whether the area with a small gradient difference shows a continuous reflectivity change. The continuity is determined by setting a threshold for the gradient difference value. For example, the threshold is set to 0.03. If the gradient difference value is less than this threshold, it is considered that the reflection distribution is continuous, and the continuity characteristics of the reflection distribution of the material surface in the scanning path are obtained.

[0019] The acoustic wave response measurement module collects the ultrasonic emission echo time values ​​of the bottom corner reinforcement and the side wall embedded layer of the medical refrigerator body based on the reflection continuous distribution characteristics, compares the time interval difference of the emission points with the structural echo stability, identifies the number of abnormal points in the conduction process, and obtains the abnormal acoustic wave response distribution; The acoustic wave transmitter is installed on the bottom corner reinforcement and the side wall embedded layer of the medical refrigerator, and ultrasonic waves are emitted. The echo time is recorded by an acoustic wave detector. It is assumed that the frequency of the emitted acoustic wave is 40kHz, the emission time interval is 100 milliseconds, and the echo time recorded by the acoustic wave detector is 20 to 30 milliseconds. Then the time interval difference between different emission points is calculated. This time interval difference is obtained by comparing the echo time difference of two consecutive acoustic wave emissions. It is assumed that the two consecutive echo times are 25 milliseconds and 27 milliseconds respectively, and the time interval difference is 2 milliseconds. Then the time interval difference is compared, and a threshold is set to determine whether there is an abnormality. For example, if the time interval difference exceeds 5 milliseconds, it is marked as an abnormal point. By counting the number of all marked abnormal points, the number of abnormal points in the conduction process is identified, and the abnormal response distribution of the acoustic wave is obtained.

[0020] The optical signal fluctuation analysis module records the light intensity gradient changes in the optical scanning path of the joint seam and the recessed area on the top of the medical refrigerator according to the abnormal response distribution of the acoustic wave, selects the jump points and analyzes the distribution interval span to obtain the light intensity jump distribution segment; An optical scanner is set to scan along the joints and recessed areas on the top of the medical refrigerator, and the light intensity value of each scanning point is recorded. It is assumed that the light intensity value range during the scanning process is 200 to 300 luminous units. The gradient change of the light intensity is then calculated. The light intensity gradient is obtained by differentiating the light intensity of consecutive scanning points. For example, the light intensity values ​​of consecutive scanning points are 250 and 260, and the light intensity gradient is 10 luminous units. Then, those points where the light intensity gradient suddenly jumps are screened, and the jump threshold is set to 15 luminous units. Only when the light intensity gradient exceeds the threshold is it marked as a jump point. Finally, the distribution interval span of these jump points is analyzed. For example, if the span of consecutive jump points exceeds 5 scanning points, the recorded segment is a light intensity jump distribution segment, and the light intensity jump distribution segment is obtained.

[0021] The surface defect recognition module uses the light intensity jump distribution segment to identify the scratches, bubbles, and detached areas on the panel and corners of the medical refrigerator body, determines the boundary morphology similarity of multiple defects in the scanned image, and marks the position coordinate group to obtain surface abnormality positioning data; The data of the light intensity jump distribution segment is analyzed by image processing software to find the area with significant changes compared with the normal light intensity distribution. It is assumed that the light intensity jump area obtained by analysis is in the upper right corner of the medical refrigerator body, with the specific coordinates of (200, 150). Then, a detailed image analysis is performed on the area. By comparing the light intensity values ​​of the surrounding areas, the specific locations of scratches, bubbles or peeling are identified. It is assumed that the bubble position is (205, 155) and the scratch is (210, 160). Then, the boundary morphological similarity of these defects is determined. By calculating the boundary shape parameters of different defect areas, such as circularity, boundary roughness, etc., and comparing them with the known defect morphological parameters, the type is determined. Finally, the defect position coordinate group is marked, and the data is marked on the structural diagram of the medical refrigerator to obtain surface abnormality positioning data.

[0022] The structural feature status assessment module determines the correlation between structural connection point defects and acoustic wave interference based on the surface anomaly positioning data and the acoustic wave path changes and reflection differences between the connection part of the medical refrigerator and the evaporator area, identifies the number of offset path intervals, and obtains the refrigerator body structural defect interference block; The acoustic wave paths of the cold storage connection and the evaporator area are measured. An acoustic wave detector is used to emit acoustic waves in these areas and record the propagation time of the acoustic waves. It is assumed that the propagation time of the acoustic wave in the cold storage connection is 15 milliseconds and the propagation time in the evaporator area is 18 milliseconds. Then the time difference of the acoustic wave paths in these two areas is calculated, and the difference is 3 milliseconds. Then the time difference is evaluated and the time difference threshold is set to 5 milliseconds. If it is less than the threshold, it is judged that the structure is well connected. If it is greater than the threshold, it is marked as a possible defect point. Then the difference in acoustic wave reflection is analyzed. By comparing the acoustic wave reflection data of different areas, such as reflection intensity, reflection time, etc., it is assumed that the reflection intensity of the evaporator area is 0.8, and the reflection intensity of the cold storage connection is 0.75, and the difference is 0.05. The defects of the structural connection points are analyzed and identified, and the interference block of the refrigerator body structure defects is obtained.

[0023] The reflection continuous distribution characteristics include reflection consistency index, spectral change trend characteristics, and local reflection difference distribution. The sound wave abnormal response distribution includes the abnormal value of the emission echo time, the structural interference intensity level, and the abnormal echo concentration area. The light intensity jump distribution segment includes the light intensity mutation amplitude value, the jump point spacing interval, and the light intensity discontinuous section. The surface anomaly positioning data includes the center coordinates of the defect area, the boundary contour morphological parameters, and the position distribution density. The refrigerator body structure defect interference block includes the abnormal path segment of the connection part, the abnormal reflection wave offset area, and the structural response imbalance area.

[0024] See also Figure 2 , coating reflection detection module includes: The reflection spectrum acquisition submodule collects the multi-band reflectivity of each point in the scanning path of the front door panel and the side panel based on the external data of the medical refrigerator, records the band position and the corresponding reflection value, and obtains the multi-band reflection value sequence; Collect the reflectivity information of the front door panel and the side panel area in multiple bands to construct the reflectivity spectrum sequence of the continuous points of each area in the scanning path. When executing it specifically, first establish a point numbering system for each part of the box to mark the spatial position relationship of each measurement position. For example, the front door panel is divided into P1 to P10, and the side panel is divided into S1 to S10. Then, use a spectrometer to measure the reflectivity of each point in multiple bands such as 400nm, 500nm, and 600nm. For example, the P3 position is recorded as 0.42, 0.37, and 0.34, and the S5 position is recorded as 0.39, 0.36, and 0.32 respectively. During the collection process, it must be ensured that each point is scanned at a uniform spacing and illuminated with the same angle of light to ensure the comparability of the data. The collected data must be uniformly converted into a reflection value curve with the band as the horizontal axis and the reflectivity as the vertical axis, and stored in a structured table with an index number. The final record constitutes a multi-band reflection value sequence arranged in path order.

[0025] The amplitude gradient analysis submodule calls the multi-band reflection value sequence, identifies the band reflection amplitude difference between adjacent scanning point positions, analyzes the positive and negative changes of the difference and the amplitude fluctuation, and obtains the reflection gradient change trend; The reflectivity changes between adjacent scanning points in each band are calculated by difference. For example, if the reflectivity of point P3 is 0.44 at 500nm and that of point P4 is 0.48 in the same band, then the difference in reflection amplitude is 0.04. A reflection difference sequence can be formed by performing similar calculations on all point pairs in the entire sequence. The continuous differences in each band are processed by absolute value to obtain their gradient intensity and compare them with the standard gradient range. If the absolute value of the gradient difference is between 0.02 and 0.05, it is considered as a slow-changing interval, and if it is above 0.05, it is considered as a sudden change interval. Then, the distribution ratio of different types of areas in the overall reflection path is counted and normalized to form a reflection gradient curve in each band. The stability of the reflection difference sequence of the measuring points must also be considered during the calculation process. If the differences of three consecutive points are in the same direction and the amplitude changes are all less than 0.01, they are defined as a reflection stable segment. The final output is summarized and organized to form a reflection gradient change trend.

[0026] The curve continuity judgment submodule compares the smoothness and jump difference of the reflection values ​​of consecutive points according to the reflection gradient change trend, and quantifies the difference between the reflection value and the gradient value and the accumulated amount of trend change using the formula: ; Calculate the reflection consistency value between continuous segments, make judgments based on the reflection fluctuation level, and obtain the reflection continuous distribution characteristics; in, Representative The band reflectance value of each point, Representative The band reflectance value of each point, Representative The reflection gradient value corresponding to the point, Representative The reflection gradient value corresponding to the point, Indicates the reflection consistency value between consecutive segments, is the total number of points in the scanning path; The smoothness and jump status of the curve of the reflection value of the continuous points in each scanning segment are quantitatively judged. In the specific implementation process, the reflectivity data of 10 consecutive scanning points (recorded as P1 to P10) measured at a wavelength of 500nm are first called. , whose values ​​are directly collected by the on-site spectrometer and recorded as: , Reflectivity itself is a dimensionless quantity and does not require additional processing; Call the reflection gradient value calculated from the reflection gradient change trend in the previous step , this gradient value is the dimensionless value obtained by dividing the reflectivity difference of adjacent points by the distance between adjacent points, and is also directly recorded as: ; Perform difference calculation and gradient combination calculation for each pair of consecutive points according to the scanning path. For example, for points P1 to P2, the reflectivity difference is , corresponding to the gradient value Substituting the gradient combination operation into , the product of the two items is the local difference product 0.02×0.1281=0.00256; The difference product from point P2 to P3 is calculated to be 0.00224, and the difference product from point P9 to P10 is calculated to be 0.00188. All local difference products are added one by one to obtain the sum of the numerator terms: ; Sum the reflectivity of all points to obtain the denominator of the first part: ; Then, the absolute value of the gradient difference is accumulated to obtain the denominator of the second part: ; Combine the two denominator terms and add them together to get the total denominator: ; Substitute the numerator and denominator values ​​calculated above into the formula to obtain the final consistency value. The calculation formula is as follows: ; The consistency value It shows that the reflection change trend of the coating on the medical refrigerator body under the current scanning path is more continuous at a wavelength of 500nm, and the reflection curve jump is weak. The obtained value can directly correspond to the reflection continuous distribution characteristics of the coating surface in a specific band. Indicates The reflectance data collected at each scanning point (such as P1, P2...P10) at a specified wavelength (500nm in this case) is dimensionless and is directly collected by the spectrometer; Indicates The reflection gradient value corresponding to each point is obtained by dividing the reflectivity difference of adjacent scanning points by the spatial spacing. The value is dimensionless and is calculated as follows: , where the spatial spacing d is a fixed value, uniformly taken as 1; Indicates that the above difference product calculation and gradient difference accumulation are performed one by one for each pair of adjacent points on the continuous scanning path, where is the total number of scanned points; It is the reflection consistency value between continuous segments. The closer its value is to 0, the more stable the reflection change on the scanning path is and the better the continuity is. By combining the dual data features of reflectivity difference and gradient change to jointly judge the continuity of the coating surface spectral curve, the characterization is made more comprehensive and detailed. By normalization processing, a unified numerical continuity evaluation is achieved, which can effectively reflect the uniformity of material distribution on the surface of medical refrigerator coating during the manufacturing process.

[0027] See also Figure 3 , the acoustic response measurement module includes: The acoustic wave time analysis submodule collects the ultrasonic emission and echo time values ​​of the bottom corner reinforcement and the side wall embedded layer of the medical refrigerator body based on the reflection continuous distribution characteristics, records the propagation speed and path changes, and obtains the acoustic wave propagation time data; The ultrasonic emission echo time of the bottom corner reinforcement and the embedded layer of the side wall of the medical refrigerator is accurately collected. Multi-point collection is performed at key parts of the refrigerator through high-precision sensors to ensure that the data at each measurement point is representative. For example, two sampling points are set at the bottom and side walls of the refrigerator through sensors to record the emission and echo time values ​​of the sound wave respectively. Each set of data collected includes the start time of the sound wave emission and the time when the echo returns. By measuring the time value, the delay of the sound wave propagation in the box can be further analyzed. The sensor will record the echo time with an accuracy of milliseconds to ensure the accuracy of the time difference. For example, in an actual scenario, the emission time of acquisition point A is 2.5 milliseconds and the echo time is 3.0 milliseconds. The emission time of acquisition point B is 2.6 milliseconds and the echo time is 3.2 milliseconds. In this way, the emission and echo time values ​​of each sampling point can be obtained, and the data can be used in subsequent analysis to compare the response characteristics of different areas and obtain the sound wave propagation time data of each sampling point.

[0028] The structural echo stability detection submodule compares the acoustic echo time value with the original stability data and analyzes the volatility of the echo time using the formula: ; Get echo stability index; in, is the echo stability index, For the The echo time of the measurement, is the original average echo time, is the total number of measurements; Each echo time is compared with the average echo time of the raw data to analyze its fluctuation. The key to the process is to ensure the accuracy and consistency of the echo time data used. In actual implementation, the echo time of each sampling point is measured and a standard value is obtained based on the raw data (for example, the average echo time of the raw stability data). 3.1 milliseconds), the echo time can be obtained through the timing data of the reflected echo, and compared with the original standard to calculate the fluctuation value. Assuming that the measured current echo time is 3.0 milliseconds, 3.2 milliseconds and 3.3 milliseconds, the original echo time average is 3.1 milliseconds; in, is the current echo time value, is the original echo time mean, is the sampling times. Specifically, assuming there are 3 sampling points, the echo time values ​​are millisecond, millisecond, milliseconds, while the original mean is milliseconds, then volatility The calculation process is as follows: Take the absolute value of the difference between each echo time and the original mean: ; ; ; Compute the mean of the differences: ; Get echo stability index milliseconds, indicating the fluctuation degree of the current echo signal, calculated The unit is milliseconds (ms), which indicates the average deviation of the current echo time from the original mean. The volatility value is a measure of the echo stability. is larger, indicating that the echo signal in this area is abnormal. Milliseconds have smaller fluctuations, indicating that the signal is relatively stable. If the fluctuation exceeds the set threshold (for example, the threshold is set to 0.15 milliseconds) during the measurement process, the echo signal is considered abnormal.

[0029] The outlier identification submodule compares the echo stability index with the preset threshold, identifies the number of outliers, locates the problem area in the sound wave propagation, and obtains the distribution of abnormal sound wave responses; The echo stability index is compared with the preset threshold to identify abnormal points. In actual operation, an echo stability threshold is set, for example 0.15 milliseconds, which means that when the volatility is greater than the threshold, the echo signal is abnormal, which means that there is a potential problem with the structure of the cabinet. When the volatility value of the echo time is 0.133 milliseconds, it is lower than the threshold, indicating that the current echo signal is stable. If the volatility reaches 0.16 milliseconds, it means that the echo signal has abnormal fluctuations. The abnormal points are recorded and their positions are marked. In this way, structural damage to the refrigerator cabinet during use can be automatically identified, which helps maintenance personnel quickly locate the problem area. The distribution of abnormal points will generate an abnormal response distribution of acoustic waves, which is convenient for subsequent maintenance and repair work.

[0030] See also Figure 4 , the optical signal fluctuation analysis module includes: The light intensity gradient acquisition submodule monitors and records the changes in light intensity based on the abnormal response distribution of the acoustic wave and the optical scanning path of the joint seam and the recessed area on the top of the medical refrigerator to obtain the light intensity gradient change data; According to the optical scanning path of the joint seam and the recessed area on the top of the medical refrigerator, the light intensity data is obtained through the optical sensor. The scanning accuracy of the optical sensor should reach 0.01Lux. The light intensity data collection time interval is 1 second. Each position collects data 100 times to ensure the stability and accuracy of the data. For example, the sensor set on the top of the refrigerator will gradually obtain the light intensity value of each position according to the established scanning path. Through the acquisition frequency of the sensor, the change of light intensity can be recorded second by second. The recording of light intensity data will reflect the light intensity fluctuation between different positions, so as to obtain more accurate light intensity gradient information. For example, at a certain position, the recorded light intensity values ​​are 5Lux and 4.8Lux, indicating that the light intensity gradient has slightly fluctuated. The change process is recorded until the light intensity change is stable. The collected data will be used to calculate the change of light intensity gradient in subsequent processing, and finally generate the corresponding light intensity gradient change data.

[0031] The jump point screening submodule screens the light intensity gradient change data, detects and marks the jump points, screens the abnormal points that meet the set threshold, and obtains the jump point data set; The collected light intensity gradient change data are screened. The threshold setting method is used in the screening process to mark the fluctuation points that exceed the set threshold as jump points. When setting the threshold, it is experimentally concluded that the threshold is 0.2Lux / second. When the light intensity change rate exceeds this value, it is considered that a jump has occurred. In actual operation, if the light intensity change rate at a certain position is 0.3Lux / second, and this value is higher than the set threshold of 0.2Lux / second, this point will be marked as a jump point. During the screening process, the data must also be judged for continuity to ensure that there is no single abnormal fluctuation point that is misjudged. By setting the threshold screening method, noise fluctuations caused by external factors or equipment failures can be eliminated, and the jump point data set can be accurately obtained. The data provides a basis for subsequent light intensity jump distribution analysis.

[0032] The light intensity jump distribution analysis submodule analyzes the distribution interval span of the jump point according to the jump point data set, identifies the associated distribution segment, and uses the formula: ; Calculate the light intensity change value to obtain the light intensity jump distribution segment; in, represents the light intensity value of the jth sampling point, Representative The light intensity value at the sampling point, represents the average light intensity, Represents the total number of sampling points, is the light intensity change value; By screening and processing the jump point data, the jump point intervals are calculated and divided according to the size and distribution of the light intensity jump points. The difference between the before and after values ​​of the light intensity jump point (i.e., the rate of change of light intensity) is used as the standard for judging whether the interval belongs to the "larger span" interval. By comparing the light intensity difference, a threshold of 0.4Lux is set as the reference value of the interval span. When the light intensity change value is greater than the threshold, it is considered that there is a large jump in the interval. Based on this, the jump point data of the light intensity change will be divided into multiple intervals. For example, assuming that the jump point data collected in a certain area are 5Lux and 4.2Lux, the difference is calculated, and the light intensity jump difference is 0.8Lux. Since the difference is greater than the set 0.4Lux threshold, the span of the interval is identified as a "larger span jump segment". After multiple calculations, the intervals of all jump point distributions are finally divided; Through the above-mentioned light intensity gradient change data, the light intensity value of each sampling point is obtained, and the data is summed. Assuming that the light intensity value data of 5 sampling points is [5.0, 4.5, 4.8, 5.2, 5.1] Lux, the light intensity difference of the calculated value can be obtained according to the terms in the formula: Calculate the light intensity difference: ; Calculate the mean light intensity: ; Calculate the squared difference in light intensity: ; Calculate the final light intensity change value: ; The obtained light intensity change value is 0.1803Lux, which represents the amplitude of the light intensity change in the interval. Since the change value is less than the set threshold of 0.4Lux, the interval will not be identified as a large span jump segment. If the light intensity change value of a certain interval exceeds 0.4Lux (for example, the difference in a certain interval is 0.6Lux), then the interval will be classified as a large jump segment, which can more accurately analyze the light intensity fluctuation of the medical refrigerator, thereby effectively monitoring the performance changes of the refrigerator in actual scenarios.

[0033] See also Figure 5 , the surface defect recognition module includes: The light intensity jump data application submodule calls the light intensity jump distribution segment, identifies the jump point data in each interval and determines the existing defect area. By comparing the similarity between the light intensity change and the scratch, bubble, and detachment area, the potential defect area is screened and the defect area data set is obtained. The light intensity jump distribution segments obtained by the optical scanning device reflect the light intensity changes in different areas. The candidate positions of the surface defect areas are determined by matching the light intensity data with the various parts of the refrigerator body and combining the physical structure information of the refrigerator body. If the light intensity fluctuates greatly and the local light intensity change has an obvious "jump" in the scanning path, the position is considered to be a potential defect position. For example, when the light intensity value at one position changes more dramatically and the light intensity fluctuation in the adjacent area is small, this indicates that the area has scratches or falls off. By using the light intensity data to identify the jump segments, the potential defect area can be preliminarily located, and then the defect morphology recognition and positioning stage is entered to obtain the defect area data set.

[0034] The defect morphology recognition submodule determines the boundary morphology similarity of defects in the region based on the defect region data set, identifies the graphic features of scratches, bubbles and detached regions, and compares them with the preset defect template using the formula: ; Calculate the defect boundary similarity and obtain defect morphological feature data; in, Representative The defect boundary coordinate value of the sampling point, Representative The preset template boundary coordinate value of the sampling point, Representative The mean of the defect boundary coordinate values ​​at the sampling points, Represents the total number of sampling points, is the defect boundary similarity; The boundary morphology of the candidate defect area is identified. The boundary features of each defect area are extracted through image processing technology to obtain boundary coordinate points, and morphological matching is performed with the preset defect template (such as scratches, bubbles, and detached areas). By matching the morphological features of the boundary, it is determined whether the candidate area meets a certain type of defect. In order to achieve this process, the coordinate values ​​of each area are normalized to ensure that the scale of all defect area data is consistent, which is convenient for subsequent morphological matching calculations. For example, assuming that the boundary coordinates of a region are (5, 10, 15, 20) and the boundary coordinates of the preset template are (5, 12, 14, 18), the coordinate values ​​are normalized to convert them into a unified scale value for morphological matching. When performing morphological matching, the defect boundary similarity is obtained by calculation; Assume that the boundary coordinates of a region are (5, 10, 15, 20) and the boundary coordinates of the preset template are (5, 12, 14, 18). In order to eliminate the scale difference, the coordinates are normalized using the following formula: ; For example, for (5, 10, 15, 20), take the minimum value , the maximum value , then the normalized coordinate value is: ; Similarly, the preset template coordinates (5, 12, 14, 18) are normalized to: ; Calculate similarity: Use the normalized coordinate values ​​above to calculate the defect shape similarity using the formula: ; Determine defect type: Based on the calculated similarity Value, if If the value is greater than the set threshold (e.g. 0.7), the area is judged to be consistent with a certain defect type (e.g. scratch), otherwise it is judged to be of the same type or defect-free. Through this calculation, the defect boundary similarity is obtained. =0.697. Since this similarity is close to the set threshold (for example, 0.7), the area is a scratch. This process provides a basis for further defect location. Next, we will enter the stage of defect position marking, which can accurately identify the morphology and classify the type of the defect area, thereby ensuring that each defect is correctly handled in subsequent processing.

[0035] The defect position marking submodule marks the position coordinates of the defect in the scanned image based on the defect morphological feature data and the defect type, identifies the defect position coordinate group, and obtains the surface abnormality positioning data; Determine the type of each identified defect, then accurately calculate the position coordinates of each defect in the image and mark the corresponding coordinate points. For example, when a defect is determined to be a scratch, the defect position in the scanned image will be marked as a straight line, and the bubble will be marked as a circular area. By recording the center position and boundary points of each defect, the specific coordinates of each defect are calibrated, and basic data is provided for subsequent defect repair work. The coordinate data is finally summarized into surface anomaly positioning data, and the data is output to the visualization interface or subsequent processing module for later maintenance and quality inspection to ensure accurate repair and monitoring of defective parts in actual production.

[0036] See also Figure 6 , the structural feature status assessment module includes: The anomaly positioning submodule extracts the coordinates of multiple points of the cold storage connection according to the surface anomaly positioning data, selects the regional point groups whose change rate exceeds the anomaly detection threshold, and counts the number of anomaly points and continuous distribution trend under each unit area to obtain the anomaly density interval value; First, a detailed surface scan is performed on the cold storage compartment connection of the medical refrigerator to identify abnormal point groups whose coordinate change rate exceeds the preset threshold. The abnormal point group indicates potential structural defects or wear problems. In practical applications, such as refrigerators used in hospitals or research institutes, the connection parts are prone to rapid wear when the doors are frequently opened and closed. By analyzing the specific location and distribution of the regional point group, the number of abnormal points per unit area and the continuous distribution trend of the points can be calculated. This involves statistical analysis of the spatial distribution of the points and comparing the magnitude and frequency of changes in each point group through set operations. The calculation of the magnitude of change can be to average the coordinate change amplitude of the abnormal points in each point group, and then compare it with the structural abnormality baseline value. In this way, it can be determined which point groups have abnormal conditions beyond the normal range and obtain the abnormal density interval value. This value is obtained by calculation. For example, if the structural abnormality baseline value is set to a change amplitude of 0.5 cm, and the average change amplitude of a point group is 0.7 cm, the abnormal density interval value of the point group will be marked as a high abnormal interval, which helps subsequent maintenance and repair personnel to quickly locate and solve problems, thereby keeping the refrigerator in a safe and effective operating state.

[0037] The acoustic path judgment submodule calls the abnormal density interval value, compares the acoustic wave reflection data from the connection part to the evaporator area, analyzes the reflection angle deviation rate and screens the path segments that exceed the interference threshold, and combines the abnormal area difference with the reflection attenuation to obtain the reflection path deviation coefficient; Calling the abnormal density interval value helps to understand which areas affect the propagation of sound waves. The sound wave propagation path is very critical in practical applications, especially in medical equipment. The integrity of sound wave propagation directly affects the diagnostic accuracy of the equipment. By analyzing the sound wave reflection data from the connection to the evaporator, the incident angle and reflection angle of the sound wave can be calculated. If the reflection angle deviation is significant, that is, the reflection angle deviation rate exceeds the set sound wave reflection interference threshold, this involves numerical calculations that compare the measured sound wave angle with the theoretical angle. Such calculations help identify structural problems or sound wave propagation obstacles. For example, if the theoretical reflection angle is 30 degrees and the measured reflection angle is 35 degrees, the reflection angle deviation rate can be calculated as (35-30) / 30=16.7%. If the interference threshold is set to 10%, the sound wave path in this area will be marked as abnormal. Combined with the reflection intensity attenuation, the reflection path deviation coefficient can be obtained. This coefficient provides a quantitative basis for subsequent adjustments and optimizations, ensuring that medical equipment can operate under the accurate sound wave path and ensure the accuracy of diagnosis and research.

[0038] The path interval identification submodule calls the reflection path deviation coefficient, identifies the segments whose deviation amplitude exceeds the stability threshold according to the path segment length and propagation speed, extracts the number of deviation segments between the start and end nodes, and obtains the path deviation interval amount; The length of the acoustic path segment and the acoustic wave propagation velocity in the evaporator area will be analyzed, which is very critical in practice. For example, in the daily maintenance of medical refrigerators, ensuring the correctness of the acoustic wave path can avoid erroneous temperature readings and drug preservation problems. By identifying and classifying all acoustic wave path segments, especially those path segments with an offset amplitude exceeding the stability change threshold, the equipment can be accurately adjusted to ensure the correct propagation of the sound waves. By calculating the offset coefficient of the path segment and comparing it with the stability change threshold, the path interval that requires special attention can be identified. For example, if the acoustic wave propagation speed varies within the range of ±5% of the theoretical speed, and the offset coefficient of a certain path segment shows a deviation of 10%, this segment will be marked and the path offset interval amount will be generated, which helps technicians quickly locate the problem and make necessary adjustments.

[0039] The interference block extraction submodule calls the path offset interval quantity, compares the number of offset segments with the overlap of abnormal dense areas, and screens the spatial range where the overlap exceeds the structural overlap threshold to obtain the interference block of the refrigerator body structure defect; Using the path offset interval as a reference, by comparing the overlap between the number of offset segments and the dense area of ​​abnormal points in detail, such as in environments such as hospitals, it is ensured that each part of the refrigerator is not subject to undetected structural interference. Especially when storing sensitive materials such as vaccines and other important medicines, by screening areas where the overlap value exceeds the interference structure overlap threshold, the blocks in the refrigerator that need further inspection can be marked. For example, if the calculated overlap value of a block is 30%, and the interference overlap threshold is set to 25%, this block will be regarded as a structural defect interference block. The information is summarized and marked, and finally the refrigerator body structural defect interference block is obtained, which helps the maintenance team to carry out targeted repairs or replacements to ensure the normal operation of the refrigerator and the stability of the internal storage environment.

[0040] The medical refrigerator case manufacturing and testing method is performed based on the above-mentioned medical refrigerator case manufacturing and testing system, and includes the following steps: S1: Based on the external data of the medical refrigerator, the reflection spectrum data of the front door panel and the side panel area are extracted, the reflection amplitude gradient difference of adjacent scanning areas is analyzed, the continuity of the band change is judged, the change trend of adjacent areas is compared, the curve jump value is identified and the reflection gradient deviation range is determined, the discontinuous change segment is screened, and the reflection distribution continuity index set is established; S2: Based on the reflection distribution continuity index set, the ultrasonic emission echo time values ​​of the bottom corner reinforcement position and the side wall embedded layer are collected, the echo time interval difference and amplitude stability value are compared, the abnormal point position is identified, and the abnormal sound wave response distribution area is obtained; S3: Based on the abnormal response distribution area of ​​the acoustic wave, record the light intensity gradient change value in the optical scanning path of the top joint seam and the concave area, determine the position and span of the light intensity jump point, and establish the light intensity jump distribution segment sequence; S4: Based on the light intensity jump distribution segment sequence, the boundary morphological feature values ​​in the panel and corner area image scanning data are extracted, the similarity of the defect boundary is identified, the spatial coordinates are marked, and the surface anomaly positioning data set is obtained; S5: Based on the surface anomaly positioning data set, the acoustic wave path data of the cold storage compartment connection and the evaporator area are extracted, the defect coordinates are matched with the path change position, the number of path offset segments is counted, and the refrigerator body structure defect interference block is obtained.

[0041] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. Medical refrigerator box manufacturing detection system, characterized in that: The system comprises: The coating reflection detection module extracts multi-band reflection spectrum data of the front door panel and side panel area based on the external data of the medical refrigerator body, compares the reflection amplitude gradient difference and the continuity of the spectrum curve change of adjacent scanning areas, determines the reflection distribution continuity of the material surface in the scanning path, and obtains the reflection continuous distribution characteristics; The acoustic wave response measurement module collects the ultrasonic emission echo time values ​​of the bottom corner reinforcement position and the side wall embedded layer of the medical refrigerator body based on the reflection continuous distribution characteristics, compares the time interval difference of the emission points with the structural echo stability, identifies the number of abnormal points in the conduction process, and obtains the acoustic wave abnormal response distribution; The optical signal fluctuation analysis module records the light intensity gradient change in the optical scanning path of the joint seam and the recessed area on the top of the medical refrigerator according to the abnormal response distribution of the acoustic wave, selects the jump point and analyzes the distribution interval span to obtain the light intensity jump distribution segment; The surface defect recognition module calls the light intensity jump distribution segment to identify the positions of scratches, bubbles, and detached areas on the panel and corners of the medical refrigerator body, determines the boundary morphological similarity of multiple defects in the scanned image, and marks the position coordinate group to obtain surface anomaly positioning data.

2. The medical refrigerator box manufacturing detection system according to claim 1, characterized in that: The reflection continuous distribution characteristics include reflection consistency index, spectral change trend characteristics, and local reflection difference distribution. The acoustic wave abnormal response distribution includes emission echo time abnormal value, structural interference intensity level, and abnormal echo concentration area. The light intensity jump distribution segment includes light intensity mutation amplitude value, jump point spacing interval, and light intensity discontinuity segment. The surface abnormality positioning data includes the center coordinates of the defect area, boundary contour morphological parameters, and position distribution density.

3. The medical refrigerator box manufacturing detection system according to claim 1, characterized in that: The coating reflection detection module comprises: The reflection spectrum acquisition submodule collects the multi-band reflectivity of each point in the scanning path of the front door panel and the side panel based on the external data of the medical refrigerator, records the band position and the corresponding reflection value, and obtains the multi-band reflection value sequence; The amplitude gradient analysis submodule calls the multi-band reflection value sequence, identifies the band reflection amplitude difference between adjacent scanning point positions, analyzes the positive and negative changes of the difference and the amplitude fluctuation, and obtains the reflection gradient change trend; The curve continuity judgment submodule compares the smoothness and jump difference of the reflection values ​​of the continuous points according to the reflection gradient change trend, and quantifies the difference between the reflection value and the gradient value and the accumulated amount of trend change using the formula: ； Calculate the reflection consistency value between continuous segments, make judgments based on the reflection fluctuation level, and obtain the reflection continuous distribution characteristics; in, Representative The band reflectance value of each point, Representative The band reflectance value of each point, Representative The reflection gradient value corresponding to the point, Representative The reflection gradient value corresponding to the point, Indicates the reflection consistency value between consecutive segments, is the total number of points in the scan path.

4. The medical refrigerator box manufacturing detection system according to claim 3, characterized in that: The acoustic wave response measurement module comprises: The acoustic wave time analysis submodule collects the ultrasonic emission and echo time values ​​of the bottom corner reinforcement and the side wall embedded layer of the medical refrigerator body based on the reflection continuous distribution characteristics, records the propagation speed and path changes, and obtains the acoustic wave propagation time data; The structural echo stability detection submodule compares the acoustic echo time value with the original stability data and analyzes the volatility of the echo time using the formula: ； Get echo stability index; in, is the echo stability index, For the The echo time of the measurement, is the original average echo time, is the total number of measurements; The abnormal point identification submodule compares the echo stability index with a preset threshold, identifies the number of abnormal points, locates the problem area in the sound wave propagation, and obtains the abnormal response distribution of the sound wave.

5. The medical refrigerator box manufacturing detection system according to claim 4, characterized in that: The optical signal fluctuation analysis module comprises: The light intensity gradient acquisition submodule monitors and records the change of light intensity according to the abnormal response distribution of the acoustic wave and the optical scanning path of the joint seam and the recessed area on the top of the medical refrigerator, and obtains the light intensity gradient change data; A jumping point screening submodule screens the light intensity gradient change data, detects and marks jumping points, screens abnormal points that meet a set threshold, and obtains a jumping point data set; The light intensity jump distribution analysis submodule analyzes the distribution interval span of the jump point according to the jump point data set, identifies the associated distribution segment, and uses the formula: ； Calculate the light intensity change value to obtain the light intensity jump distribution segment; in, represents the light intensity value of the jth sampling point, Representative The light intensity value at the sampling point, represents the average light intensity, Represents the total number of sampling points, is the light intensity change value.

6. The medical refrigerator box manufacturing detection system according to claim 5, characterized in that: The surface defect recognition module comprises: The light intensity jump data application submodule calls the light intensity jump distribution segment, identifies the jump point data in each interval and determines the existing defect area, and screens the potential defect area by comparing the similarity between the light intensity change and the scratch, bubble, and detachment area, and obtains the defect area data set; The defect morphology recognition submodule determines the boundary morphology similarity of defects in the region according to the defect region data set, identifies the graphic features of scratches, bubbles and detached regions, and compares them with the preset defect template using the formula: ； Calculate the defect boundary similarity and obtain defect morphological feature data; in, Representative The defect boundary coordinate value of the sampling point, Representative The preset template boundary coordinate value of the sampling point, Representative The mean of the defect boundary coordinate values ​​at the sampling points, Represents the total number of sampling points, is the defect boundary similarity; The defect position marking submodule marks the position coordinates of the defect in the scanned image based on the defect morphological feature data and the defect type, identifies the defect position coordinate group, and obtains surface anomaly positioning data.

7. The medical refrigerator box manufacturing and detection system according to claim 1, characterized in that: The system also includes a structural feature status assessment module: The structural feature status assessment module determines the association between the structural connection point defect and the acoustic wave interference based on the surface abnormality positioning data and the acoustic wave path change and reflection difference between the connection part of the medical refrigerator and the evaporator area, identifies the number of offset path intervals, and obtains the refrigerator body structural defect interference block; The refrigerator body structural defect interference block includes an abnormal path section of the connection part, an abnormal reflection wave offset area, and a structural response imbalance area.

8. The medical refrigerator box manufacturing and detection system according to claim 7, characterized in that: The structural feature status assessment module comprises: The abnormality positioning submodule extracts the coordinates of multiple points of the cold storage connection part according to the surface abnormality positioning data, selects the regional point group whose change rate exceeds the abnormality detection threshold, counts the number of abnormal points and continuous distribution trend under each unit area, and obtains the abnormality density interval value; The acoustic wave path judgment submodule calls the abnormal density interval value, compares the acoustic wave reflection data from the connection part to the evaporator area, analyzes the reflection angle deviation rate and screens the path segment exceeding the interference threshold, and combines the abnormal area difference with the reflection attenuation to obtain the reflection path deviation coefficient; The path interval identification submodule calls the reflection path deviation coefficient, identifies the segment whose deviation amplitude exceeds the stability threshold according to the path segment length and the propagation speed, extracts the number of deviation segments between the start and end nodes, and obtains the path deviation interval amount; The interference block extraction submodule calls the path offset interval quantity, compares the number of offset segments with the overlap of abnormal dense areas, screens the spatial range whose overlap exceeds the structural overlap threshold, and obtains the interference block of the refrigerator body structure defect.

9. A medical refrigerator box manufacturing detection method, based on the medical refrigerator box manufacturing detection system according to any one of claims 1 to 8, characterized in that: The following steps are involved: S1: Based on the external data of the medical refrigerator, the reflection spectrum data of the front door panel and the side panel area are extracted, the reflection amplitude gradient difference of adjacent scanning areas is analyzed, the continuity of the band change is judged, the change trend of adjacent areas is compared, the curve jump value is identified and the reflection gradient deviation range is determined, the discontinuous change segment is screened, and the reflection distribution continuity index set is established; S2: Based on the reflection distribution continuity index set, the ultrasonic emission echo time values ​​of the bottom corner reinforcement position and the side wall embedded layer are collected, the echo time interval difference and the amplitude stability value are compared, the abnormal point position is identified, and the sound wave abnormal response distribution area is obtained; S3: Based on the abnormal acoustic wave response distribution area, record the light intensity gradient change value in the optical scanning path of the top joint seam and the concave area, determine the position and span of the light intensity jump point, and establish a light intensity jump distribution segment sequence; S4: based on the light intensity jump distribution segment sequence, extracting the boundary morphological feature values ​​in the panel and corner area image scanning data, identifying the similarity of the defect boundary, marking the spatial coordinates, and obtaining the surface anomaly positioning data set; S5: Based on the surface anomaly positioning data set, the acoustic wave path data of the cold storage compartment connection part and the evaporator area are extracted, the defect coordinates are matched with the path change position, the number of path offset segments is counted, and the refrigerator body structure defect interference block is obtained.

Images