An ointment coating quality detection method, system and device

Abstract

The present invention relates to the field of detection technology, and specifically discloses a method, system and device for detecting the quality of ointment coating. The method includes: collecting image data of the coating area in real time during the ointment coating process, and reading the real-time operating parameter data of the coating device; analyzing and processing the image data based on edge detection methods and image segmentation methods to obtain multi-dimensional coating features; analyzing the defect location data of the ointment coating layer based on the multi-dimensional coating features and the real-time operating parameter data of the coating device; generating a coating quality inspection report based on the defect location data of the ointment coating layer; this method realizes the real-time, accurate detection and evaluation of the quality of ointment coating, helps to timely discover problems and take improvement measures, and improves product quality and coating process efficiency.

Description

Technical Field

[0001] The present invention relates to the field of detection technology, and particularly relates to a method, system and device for detecting the quality of ointment coating. Background Art

[0002] Currently, in the medical field, ointment coating operations are widely present. For example, during the treatment process, the ointment needs to be evenly coated on the patient's skin surface. Traditional methods for detecting the quality of ointment coating mostly rely on manual visual inspection. The inspectors directly observe the coating surface with the naked eye to judge whether the ointment coating is uniform and whether there are defects such as missed coating. With the improvement of the degree of automation, the medical industry has begun to introduce machine vision technology for detection. By collecting images of the coating area and using simple image processing algorithms to analyze the coating situation. In addition, some detection systems have begun to pay attention to the operating parameters of the coating device, such as coating speed, pressure, etc., and try to infer the coating quality by monitoring these parameters.

[0003] However, even with the introduction of machine vision technology, most of the existing methods detect after the coating is completed and cannot detect problems in real time during the coating process. Once coating quality problems occur, it will increase the rework time of the entire coating link and greatly reduce the coating efficiency. And currently, the detection methods based on machine vision are mostly limited to simple analysis of images. At the same time, the correlation analysis between the real-time operating parameters of the coating device and the coating quality is not deep enough, and the operating parameters are not fully utilized to comprehensively evaluate the coating quality. Moreover, after the existing detection technology discovers coating quality problems, it is difficult to accurately locate the defect positions. This makes it difficult to accurately perform the supplementary coating work later, affecting the overall quality of the product and the efficiency of the coating process.

[0004] Therefore, the present invention proposes a method, system and device for detecting the quality of ointment coating. Summary of the Invention

[0005] The present invention provides a method, system and device for detecting the quality of ointment coating. The method collects image data of the coating area and real-time operating parameter data of the coating device in real time during the ointment coating process, overcomes the problem of insufficient real-time detection of the prior art, and can timely detect coating quality problems during the coating process. Based on edge detection methods and image segmentation methods, the image data is analyzed and processed to obtain multi-dimensional coating features, making up for the defect of single and incomplete existing analysis methods and more comprehensively and deeply extracting the features of the coating layer. Based on the multi-dimensional coating features and the real-time operating parameter data of the coating device, the defect location data of the ointment coating layer is analyzed, and the operating parameters are fully utilized to comprehensively evaluate the coating quality, achieving accurate defect location and facilitating subsequent precise repair and adjustment. Based on the defect location data of the ointment coating layer, a coating quality detection report is generated, meeting the requirement of generating a comprehensive and detailed report, providing comprehensive and accurate information for medical staff, and helping to improve the coating quality and coating efficiency.

[0006] The present invention provides a method for detecting the quality of ointment coating, including:

[0007] S1: During the ointment coating process, image data of the coating area is collected in real time, and real-time operation parameter data of the coating device is read;

[0008] S2: The image data is analyzed and processed based on edge detection method and image segmentation method to obtain multi-dimensional coating features;

[0009] S3: Based on the multi-dimensional coating features and the real-time operation parameter data of the coating device, defect location data of the ointment coating layer is analyzed;

[0010] S4: A coating quality inspection report is generated based on the defect location data of the ointment coating layer.

[0011] Preferably, for the method for detecting the quality of ointment coating, S1: During the ointment coating process, image data of the coating area is collected in real time, and real-time operation parameter data of the coating device is read, including:

[0012] Based on a high-resolution camera, image data of the coating area is collected in real time during the ointment coating process;

[0013] The real-time ointment extrusion amount and real-time coating speed of the coating device are read as the real-time operation parameter data of the coating device.

[0014] Preferably, for the method for detecting the quality of ointment coating, S2: The image data is analyzed and processed based on edge detection method and image segmentation method to obtain multi-dimensional coating features, including:

[0015] The image data is analyzed and processed based on edge detection method and image segmentation method to obtain the boundary shape features of the target coating area and the boundary shape features of the actual coating area;

[0016] Based on the color change features of the actual coating area included in the image data, the thickness distribution features of the actual coating area are analyzed;

[0017] Among them, the multi-dimensional coating features include the boundary shape features of the target coating area, the boundary shape features of the actual coating area, and the thickness distribution features.

[0018] Preferably, for the method for detecting the quality of ointment coating, based on the color change features of the actual coating area included in the image data, the thickness distribution features of the actual coating area are analyzed, including:

[0019] Based on the boundary morphological characteristics of the actual coating area contained in the image data, the actual coating area image is divided, and the actual coating area image is converted from the RGB color model to the CIELAB color space to obtain all color components of each pixel in the actual coating area image in the CIELAB color space;

[0020] Based on all color components of each pixel in the actual coating area image in the CIELAB color space, the color change amount of each pixel in the actual coating area image in the CIELAB color space is calculated;

[0021] Based on the color change amount of each pixel in the actual coating area image in the CIELAB color space, the thickness distribution characteristics of the actual coating area are analyzed.

[0022] Preferably, for the ointment coating quality detection method, based on the color change amount of each pixel in the actual coating area image in the CIELAB color space, the thickness distribution characteristics of the actual coating area are analyzed, including:

[0023] A relationship model between the color change amount of each pixel in the actual coating area image in the CIELAB color space and the absorbance is established as , where is the color change amount of a single pixel in the CIELAB color space and is dimensionless, is a dimensionless proportionality coefficient, is the absorbance at the corresponding pixel in the actual coating area image;

[0024] Based on the color change amount of each pixel in the actual coating area image in the CIELAB color space, the relationship model, and the extinction coefficient of the ointment, the thickness value of each pixel in the actual coating area image is calculated: ,

[0025] where is the thickness value of each pixel in the actual coating area image and the unit is meter, is the extinction coefficient of the ointment and the unit is , is the amount-of-substance concentration and the unit is ;

[0026] Based on the thickness values of all pixels in the actual coating area image, the thickness distribution characteristics of the actual coating area are obtained.

[0027] Preferably, for the ointment coating quality detection method, S3: Based on the multi-dimensional coating characteristics and the real-time operating parameter data of the coating device, the defect location data of the ointment coating layer are analyzed, including:

[0028] Determine all boundary pixel coordinates in the target coating area image based on the boundary morphological features of the target coating area in the multi-dimensional coating features, and determine all boundary pixel coordinates in the actual coating area image based on the boundary morphological features of the actual coating area in the multi-dimensional coating features;

[0029] Quantify all boundary pixel coordinates in the target coating area image and all boundary pixel coordinates in the actual coating area image using Fourier descriptors to obtain all Fourier coefficients of the discrete Fourier transform of all boundary pixel coordinates in the target coating area image and all Fourier coefficients of the discrete Fourier transform of boundary pixel coordinates in the actual coating area image;

[0030] Construct a defect determination function based on all Fourier coefficients of the discrete Fourier transform of all boundary pixel coordinates in the target coating area image, all Fourier coefficients of the discrete Fourier transform of boundary pixel coordinates in the actual coating area image, and the thickness distribution feature in the multi-dimensional coating features :

[0031] ,

[0032] wherein, is the pixel coordinate in the actual coating area image, is the weight coefficient of the boundary morphological feature, is the total number of Fourier coefficients, is the th Fourier coefficient of the discrete Fourier transform of all boundary pixel coordinates in the target coating area image, is the th Fourier coefficient of the discrete Fourier transform of boundary pixel coordinates in the actual coating area image, is the weight coefficient of the thickness distribution feature, is the set of pixels in the neighborhood of the pixel with coordinate in the actual coating area image, is the th thickness value at the th pixel in the set of pixels in the neighborhood of the pixel with coordinate in the actual coating area image, is the average value of the thickness values at all pixels in the set of pixels in the neighborhood of the pixel with coordinate

[0033] Determine the defect determination value at each pixel in the actual coating area image based on the defect determination function;

[0034] Analyze the defect location data of the ointment coating layer based on the defect determination value at each pixel in the actual coating area image and the real-time operation parameter data of the coating device.

[0035] Preferably, for the ointment coating quality detection method, based on the defect determination values at each pixel in the image of the actual coating area and the real-time operation parameter data analysis of the coating device, the defect location data of the ointment coating layer is obtained, including:

[0036] Based on the real-time operation parameter data of the coating device and the defect possibility evaluation model, the defect possibility score value at each pixel in the image of the actual coating area is analyzed;

[0037] Based on the defect determination value and the defect possibility score value at each pixel in the image of the actual coating area, the final defect determination value at each pixel in the image of the actual coating area is calculated;

[0038] Based on the final defect determination values at all pixels in the image of the actual coating area, the defect location data of the ointment coating layer is obtained.

[0039] Preferably, for the ointment coating quality detection method, S4: Based on the defect location data of the ointment coating layer, a coating quality detection report is generated, including:

[0040] Based on the defect location data of the ointment coating layer, the recommended improved operation parameters of the coating device are generated;

[0041] Based on the defect location data analysis of the ointment coating layer, multiple defect parameters are obtained, and based on all defect parameters and the recommended improved operation parameters of the coating device, a coating quality detection report is generated.

[0042] The present invention provides an ointment coating quality detection system for performing any one of the above ointment coating quality detection methods, including:

[0043] An image and operation parameter acquisition module, configured to collect image data of the coating area in real time during the ointment coating process and read the real-time operation parameter data of the coating device;

[0044] A multi-dimensional coating feature acquisition module, configured to analyze and process the image data based on an edge detection method and an image segmentation method to obtain multi-dimensional coating features;

[0045] A coating defect location module, configured to analyze the defect location data of the ointment coating layer based on the multi-dimensional coating features and the real-time operation parameter data of the coating device;

[0046] A coating quality detection output module, configured to generate a coating quality detection report based on the defect location data of the ointment coating layer.

[0047] The present invention provides an ointment coating quality detection device, including:

[0048] A processor and a storage device;

[0049] The storage device is used to store instructions;

[0050] When the processor executes an instruction, any of the above ointment coating quality detection methods is implemented.

[0051] The beneficial effects of the present invention compared with the prior art are as follows: By collecting the image data of the coating area and the real-time operation parameter data of the coating device in real time during the ointment coating process, the problem of insufficient real-time detection in the prior art is overcome, and coating quality problems can be found in time during the coating process. Based on the edge detection method and the image segmentation method, the image data is analyzed and processed to obtain multi-dimensional coating features, making up for the defect of single and incomplete existing analysis methods, and extracting the coating layer features more comprehensively and deeply. Based on the multi-dimensional coating features and the real-time operation parameter data of the coating device, the defect location data of the ointment coating layer is analyzed, and the operation parameters are fully utilized to comprehensively evaluate the coating quality, realizing accurate defect location and facilitating subsequent precise repair and adjustment. Based on the defect location data of the ointment coating layer, a coating quality detection report is generated, meeting the requirement of generating a comprehensive and detailed report, providing comprehensive and accurate information for medical staff, and helping to improve the coating quality and coating efficiency.

[0052] Other features and advantages of the present invention will be described in the following specification, and, in part, will be obvious from the specification, or will be understood by implementing the present invention. The objectives and other advantages of the present invention can be achieved and obtained through the structure specifically pointed out in this application document.

[0053] The technical solutions of the present invention will be further described in detail below through the drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] The drawings are used to provide a further understanding of the present invention, and constitute a part of the specification, and are used to explain the present invention together with the embodiments of the present invention, and do not constitute a limitation to the present invention. In the drawings:

[0055] Figure 1 is the flowchart of the ointment coating quality detection method in the embodiment of the present invention;

[0056] Figure 2 is the schematic diagram of the ointment coating quality detection system in the embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0057] The following describes the preferred embodiments of the present invention with reference to the drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention. Embodiment 1

[0058] Referring to Figure 1 , the present invention provides an ointment coating quality detection method, including:

[0059] S1: During the ointment coating process, collect the image data of the coating area in real time and read the real-time operating parameter data of the coating device;

[0060] S2: Analyze and process the image data based on edge detection methods and image segmentation methods to obtain multi-dimensional coating features;

[0061] S3: Analyze the defect location data of the ointment coating layer based on the multi-dimensional coating features and the real-time operating parameter data of the coating device;

[0062] S4: Generate a coating quality inspection report based on the defect location data of the ointment coating layer.

[0063] In this embodiment, the image data of the coating area refers to the image information of the coating part collected through specific equipment (such as a high-resolution camera) during the ointment coating process, including visual features such as the shape, color, and texture of the coating area.

[0064] In this embodiment, the coating device is a device used to apply the ointment to the specified area, and it has components with functions such as controlling the extrusion amount and coating speed of the ointment.

[0065] In this embodiment, the defect location data of the ointment coating layer refers to the relevant data that determines the specific location of defects in the ointment coating layer through the analysis of coating features and the operating parameters of the coating device.

[0066] In this embodiment, the coating quality inspection report is a document generated based on the inspection and analysis results of the ointment coating quality, including the operating conditions of the coating device, defect information of the coating layer, multiple defect parameters, improvement suggestions, etc., providing a reference for evaluating the coating quality and improving the coating process.

[0067] The beneficial effects of the above technology are as follows: By collecting the image data of the coating area and the real-time operating parameter data of the coating device in real time during the ointment coating process (S1), the problem of insufficient real-time detection in the prior art is overcome, and coating quality problems can be detected in a timely manner during the coating process. Analyze and process the image data based on edge detection methods and image segmentation methods to obtain multi-dimensional coating features (S2), making up for the defect of single and incomplete existing analysis methods and extracting coating layer features more comprehensively and deeply. Analyze the defect location data of the ointment coating layer based on the multi-dimensional coating features and the real-time operating parameter data of the coating device (S3), making full use of the operating parameters to comprehensively evaluate the coating quality, achieving accurate defect location, and facilitating subsequent precise repair and adjustment. Generate a coating quality inspection report based on the defect location data of the ointment coating layer (S4), meeting the need to generate a comprehensive and detailed report, providing comprehensive and accurate information for medical staff, and helping to improve the coating quality and coating efficiency. Example 2

[0068] Based on Embodiment 1, a method for detecting the quality of ointment coating, S1: During the process of ointment coating, image data of the coating area is collected in real time, and real-time operation parameter data of the coating device is read, including:

[0069] Based on a high-resolution camera, image data of the coating area is collected in real time during the process of ointment coating;

[0070] The real-time ointment extrusion amount and real-time coating speed of the coating device are read as the real-time operation parameter data of the coating device.

[0071] In this embodiment, the real-time ointment extrusion amount and real-time coating speed of the coating device refer to the quantity of ointment extruded by the coating device at each moment and the speed of the coating operation at each moment when the ointment coating is in progress.

[0072] The beneficial effects of the above technical solutions are as follows: By using a high-resolution camera to collect image data in real time, clear and detailed images of the coating area can be obtained, improving the accuracy of detection. Reading the two key operation parameter data of the real-time ointment extrusion amount and coating speed of the coating device provides an important reference basis for accurately analyzing the coating quality. The acquisition of high-resolution image data and key operation parameters enables a more comprehensive and accurate assessment of the coating quality. It helps to promptly discover problems in the coating process, such as quality defects caused by insufficient extrusion amount or inappropriate coating speed. It can provide higher-quality and more valuable data support for the detection and analysis of the quality of ointment coating. Embodiment 3

[0073] Based on Embodiment 1, a method for detecting the quality of ointment coating, S2: Analyze and process the image data based on edge detection methods and image segmentation methods to obtain multi-dimensional coating features, including:

[0074] Analyze and process the image data based on edge detection methods and image segmentation methods to obtain the boundary shape features of the target coating area and the boundary shape features of the actual coating area;

[0075] Analyze the thickness distribution features of the actual coating area based on the color change features of the actual coating area included in the image data;

[0076] Among them, the multi-dimensional coating features include the boundary shape features of the target coating area, the boundary shape features of the actual coating area, and the thickness distribution features.

[0077] In this embodiment, analyzing and processing the image data based on the edge detection method and the image segmentation method to obtain the boundary morphological features of the target coating area and the boundary morphological features of the actual coating area means using the technical means of edge detection and image segmentation to analyze the image data of the ointment coating, so as to obtain the boundary shape, line trend and other characteristics of the area that should be coated (i.e., the target coating area) set in advance, as well as the corresponding boundary morphological characteristics of the area where the coating is actually completed (i.e., the actual coating area).

[0078] The beneficial effects of the above technical solutions are as follows: Obtaining the boundary morphological features of the target coating area and the actual coating area through the edge detection and image segmentation methods helps to accurately evaluate the accuracy and consistency of the coating. Analyzing the thickness distribution characteristics by analyzing the color change characteristics of the actual coating area can more comprehensively understand the uniformity and quality of the coating layer. Considering multiple-dimensional coating characteristics such as boundary morphological features and thickness distribution characteristics comprehensively makes the evaluation of the coating quality more comprehensive and in-depth. It helps to discover subtle defects and unevenness in the coating process and improve the control level of the ointment coating quality. It can provide multi-dimensional and detailed feature information for the accurate evaluation of the ointment coating quality. Embodiment 4

[0079] On the basis of Embodiment 3, the method for detecting the quality of ointment coating analyzes the thickness distribution characteristics of the actual coating area based on the color change characteristics of the actual coating area included in the image data, including:

[0080] Dividing the actual coating area image based on the boundary morphological features of the actual coating area included in the image data, converting the actual coating area image from the RGB color model to the CIELAB color space, and obtaining all color components of each pixel in the actual coating area image in the CIELAB color space;

[0081] Calculating the color change amount of each pixel in the actual coating area image in the CIELAB color space based on all color components of each pixel in the actual coating area image in the CIELAB color space;

[0082] Analyzing the thickness distribution characteristics of the actual coating area based on the color change amount of each pixel in the actual coating area image in the CIELAB color space.

[0083] In this embodiment, dividing the actual coating area image based on the boundary morphological features of the actual coating area included in the image data means separating this part of the area from the whole image according to the edge shape and other characteristics of the actual coating area presented in the image data, forming a specific actual coating area image.

[0084] In this embodiment, the image of the actual coating area is converted from the RGB color model to the CIELAB color space. Obtaining all the color components of each pixel in the actual coating area image in the CIELAB color space means transforming the image of the actual coating area from the common RGB color mode to the CIELAB color space, so as to obtain all the color components corresponding to each pixel point in the CIELAB color space in this area image.

[0085] First, convert the collected image from the common RGB color model to a color model more suitable for analysis, such as the CIELAB color space. Convert the collected RGB image to the CIELAB color space because the CIELAB color space is more in line with human visual perception and has more advantages in analyzing color differences. In the CIELAB color space, color is represented by three components: represents luminance, represents the color component from green to red, represents the color component from blue to yellow; for the RGB color space (value range 0 - 255), the conversion steps are as follows:

[0086] 1. First, normalize the RGB values to the range of 0 - 1, that is: , , .

[0087] 2. Perform gamma correction. If , then ; if , then , and similarly perform similar operations on and to obtain and .

[0088] 3. Convert to the XYZ color space:

[0089] , , .

[0090] 4. Finally, convert to the CIELAB color space:

[0091] ,

[0092] ,

[0093] ,

[0094] where, , , They are the tristimulus values under the standard illuminant.

[0095] In this embodiment, the color change amount of each pixel in the CIELAB color space of the actual coating area image is calculated based on all color components of each pixel in the CIELAB color space of the actual coating area image, which means relying on all color component data of each pixel in the CIELAB color space and further calculating the numerical value of the color change of each pixel in this color space. For example, the three color components of each pixel in the CIELAB color space of the actual coating area image , , After taking the square root of the sum of squares, the obtained value is regarded as the numerical value of the color change of each pixel in this color space.

[0096] The beneficial effects of the above technical solutions are as follows: Converting the actual coating area image from the RGB color model to the CIELAB color space can more accurately analyze color changes. Calculating the color change amount of each pixel in the CIELAB color space provides a quantitative data basis for analyzing the thickness distribution. Analyzing the thickness distribution characteristics based on the color change amount improves the scientificity and accuracy of the analysis. It can more precisely detect the change of the coating layer thickness and discover potential quality problems. It provides a more accurate and reliable thickness distribution characteristic analysis method for the evaluation of ointment coating quality.

[0097] Embodiment 5:

[0098] Based on the color change amount of each pixel in the CIELAB color space of the actual coating area image, on the basis of Embodiment 4, an ointment coating quality detection method analyzes the thickness distribution characteristics of the actual coating area, including:

[0099] Establish a relationship model between the color change amount of each pixel in the CIELAB color space of the actual coating area image and the absorbance as , where is the color change amount of a single pixel in the CIELAB color space and is dimensionless, is a dimensionless proportionality coefficient, is the absorbance at the corresponding pixel in the actual coating area image; based on the color change amount of each pixel in the CIELAB color space of the actual coating area image, the relationship model, and the extinction coefficient of the ointment, calculate the thickness value of each pixel in the actual coating area image: , where is the thickness value of each pixel in the actual coating area image and the unit is meter, is the extinction coefficient of the ointment and the unit is , is the amount-of-substance concentration with the unit of ; the thickness distribution characteristics of the actual coating area are obtained based on the thickness values of all pixels in the actual coating area image.

[0100] In this embodiment, the absorbance at the corresponding pixel in the actual coating area image refers to a measure of the degree of light absorption by the substance at a specific pixel position in the actual coating area image.

[0101] In this embodiment, the extinction coefficient of the ointment refers to a quantitative index of the light absorption ability of the ointment substance for light of a specific wavelength, indicating the proportion of light absorbed when passing through the ointment with a unit concentration and a unit thickness.

[0102] The beneficial effects of the above technical solutions are as follows: Establishing a relationship model between the color change amount and the absorbance provides a theoretical basis for the calculation of the thickness value. Calculating the thickness value of each pixel through the model and related parameters realizes the accurate quantification of the coating thickness. Obtaining the thickness distribution characteristics based on the thickness values of all pixels can comprehensively and accurately reflect the thickness of the actual coating area. It helps to more accurately evaluate the uniformity and quality of the ointment coating and discover local thickness abnormalities. It provides an effective method and data support for the fine analysis and evaluation of the ointment coating quality.

[0103] Embodiment 6:

[0104] Based on the embodiment 1, for the ointment coating quality detection method, S3: Analyzing the defect location data of the ointment coating layer based on the multi-dimensional coating characteristics and the real-time operation parameter data of the coating device, including:

[0105] Determining all the boundary pixel coordinates in the target coating area image based on the boundary shape characteristics of the target coating area in the multi-dimensional coating characteristics, and determining all the boundary pixel coordinates in the actual coating area image based on the boundary shape characteristics of the actual coating area in the multi-dimensional coating characteristics;

[0106] Quantifying all the boundary pixel coordinates in the target coating area image and all the boundary pixel coordinates in the actual coating area image by using the Fourier descriptor, obtaining all the Fourier coefficients of the discrete Fourier transform of all the boundary pixel coordinates in the target coating area image and all the Fourier coefficients of the discrete Fourier transform of the boundary pixel coordinates in the actual coating area image;

[0107] Constructing a defect determination function based on all the Fourier coefficients of the discrete Fourier transform of all the boundary pixel coordinates in the target coating area image and all the Fourier coefficients of the discrete Fourier transform of the boundary pixel coordinates in the actual coating area image and the thickness distribution characteristics in the multi-dimensional coating characteristics :

[0108] , where, is the pixel coordinate in the actual coating area image, is the weight coefficient of the boundary morphological feature, is the total number of Fourier coefficients, is the th Fourier coefficient of the discrete Fourier transform of all boundary pixel coordinates in the target coating area image, is the th Fourier coefficient of the discrete Fourier transform of the boundary pixel coordinates in the actual coating area image, is the weight coefficient of the thickness distribution feature, is the set of pixels in the neighborhood of the pixel with coordinates in the actual coating area image, is the th pixel in the set of pixels in the neighborhood of the pixel with coordinates in the actual coating area image, is the thickness value at the th pixel in the set of pixels in the neighborhood of the pixel with coordinates and in the actual coating area image, and the sum of

[0109] Based on the defect determination function, the defect determination value at each pixel in the actual coating area image is determined;

[0110] Based on the defect determination value at each pixel in the actual coating area image and the real-time operation parameter data of the coating device, the defect location data of the ointment coating layer is analyzed.

[0111] In this embodiment, all boundary pixel coordinates in the target coating area image are determined based on the boundary morphological feature of the target coating area in the multi-dimensional coating features, and all boundary pixel coordinates in the actual coating area image are determined based on the boundary morphological feature of the actual coating area in the multi-dimensional coating features: It means that according to the boundary shape characteristics of the target coating area and the actual coating area in the multi-dimensional coating features, the position coordinates of all pixel points forming the boundary in the images of these two areas are determined.

[0112] In this embodiment, the Fourier descriptor is used to quantify all boundary pixel coordinates in the target coating area image and all boundary pixel coordinates in the actual coating area image, and all Fourier coefficients of the discrete Fourier transform of all boundary pixel coordinates in the target coating area image and all Fourier coefficients of the discrete Fourier transform of the boundary pixel coordinates in the actual coating area image are obtained. Using the Fourier descriptor as a mathematical tool, the boundary pixel coordinates of the target coating area and the actual coating area are numerically processed, so as to obtain all Fourier coefficients corresponding to their discrete Fourier transforms.

[0113] The boundary morphology of the target coating area and the actual coating area is quantified using Fourier descriptors. For a boundary curve, its Fourier descriptors can be obtained by performing a discrete Fourier transform on the coordinates of the boundary points. Let the coordinate sequence of the boundary points be , then the discrete Fourier transform is:

[0114] , where is the th Fourier coefficient, . These coefficients form a boundary morphology feature vector used to describe the shape of the boundary. takes the value of 2.71828, and takes the value of 3.14.

[0115] In this embodiment, the weight coefficient of the boundary morphology feature is a numerical value representing the relative importance given to the boundary morphology feature during the comprehensive analysis and calculation process.

[0116] In this embodiment, the weight coefficient of the thickness distribution feature refers to the numerical value set to represent its importance during the relevant calculations and evaluations of the thickness distribution feature.

[0117] In this embodiment, the set of pixels within the neighborhood of a pixel refers to the set composed of other pixel points within a certain range around a specific pixel point.

[0118] The beneficial effects of the above technical solutions are as follows: By determining the boundary pixel coordinates of the target and actual coating areas and performing Fourier descriptor quantization, the accurate quantization and comparison of the boundary morphology are achieved. The constructed defect determination function comprehensively considers the boundary morphology, Fourier coefficients, and thickness distribution features, making the defect determination more comprehensive and accurate. By calculating the defect determination value at each pixel, the quality status of the coating area can be evaluated in detail. Combining with the real-time operation parameter data of the coating device further improves the accuracy and reliability of the defect location data. The defects of the ointment coating layer can be accurately analyzed and located, providing a targeted basis for improving the coating quality.

[0119] Example 7:

[0120] Based on the defect determination value at each pixel in the actual coating area image and the real-time operation parameter data of the coating device, on the basis of Example 6, the defect location data of the ointment coating layer is analyzed, including:

[0121] Analyzing the defect probability score value at each pixel in the actual coating area image based on the real-time operation parameter data of the coating device and the defect probability assessment model;

[0122] Calculate the final defect determination value at each pixel in the actual coating area image based on the defect determination value and the defect possibility score value at each pixel in the actual coating area image;

[0123] Obtain the defect location data of the ointment coating layer based on the final defect determination values at all pixels in the actual coating area image.

[0124] In this embodiment, the defect possibility evaluation model is a mathematical model or algorithm for evaluating the possibility of defects at each pixel in the actual coating area according to the input data (such as the real-time operating parameters of the coating device, etc.).

[0125] In this embodiment, the defect possibility score value at each pixel in the actual coating area image analyzed based on the real-time operating parameter data of the coating device and the defect possibility evaluation model refers to inputting the real-time operating parameter data of the coating device into the defect possibility evaluation model, and obtaining the quantitative score of the possibility of defects at each pixel position in the actual coating area image through the calculation of the model.

[0126] In this embodiment, calculating the final defect determination value at each pixel in the actual coating area image based on the defect determination value and the defect possibility score value at each pixel in the actual coating area image means comprehensively considering the preliminary defect determination value of each pixel and the defect possibility score value obtained through the model, and obtaining the final defect determination value of each pixel through a specific calculation method. For example, taking the average of the defect determination value and the defect possibility score value at each pixel in the actual coating area image as the final defect determination value at each pixel in the actual coating area image.

[0127] In this embodiment, obtaining the defect location data of the ointment coating layer based on the final defect determination values at all pixels in the actual coating area image is to determine the specific location and other relevant data of the defects in the ointment coating layer by summarizing the final defect determination values of all pixels in the actual coating area image.

[0128] The beneficial effects of the above technical solutions are as follows: By using the defect possibility evaluation model in combination with the real-time operating parameter data of the coating device, the defect possibility score value at each pixel is obtained, which increases the dimension and accuracy of defect evaluation. Calculating the final defect determination value through the defect determination value and the possibility score value makes the evaluation result more comprehensive and reliable. Obtaining the defect location data based on the final defect determination values of all pixels can more accurately determine the location of the defects in the ointment coating layer. It helps to detect and locate tiny defects in a timely manner and improve the precision of the quality control of the ointment coating. It further improves the detection and location method of the defects in the ointment coating layer and provides strong support for ensuring the coating quality.

[0129] Example 8:

[0130] Based on the defect location data of the ointment coating layer, the method for detecting the coating quality of the ointment, S4: generating a coating quality inspection report, including:

[0131] Generating recommended improved operating parameters for the coating device based on the defect location data of the ointment coating layer;

[0132] Analyzing multiple defect parameters based on the defect location data of the ointment coating layer, and generating a coating quality inspection report based on all defect parameters and the recommended improved operating parameters of the coating device.

[0133] In this embodiment, generating the recommended improved operating parameters for the coating device based on the defect location data of the ointment coating layer means proposing suggestions for adjusting the operating parameters of the coating device that can improve the coating effect according to relevant data such as the location of defects in the ointment coating layer.

[0134] In this embodiment, the recommended improved operating parameters of the coating device refer to the parameter values for optimization and adjustment in aspects such as coating speed and ointment extrusion amount given for the coating device in order to improve the coating quality of the ointment.

[0135] In this embodiment, analyzing multiple defect parameters based on the defect location data of the ointment coating layer means calculating and summarizing multiple quantitative indicators that can describe aspects such as the degree, type, and distribution of defects according to information such as the location of defects in the ointment coating layer.

[0136] In this embodiment, generating a coating quality inspection report based on all defect parameters and the recommended improved operating parameters of the coating device means comprehensively considering the quantitative indicators of each defect and the suggestions for adjusting the operating parameters of the coating device to form a detailed inspection report on the coating quality of the ointment.

[0137] The beneficial effects of the above technical solutions are as follows: Generating the recommended improved operating parameters for the coating device based on the defect location data can provide direct guidance for optimizing the coating process. Analyzing multiple defect parameters from the defect location data makes the report content more rich and comprehensive. Generating an inspection report based on the defect parameters and the recommended improved operating parameters provides detailed and targeted basis for adjusting the coating process and improving the quality. It helps medical staff quickly understand the key points of the coating quality problems and take effective improvement measures. It can generate a coating quality inspection report with practical value, promoting the continuous optimization of the ointment coating process and the improvement of the coating quality.

[0138] Embodiment 9:

[0139] The present invention provides an ointment coating quality detection system for performing any one of the ointment coating quality detection methods in Embodiments 1 to 8, refer to Figure 2 , including:

[0140] An image and operation parameter acquisition module, which is used to collect image data of the coating area in real time during the ointment coating process and read the real-time operation parameter data of the coating device;

[0141] A multi-dimensional coating feature acquisition module, which is used to analyze and process the image data based on edge detection methods and image segmentation methods to obtain multi-dimensional coating features;

[0142] A coating defect location module, which is used to analyze the defect location data of the ointment coating layer based on the multi-dimensional coating features and the real-time operation parameter data of the coating device;

[0143] A coating quality detection output module, which is used to generate a coating quality detection report based on the defect location data of the ointment coating layer.

[0144] The beneficial effects of the above technologies are as follows: By collecting the image data of the coating area in real time and the real-time operation parameter data of the coating device during the ointment coating process, the problem of insufficient real-time detection in the prior art is overcome, and coating quality problems can be detected in time during the coating process. Analyzing and processing the image data based on edge detection methods and image segmentation methods to obtain multi-dimensional coating features makes up for the defects of single and incomplete existing analysis methods, and extracts coating layer features more comprehensively and deeply. Analyzing the defect location data of the ointment coating layer based on the multi-dimensional coating features and the real-time operation parameter data of the coating device, making full use of the operation parameters to comprehensively evaluate the coating quality, realizing accurate defect location, and facilitating subsequent precise repair and adjustment. Generating a coating quality detection report based on the defect location data of the ointment coating layer meets the requirement of generating a comprehensive and detailed report, provides comprehensive and accurate information for medical staff, and helps improve the coating quality and coating efficiency.

[0145] Embodiment 10:

[0146] The present invention provides an ointment coating quality detection device, including:

[0147] A processor and a storage device;

[0148] The storage device is used to store instructions;

[0149] When the processor executes the instructions, any of the above ointment coating quality detection methods is implemented.

[0150] The beneficial effects of the above technical solution are as follows: Through the cooperation of the processor and the storage device, the quality inspection method of ointment coating can be efficiently executed, ensuring the timeliness and accuracy of the inspection. A series of precise and comprehensive inspection steps can be achieved to ensure the effective evaluation of the quality of ointment coating. Real-time online inspection is realized, improving the coating efficiency and the level of coating quality control. The integrated design of the device reduces the compatibility problems between devices and improves the stability and reliability of the system. It provides an efficient, stable and easy-to-apply solution for the quality inspection of ointment coating.

[0151] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention also intends to include these changes and modifications.

Claims

1. A method for detecting the quality of ointment coating, characterized in that: include: S1: During the ointment coating process, real-time image data of the coating area is collected, and real-time operating parameter data of the coating device is read; S2: Analyze and process the image data based on edge detection method and image segmentation method to obtain multi-dimensional coating features, including: Analyze and process the image data based on edge detection method and image segmentation method to obtain the boundary morphological features of the target coating area and the boundary morphological features of the actual coating area; The thickness distribution characteristics of the actual coating area are analyzed based on the color change characteristics of the actual coating area contained in the image data, including: Based on the boundary morphological features of the actual coated area contained in the image data, an actual coated area image is divided, and the actual coated area image is converted from the RGB color model to the CIELAB color space to obtain all color components of each pixel in the actual coated area image in the CIELAB color space; Calculate the color change of each pixel in the actual coated area image in the CIELAB color space based on all color components of each pixel in the actual coated area image in the CIELAB color space; Based on the color change of each pixel in the actual coating area image in the CIELAB color space, the thickness distribution characteristics of the actual coating area are analyzed, including: The relationship model between the color change and absorbance of each pixel in the CIELAB color space in the actual coating area image is established as follows: ,in, is the color change of a single pixel in the CIELAB color space and is dimensionless. is the dimensionless proportionality coefficient, is the absorbance at the corresponding pixel in the actual coating area image; Based on the color change of each pixel in the actual coating area image in the CIELAB color space, the relationship model, and the extinction coefficient of the ointment, the thickness value of each pixel in the actual coating area image is calculated: ,in, The actual coating area thickness value of each pixel in the image and the unit is meter, is the extinction coefficient of the ointment and its unit is , is the concentration of the substance and its unit is ; Obtaining the thickness distribution characteristics of the actual coating area based on the thickness values ​​of all pixels in the actual coating area image; The multi-dimensional coating features include the boundary morphological features of the target coating area and the boundary morphological features and thickness distribution features of the actual coating area; S3: Analyze the defect location data of the ointment coating layer based on the multi-dimensional coating characteristics and the real-time operating parameter data of the coating device, including: Determine all boundary pixel coordinates in the target coating area image based on the boundary morphological features of the target coating area in the multi-dimensional coating feature, and determine all boundary pixel coordinates in the actual coating area image based on the boundary morphological features of the actual coating area in the multi-dimensional coating feature; quantizing all boundary pixel coordinates in the target coating area image and all boundary pixel coordinates in the actual coating area image using Fourier descriptors to obtain all Fourier coefficients of discrete Fourier transform of all boundary pixel coordinates in the target coating area image and all Fourier coefficients of discrete Fourier transform of boundary pixel coordinates in the actual coating area image; A defect judgment function is constructed based on all Fourier coefficients of the discrete Fourier transform of all boundary pixel coordinates in the target coating area image, all Fourier coefficients of the discrete Fourier transform of the boundary pixel coordinates in the actual coating area image, and the thickness distribution characteristics in the multi-dimensional coating characteristics. : , where is the pixel coordinate in the actual coating area image, is the weight coefficient of the boundary morphological feature, is the total number of Fourier coefficients, is the discrete Fourier transform of all boundary pixel coordinates in the target coating area image. Fourier coefficients, is the discrete Fourier transform of the boundary pixel coordinates in the actual coating area image. Fourier coefficients, is the weight coefficient of thickness distribution characteristics, The coordinates of the actual coating area in the image are The set of pixels in the neighborhood of the pixel, The coordinates of the actual coating area in the image are The first pixel in the neighborhood of the pixel The thickness value at pixels, The coordinates of the actual coating area in the image are The mean thickness value of all pixels in the pixel set in the neighborhood of the pixel; Determine a defect judgment value at each pixel in the actual coating area image based on the defect judgment function; Based on the defect judgment value at each pixel in the actual coating area image and the real-time operating parameter data of the coating device, the defect location data of the ointment coating layer is analyzed, including: Analyze the defect possibility score value of each pixel in the actual coating area image based on the real-time operating parameter data of the coating device and the defect possibility assessment model; Calculating a final defect judgment value at each pixel in the actual coating area image based on the defect judgment value and the defect possibility score value at each pixel in the actual coating area image; Obtaining defect location data of the paste coating layer based on final defect determination values ​​at all pixels in the actual coating area image; S4: Generate a coating quality inspection report based on the defect location data of the ointment coating layer.

2. The ointment coating quality detection method according to claim 1, characterized in that: S1: During the ointment coating process, real-time image data of the coating area is collected and real-time operating parameter data of the coating device is read, including: Based on a high-resolution camera, real-time image data of the coating area is collected during the ointment coating process; The real-time ointment extrusion amount and the real-time application speed of the application device are read as the real-time operation parameter data of the application device.

3. The ointment coating quality detection method according to claim 1, characterized in that: S4: Generate a coating quality inspection report based on the defect location data of the ointment coating layer, including: generating suggested improved operating parameters of the coating device based on the defect location data of the paste coating layer; Based on the defect location data of the ointment coating layer, multiple defect parameters are analyzed, and a coating quality inspection report is generated based on all defect parameters and the recommended improved operating parameters of the coating device.

4. An ointment coating quality detection system, characterized in that: Used to perform any one of the ointment coating quality detection methods described in claims 1 to 3, comprising: An image and operation parameter acquisition module is used to collect image data of the coating area in real time during the ointment coating process and read real-time operation parameter data of the coating device; A multi-dimensional coating feature acquisition module is used to analyze and process image data based on edge detection methods and image segmentation methods to obtain multi-dimensional coating features; A coating defect location module is used to analyze defect location data of the ointment coating layer based on multi-dimensional coating characteristics and real-time operating parameter data of the coating device; The coating quality detection output module is used to generate a coating quality detection report based on the defect location data of the ointment coating layer.

5. An ointment coating quality detection device, characterized in that: include: Processors and storage devices; The storage device is used to store instructions; When the processor executes the instructions, the ointment coating quality detection method as described in any one of claims 1 to 3 is implemented.

Images