Polishing quality detection method for processing LED braider track

Abstract

The invention relates to the technical field of image data processing, in particular to a grinding quality detection method for processing an LED braider track, which comprises the following steps of: acquiring an LED braider track image to be detected, classifying the image through an improved support vector machine model to obtain a corresponding label, and judging the grinding quality according to the label; wherein category weights are introduced into the improved support vector machine model, the category weights are calculated according to the content richness of each to-be-detected image, and the content richness is measured through the difference between the image gray gradient and the color channel information; specifically, the gray gradient variance and the color channel gray mean value difference of the image are used for calculating the content richness of each image, and then the category weight of each image is influenced. The problem that the accuracy is not high in the grinding quality detection process of the LED braider track is solved.

Description

Technical Field

[0001] The present invention relates to the technical field of image data processing, and more particularly to a method for detecting the polishing quality of a track of an LED taping machine. Background Art

[0002] With the rapid development of the LED industry, LED taping machines are playing an increasingly important role in the production process. As one of its key components, the track of the LED taping machine is responsible for transporting and securing the LED light strips. Therefore, the processing quality of the track directly affects the efficiency of the entire production line and the product qualification rate. To ensure high-precision processing of the LED taping machine track, track surface quality inspection has become a critical step. However, during the track polishing process, due to equipment accuracy limitations, differences in worker operation, and different materials, uneven wear, scratches, or other defects may appear on the track surface. These defects are often difficult to efficiently and accurately identify through traditional manual inspection methods.

[0003] Therefore, automated and precise detection of the polishing quality of LED taping machine tracks has become a technical problem that needs to be solved urgently in the current production process. By adopting advanced image processing, machine vision and other technologies, various defects on the track surface can be identified quickly and accurately, thereby improving production efficiency and product quality and reducing the occurrence of unqualified products. Support vector machine is a supervised learning method widely used in image classification, pattern recognition and other fields. In the image binary classification task, support vector machine distinguishes two types of samples by finding the optimal hyperplane. Its classification effect depends on factors such as the feature representation of the data, the choice of kernel function and the imbalance of categories. The introduction of category weights is an important means to solve the problem of category imbalance. It can optimize the performance of the classifier by adjusting the misclassification cost of different categories during the training process.

[0004] Existing technology can effectively alleviate the problem of category imbalance by adjusting category weights, but this method has some shortcomings. How to select appropriate category weights often depends on manual experience or parameter adjustment, and the setting of different weights may lead to overfitting or underfitting, which may lead to the problem of low accuracy of classification results in practical applications, and further lead to low accuracy in the process of polishing quality inspection of LED taping machine tracks. Summary of the Invention

[0005] In order to solve the problems of low accuracy and insufficient reliability of detection results raised in the above background technology, the present invention provides the following solutions.

[0006] The present invention provides a method for detecting the polishing quality of the track of an LED taping machine, comprising: collecting images to be detected of each track of the LED taping machine; classifying the images to be detected using an improved support vector machine model to obtain labels, and judging the polishing quality of the track of the LED taping machine based on the labels; wherein the improved support vector machine model includes category weights, the first The category weight of the image to be detected is the same as the The content richness of the image to be detected is positively correlated with the content richness of the first image to be detected, and is negatively correlated with the sum of the content richness of each image to be detected; All pixels in the image to be detected are The difference between the gray value and the gray mean under the channel and the variance of the gray gradient are positively correlated. All pixels in the image to be detected are The grayscale value under the channel is inversely correlated with the grayscale mean difference.

[0007] The above technical solution can intelligently adjust the focus on different images during the classification process, enhance the sensitivity to highly complex images and subtle defects, and reduce interference in low-information areas. Through the improved support vector machine model and combined with the dynamic weight distribution of image content richness, the detection accuracy of the LED taping machine track polishing quality is improved.

[0008] Further, the The category weight of the image to be detected for, , where For the The content richness of the image to be detected, is the total number of images to be detected, is an empirical constant.

[0009] Further, the Content richness of the image to be detected for, , where For the The variance of the grayscale gradient of the image to be detected, For the The first image in the image to be detected Pixels in Grayscale value under the channel, For the All pixels in the image to be detected are The grayscale mean under the channel, For the The first image in the image to be detected Pixels in Grayscale value under the channel, For the All pixels in the image to be detected are The grayscale mean under the channel, For the The total number of pixels in the image to be detected.

[0010] The above technical solution significantly improves the ability to quantify image complexity by introducing a content richness calculation method based on the grayscale gradient variance and the grayscale mean difference between color channels. By combining the grayscale gradient and the grayscale difference between color channels, it is possible to more accurately measure the details and potential defects in the image. In particular, when processing images with complex reflective characteristics such as metal surfaces, it can effectively capture subtle surface changes. Not only does it enhance the response to highly complex images, it also effectively increases the sensitivity to tiny defects in the image, which can help the support vector machine model more accurately adjust the weights during the classification process, thereby improving the accuracy of judging the polishing quality.

[0011] Further, the Content richness of the image to be detected for, , where For the The variance of the grayscale gradient of the image to be detected, For the The first image in the image to be detected Pixels in Grayscale value under the channel, For the All pixels in the image to be detected are The grayscale mean under the channel, For the The first image in the image to be detected Pixels in Grayscale value under the channel, For the All pixels in the image to be detected are The grayscale mean under the channel, For the The total number of pixels in the image to be detected, For the natural constant The exponential function with base , For the natural constant Logarithmic function with base .

[0012] The above technical solution further optimizes the measurement method of image content richness by introducing logarithmic and exponential operations, combining the grayscale gradient variance with the grayscale difference between color channels. By accurately quantifying the image details, subtle changes in the image can be more effectively captured, especially in complex images, such as defects or surface irregularities on metal surfaces. This calculation method enables more sensitive identification of high-complexity areas and potential quality problems, thereby improving the accuracy and robustness of defect detection. Especially when processing subtle differences in images, nonlinear operations can enhance sensitivity to details and avoid interference with low-information areas, enabling the model to better cope with changing detection scenarios in actual applications, improving the quality control capabilities of the production line, and ensuring more stable and accurate detection results.

[0013] Furthermore, the tags are divided into lossy and lossless.

[0014] Furthermore, when the tag is damaged, an early warning is issued.

[0015] Furthermore, a CCD camera or a CMOS camera is used to collect the image to be inspected of each LED taping machine track.

[0016] Furthermore, the method further includes performing grayscale processing on the image to be detected.

[0017] By introducing the grayscale processing step, the above technical solution can effectively simplify the complexity of the image, remove the interference of color information, and enable the model to focus more on the structure and texture features of the image. By converting the image to a grayscale image, the amount of calculation can be reduced and the processing speed can be improved. At the same time, without losing key information, the sensitivity to grayscale changes and edge details can be enhanced. This is particularly important for defect detection, because many defects are usually manifested as sudden changes or changes in grayscale values in grayscale images, and the grayscaled image can more clearly highlight these subtle differences, thereby improving the accuracy and robustness of detection. In particular, when dealing with metal rails with high reflectivity or complex surface textures, potential problems can be identified more accurately, optimizing overall quality control and detection efficiency.

[0018] Furthermore, the early warning is a reminder via a buzzer.

[0019] The aforementioned technical solution, through the use of a buzzer for early warning, provides immediate and intuitive feedback, ensuring that operators are aware of any abnormalities or defects in polishing quality as quickly as possible. This audible alarm effectively improves the production line's response speed, avoiding product quality issues or production delays caused by undetected defects. Furthermore, the buzzer's high audibility in noisy production environments ensures that operators clearly receive warnings even in complex or busy work environments, thereby reducing the risk of human error and improving overall production efficiency and safety.

[0020] Furthermore, the CCD camera or CMOS camera is installed above the track of each LED taping machine.

[0021] The beneficial effects of the present invention are: The present invention effectively improves the detection accuracy of the track polishing quality of LED taping machines by combining an improved support vector machine model with content richness analysis. By classifying images and assigning weights to different categories, tracks of different qualities can be distinguished more accurately, especially when damage is detected, early warnings can be issued in a timely manner to prevent production defects caused by quality problems. In addition, CCD or CMOS cameras are used for image acquisition, and grayscale processing is used to improve the stability and consistency of image processing. This method can monitor the track polishing quality in real time during the production process, ensuring production efficiency and product quality. At the same time, a buzzer is used to provide reminders, which helps to quickly respond and improve safety in the working environment, reducing the risk of human negligence. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 The flowchart schematically shows a method for detecting the polishing quality of a track of an LED taping machine according to an embodiment of the present invention. DETAILED DESCRIPTION

[0023] An embodiment of a method for detecting the polishing quality of a track for processing an LED taping machine.

[0024] like Figure 1 As shown in FIG. 1 , a flow chart of a method for detecting the polishing quality of a track of an LED taping machine according to an embodiment of the present invention includes the following steps: S1: Collect the images to be inspected of each LED taping machine track.

[0025] In one embodiment, to achieve efficient detection and monitoring of the LED taping machine track, a high-resolution CCD camera or CMOS camera can be used for image acquisition. These cameras are precisely mounted above the LED taping machine track to ensure that the entire image to be detected is captured from the optimal angle. The advantage of using a CCD camera or CMOS camera is its excellent image clarity and imaging quality. CCD and CMOS sensors have high photoelectric conversion efficiency and can capture extremely fine image details, making them particularly suitable for use in complex industrial environments. This high-quality image can effectively avoid false detection and missed detection problems caused by image blur or excessive noise.

[0026] The captured image to be inspected is then grayscaled. As a fundamental operation in image preprocessing, grayscale processing eliminates redundant information in color images, thereby reducing computational complexity. This significantly improves processing efficiency, particularly when processing large amounts of image data. Converting color images to grayscale not only reduces data storage requirements but also makes subsequent operations such as image feature extraction, edge detection, and defect analysis simpler and more accurate.

[0027] S2: Using the improved support vector machine model to classify the image to be detected to obtain a label.

[0028] In one embodiment, the improved support vector machine model includes class weights, The category weight of the image to be detected for, , where For the The content richness of the image to be detected, is the total number of images to be detected; by introducing category weights to improve the support vector machine model, the weight of each image to be detected can be dynamically adjusted based on its content richness, allowing the model to pay more attention to images with richer information during training. This method can effectively improve the model's accuracy and robustness when processing complex image data, especially in image sets with large differences or noise. By assigning higher weights to images with richer content, the impact of low-quality data on model performance can be reduced, thereby enhancing the overall classification effect and improving the accuracy and reliability of detection.

[0029] No. Content richness of the image to be detected for, , where For the The variance of the grayscale gradient of the image to be detected, For the The first image in the image to be detected Pixels in Grayscale value under the channel, For the All pixels in the image to be detected are The grayscale mean under the channel, For the The first image in the image to be detected Pixels in Grayscale value under the channel, For the All pixels in the image to be detected are The grayscale mean under the channel, For the The total number of pixels in the image to be detected.

[0030] It's important to note that rails are typically made of metal, and metal surfaces have unique light reflection properties. In the visible spectrum, red light (R channel) has a longer wavelength than blue light (B channel), making it more easily reflected by metal surfaces. When defects such as cracks, scratches, or pits exist on the rail surface, these defects alter the metal's reflective properties, causing red light to reflect differently from normal areas, making the defect more visible in the R channel. Furthermore, the morphological characteristics of defects, such as cracks and scratches, often cause subtle deformations or damage to the metal surface, which can affect the scattering and reflection direction of light. Due to its longer wavelength, red light is more susceptible to these subtle changes, resulting in a more pronounced contrast in the R channel. In contrast, blue light, with its shorter wavelength, is less sensitive to subtle deformations or damage, making defects less noticeable in the B channel. Therefore, when there is a significant difference between the images in the R and B channels, it can be inferred that there is a high probability of a defect.

[0031] Furthermore, the presence of defects increases the number of edge pixels in an image, and edge pixels typically have larger grayscale gradients. Therefore, a greater variance in the gradients of pixels in an image indicates a greater number of edge pixels, richer edge information, and more complex overall image content. This characteristic further demonstrates that calculating the variance of grayscale gradients in an image can effectively reflect the richness of image content and help identify and locate defective areas within the image.

[0032] By incorporating the differences between the image's grayscale gradient variance and the color channel grayscale mean, the content richness of each image to be detected is calculated, quantifying the image's complexity. This approach more accurately measures the differences in the distribution of spatial and color information across images, thereby assigning greater weight to images with richer content. This processing not only increases the model's focus on highly complex images while avoiding excessive influence on low-information images, but also effectively improves the accuracy of image classification and anomaly detection, particularly in complex scenarios, helping to optimize the model's robustness and adaptability.

[0033] In another embodiment, the Content richness of the image to be detected for, , where For the The variance of the grayscale gradient of the image to be detected, For the The first image in the image to be detected Pixels in Grayscale value under the channel, For the All pixels in the image to be detected are The grayscale mean under the channel, For the The first image in the image to be detected Pixels in Grayscale value under the channel, For the All pixels in the image to be detected are The grayscale mean under the channel, For the The total number of pixels in the image to be detected, For the natural constant The exponential function with base , For the natural constant Logarithmic function with base .

[0034] By combining the grayscale gradient variance and the exponential function of the grayscale value difference of the color channel, the image's content richness can be further accurately measured. By introducing logarithmic and exponential operations, the image's gradient changes and the reflectance differences of the color channels can be effectively balanced, so that areas of the image with significant changes receive higher weight in the calculation. This approach not only improves sensitivity to subtle defects, but also adjusts the response to different types of defects through nonlinear functions, making the model more stable and accurate when facing complex image data, thereby improving the accuracy and robustness of defect detection and image classification. In particular, it can more effectively avoid misjudgments and missed detections when processing complex or noisy images.

[0035] The labels are classified into lossy and lossless; S3: Determine the polishing quality of the LED taping machine track based on the label.

[0036] In one embodiment, if the polishing quality label on the track of an LED taping machine is detected as damaged, an immediate warning is issued to ensure that necessary adjustments are taken in a timely manner during production. This warning, delivered via a buzzer, quickly and effectively draws the operator's attention, ensuring the issue is addressed as quickly as possible. As an intuitive and easily recognizable alarm method, the buzzer not only quickly conveys information in noisy industrial environments, but also indicates the urgency of the issue through changes in the audio frequency, helping operators to respond appropriately.

[0037] Timely early warnings significantly reduce the risk of production failures and substandard product quality due to polishing quality issues, avoiding further losses caused by undetected defects. Furthermore, this early warning mechanism enhances the intelligence of the production line, enabling equipment to automatically respond when problems arise, reducing reliance on manual inspection and improving reliability.

[0038] The solution of the present invention can accurately detect the polishing quality of the LED taping machine track through an improved support vector machine model combined with the evaluation of image content richness. By classifying and distinguishing the collected images, defects or undesirable phenomena in the polishing process can be accurately identified, thereby improving the accuracy and efficiency of detection. By introducing the analysis of grayscale value differences and gradient variance in grayscale processing and image classification, the sensitivity to different types of defects is further enhanced, so that the detection process can effectively distinguish between normal and bad areas. In addition, combined with the early warning mechanism, timely reminders can be issued when quality problems are discovered, thereby improving the response speed of the production line and the real-time monitoring capabilities of operators, thereby effectively avoiding potential production quality risks and ensuring the stability of the polishing process and the consistency of the product.

[0039] In the description of this specification, "multiple" and "several" mean at least two, such as two, three or more, unless otherwise clearly defined.

[0040] While several embodiments of the present invention have been shown and described herein, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Numerous modifications, variations, and alternatives will occur to those skilled in the art without departing from the concept and spirit of the present invention. It should be understood that various alternatives to the embodiments of the present invention described herein may be employed in practicing the present invention.

Claims

1. A method for detecting the polishing quality of a track of an LED taping machine, characterized in that: include: Collect images of the tracks of each LED taping machine to be tested; classify the images to be tested using an improved support vector machine model to obtain labels, and judge the polishing quality of the tracks of the LED taping machine based on the labels; Among them, the improved support vector machine model includes category weights, The category weight of the image to be detected is the same as the The content richness of each image to be detected is positively correlated, and negatively correlated with the sum of the content richness of all images to be detected. The content richness is the same as All pixels in the image to be detected are The difference between the gray value and the gray mean under the channel and the variance of the gray gradient are positively correlated. All pixels in the image to be detected are The grayscale value under the channel is inversely correlated with the grayscale mean difference.

2. A method for detecting the polishing quality of a track for processing an LED taping machine according to claim 1, characterized in that: No. The category weight of the image to be detected for, , where For the The content richness of the image to be detected, is the total number of images to be detected, is an empirical constant.

3. A method for detecting the polishing quality of a track for processing an LED taping machine according to claim 1, characterized in that: No. Content richness of the image to be detected for, , where For the The variance of the grayscale gradient of the image to be detected, For the The first image in the image to be detected Pixels in Grayscale value under the channel, For the All pixels in the image to be detected are The grayscale mean under the channel, For the The first image in the image to be detected Pixels in Grayscale value under the channel, For the All pixels in the image to be detected are The grayscale mean under the channel, For the The total number of pixels in the image to be detected.

4. A method for detecting the polishing quality of a track for processing an LED taping machine according to claim 1, characterized in that: No. Content richness of the image to be detected for, , where For the The variance of the grayscale gradient of the image to be detected, For the The first image in the image to be detected Pixels in Grayscale value under the channel, For the All pixels in the image to be detected are The grayscale mean under the channel, For the The first image in the image to be detected Pixels in Grayscale value under the channel, For the All pixels in the image to be detected are The grayscale mean under the channel, For the The total number of pixels in the image to be detected, For the natural constant The exponential function with base , For the natural constant Logarithmic function with base .

5. A method for detecting the polishing quality of a track for processing an LED taping machine according to claim 1, characterized in that: The tags are classified into lossy and lossless.

6. A method for detecting the polishing quality of a track for processing an LED taping machine according to claim 5, characterized in that: When the tag is damaged, an early warning is issued.

7. A method for detecting the polishing quality of a track for processing an LED taping machine according to claim 1, characterized in that: Use a CCD camera or a CMOS camera to collect the images to be inspected of each LED taping machine track.

8. A method for detecting the polishing quality of a track of an LED taping machine according to claim 1, characterized in that: The method further includes performing grayscale processing on the image to be detected.

9. A method for detecting the polishing quality of a track for processing an LED taping machine according to claim 6, characterized in that: The early warning is to remind through a buzzer.

10. A method for detecting the polishing quality of a track of an LED taping machine according to claim 7, characterized in that: The CCD camera or CMOS camera is installed above the track of each LED taping machine.

Images