Intelligent factory production line visual inspection system and method

Through the intelligent factory production line visual inspection system, multi-spectral cameras and deep learning algorithms are used to identify product defects in real time and automatically eliminate them, solving the problems of low manual detection efficiency and missed detection risks, and achieving high-precision, stability and efficient product detection. Reduced production costs.

CN119023577APending Publication Date: 2024-11-26CHONGQING COLLEGE OF ELECTRONICS ENG
View PDF 0 Cites 3 Cited by

Patent Information

Application Number
CN202411173365.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-26
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

In industrial automation, assembly line product defect monitoring and identification relies on manual methods, resulting in inefficiency, high costs and missed inspection risks, limiting the improvement of production efficiency and product quality stability.

Method used

A visual inspection system for intelligent factory production line is designed, including environment perception module, system control module, dynamic adjustment lighting device, multi-spectral camera module, intelligent identification and processing module, automatic removal module and remote interactive control panel module, using deep learning algorithms Identify product defects in real time and realize automatic removal to reduce manual intervention.

Benefits of technology

It realizes high-precision, stability and consistency product defect detection, improves production efficiency and product quality, reduces labor costs and the risk of missed inspections, and supports 24-hour uninterrupted operation.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119023577A_ABST
    Figure CN119023577A_ABST
Patent Text Reader

Abstract

The invention provides an intelligent factory production line visual detection system and method. The method comprises the steps that an environment sensing module obtains illumination intensity around products on a production line in real time; the system integrated control module receives the illumination intensity around the product in real time and compares the illumination intensity with a preset threshold value; the dynamic adjustment illuminator dynamically adjusts the illumination intensity around the product on the production line according to the comparison result; the multispectral camera module acquires an image of a product meeting the illumination intensity in real time; the intelligent identification and processing module is used for acquiring and analyzing the image data from the multispectral camera in real time so as to identify defects on the surface of the product; if defects exist on the surfaces of the identified products, the products with the defects are removed through an automatic removing module and placed in a defective product placing area; if the number of products placed in the defective product placing area exceeds two thirds, the result is fed back to the remote interaction control panel module, and production personnel are informed of detecting the products in the defective product placing area through the early warning module to judge whether false detection exists or not.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of industrial automation and machine vision technology, and in particular to a smart factory production line visual inspection system and method. Background Art

[0002] With the rapid development of industrial automation and intelligent technology, intelligent control systems have become the core driving force in the modern industrial field. The widespread application of this technology has not only significantly improved production efficiency, but also greatly optimized the production process, reduced labor costs, and ensured the stability of product quality. However, in the current industrial automation practice, although intelligent control systems play a key role in many links, in the field of product defect monitoring and identification on the assembly line, manual methods are still generally relied on. This situation not only limits the further improvement of production efficiency, but also increases costs and may introduce the risk of missed detection.

[0003] In traditional assembly line operations, the detection of product defects often relies on workers' visual inspection or the use of simple auxiliary tools such as magnifying glasses. Although this method can identify defects such as scratches, stains, and dimensional deviations of products to a certain extent, its limitations are also obvious. First, manual inspection is inefficient, especially on large-scale, high-speed production lines. It is difficult for workers to maintain a high level of attention and accuracy, which can easily lead to fatigue and misjudgment. Secondly, labor costs are high. With the rise in labor costs, companies are facing increasing cost pressures. What's more serious is that manual inspection has the risk of missed inspections, and some minor or hidden defects may be difficult to be discovered in time, resulting in unqualified products entering the market, damaging the company's reputation and customer interests. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention proposes a smart factory production line visual inspection system and method to solve the technical problems existing in the prior art.

[0005] The technical solution adopted by the present invention is:

[0006] In a first aspect, a smart factory production line visual inspection system is provided, comprising:

[0007] Environmental perception module, system control module, dynamic adjustment illuminator, multi-spectral camera module, intelligent recognition and processing module, automatic rejection module, remote interactive control panel module and early warning module;

[0008] The environment sensing module is used to sense and obtain the light intensity around the products on the production line in real time, and feed back the real-time obtained light intensity to the system control module;

[0009] The system control module, as the center of the system, is responsible for communication management between devices and control of sending and receiving instructions, as well as receiving the light intensity of the environment perception module and comparing it with a preset threshold;

[0010] The dynamic adjustment lighting device is used to receive the comparison result of the system control module and dynamically adjust the light intensity around the products on the production line according to the comparison result;

[0011] The multispectral camera module is used for multispectral imaging to obtain product image data on the production line in real time;

[0012] The intelligent recognition and processing module is used to receive and analyze the image data from the multispectral camera in real time, and process and analyze the acquired image data to identify defects on the surface of the product;

[0013] The automatic rejection module is used to reject defective products on the production line and place the defective products in the defective product placement area;

[0014] The early warning module is used to notify production personnel to inspect the products in the defective product placement area to determine whether there is a false detection;

[0015] The remote interactive control panel module, as the main interactive interface for users, displays the system status and sends operation instructions to the system control module through a graphical interface, while displaying feedback information.

[0016] Furthermore, the system control module communicates with the remote interactive control panel module via a wireless communication module.

[0017] Furthermore, when the system control module compares the preset threshold and obtains the comparison result that the light intensity around the detected product is lower than the minimum light intensity of the multi-spectral camera module or higher than the maximum light intensity of the multi-spectral camera module, the system control module sends an instruction to the dynamic adjustment illuminator to adjust the light intensity around the product to a light intensity range between the minimum and maximum light intensities of the multi-spectral camera module through the dynamic adjustment illuminator.

[0018] Furthermore, the product image data acquired in real time by the multi-spectral camera module is product image data having a light intensity that meets a threshold range of the system control module.

[0019] Furthermore, the intelligent recognition and processing module receives and analyzes the image data from the multispectral camera in real time, and uses a deep learning algorithm to perform real-time analysis on the acquired image data to identify defects on the product surface.

[0020] Furthermore, the deep learning algorithm is any one of: a convolutional neural network, a regional convolutional neural network, a generative adversarial network and an autoencoder.

[0021] Furthermore, the defects on the surface of the product include scratches, stains, dimensional deviations, cracks, dents and bent pins.

[0022] Furthermore, the remote interactive control panel module supports touch or voice input.

[0023] Furthermore, the wireless communication module adopts Wi-Fi 6 or 5G communication technology to realize data transmission between the remote interactive control panel module and the system control module.

[0024] In a second aspect, a method for visual inspection of a production line in an intelligent factory is provided, comprising:

[0025] The environmental perception module obtains the light intensity around the products on the production line in real time;

[0026] The system integrated control module receives the light intensity around the product in real time and compares the light intensity with a preset threshold;

[0027] Dynamically adjust the lighting intensity around the product on the production line according to the threshold comparison results;

[0028] The multi-spectral camera module acquires images of products that meet the light intensity in real time;

[0029] Intelligent recognition and processing module, which acquires and analyzes image data from multispectral cameras in real time, processes and analyzes the acquired image data to identify defects on the product surface;

[0030] If there are defects on the surface of the identified product, the defective products on the production line will be removed through the automatic rejection module and placed in the defective product placement area;

[0031] If the defective product placement area contains more than two-thirds of the products, real-time feedback will be given to the remote interactive control panel module, and the early warning module will notify the production staff to inspect the products in the defective product placement area to determine whether there is any false detection.

[0032] It can be seen from the above technical solution that the beneficial technical effects of the present invention are as follows:

[0033] 1. Through the multi-spectral camera module and the intelligent recognition and processing module, the system of the present invention can capture and analyze product images in real time and accurately, identify various defects on the product surface, such as scratches, stains, dimensional deviations, cracks, dents, etc., thereby improving the overall quality of the product. Compared with traditional manual inspection, the machine vision inspection system has higher accuracy and stability, can reduce false detection and missed detection caused by human factors, and ensure the stability and consistency of product quality.

[0034] 2. The present invention can realize the automatic detection process without manual intervention, greatly improving the detection speed and production efficiency. At the same time, it can also work 24 hours a day, further improving the overall production capacity of the production line, and when unqualified products are detected, the automatic rejection module can quickly remove them from the production line, avoiding the impact on the subsequent production process, thereby ensuring the smooth operation of the production line.

[0035] 3. The present invention reduces the need for manual inspectors and reduces labor costs through automated inspection. At the same time, since the machine vision inspection system can work stably for a long time without rest, it also reduces production losses caused by employee fatigue or negligence. Moreover, through real-time monitoring and data analysis, the system can help enterprises better understand the operating status of the production line and product quality, thereby optimizing resource allocation and production planning and reducing production costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following is a brief introduction to the drawings required for the specific embodiments or the prior art description. In all the drawings, similar elements or parts are generally identified by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn according to the actual scale.

[0037] Figure 1 This is a system diagram of an embodiment of the present invention. DETAILED DESCRIPTION

[0038] The following embodiments of the technical solution of the present invention are described in detail in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and are therefore only used as examples, and cannot be used to limit the protection scope of the present invention.

[0039] It should be noted that, unless otherwise specified, the technical terms or scientific terms used in this application should have the common meanings understood by those skilled in the art to which the present invention belongs.

[0040] Example

[0041] This embodiment provides a smart factory production line visual inspection system, such as Figure 1As shown, the working principle of the embodiment is described in detail below:

[0042] In order to improve product quality and production efficiency, an electronic product manufacturing plant decided to introduce a smart factory production line visual inspection system. The system aims to accurately detect and remove surface defects of products on the production line through automation and intelligent means, thereby improving the yield and efficiency of the overall production line.

[0043] System composition and configuration

[0044] The system includes: environmental perception module, system control module, dynamic adjustment lighting, multi-spectral camera module, intelligent recognition and processing module, automatic rejection module, remote interactive control panel module and early warning module;

[0045] Environmental perception module:

[0046] Used to sense and obtain the light intensity around the products on the production line in real time, and feed back the real-time obtained light intensity to the system control module;

[0047] In some embodiments, the environmental sensing module is a high-precision photosensor, such as a RY-G / W light sensor, which senses the light intensity around the products on the production line in real time and sends the data to the system control module at a frequency of 10 times per second.

[0048] System control module:

[0049] As the center of the system, it is responsible for communication management between devices and control of sending and receiving instructions, as well as receiving the light intensity of the environment sensing module and comparing it with the preset threshold;

[0050] In some embodiments, the system control module serves as the core of the system, integrating a high-performance processor and a communication interface, and is responsible for communication and command control between modules; according to the factory manual of the multi-spectral camera, the minimum light intensity of the multi-spectral camera module is 300 lux, and the maximum light intensity of the multi-spectral camera module is 800 lux, so the preset light intensity threshold in the system control module is set to 300-800 lux.

[0051] In some embodiments, the system control module and the remote interactive control panel are connected via Wi-Fi 6 wireless communication technology to achieve remote monitoring and operation.

[0052] Dynamically adjust illuminators:

[0053] Used to receive the comparison result of the system control module and dynamically adjust the light intensity around the products on the production line according to the comparison result;

[0054] In some embodiments, the dynamically adjustable lighting device is an LED light source and an intelligent dimming system, which dynamically adjusts the light intensity around the product to a preset range (300-800 lux) according to the instructions of the system control module.

[0055] Multispectral camera module:

[0056] Used for multispectral imaging to obtain product image data on the production line in real time;

[0057] In some embodiments, the multispectral camera module is a high-resolution multispectral camera that can simultaneously capture image data in multiple bands such as visible light and infrared light, ensuring that clear product images can be obtained under various lighting conditions.

[0058] Intelligent identification and processing module:

[0059] It is used to receive and analyze the image data from the multispectral camera in real time, and use a deep learning algorithm to perform real-time analysis on the acquired image data to identify defects on the product surface.

[0060] The deep learning algorithm is any one of a convolutional neural network, a regional convolutional neural network, a generative adversarial network and an autoencoder;

[0061] Convolutional Neural Networks (CNNs):

[0062] CNNs automatically extract high-level feature representations from raw images through structures such as convolutional layers, pooling layers, and fully connected layers for tasks such as image classification, object detection, and image segmentation.

[0063] In product defect detection, CNNs can learn the normal and abnormal features of the product surface, thereby effectively identifying defects such as scratches and stains.

[0064] Region-based Convolutional Neural Networks (R-CNNs):

[0065] R-CNNs and its variants (such as Fast R-CNN, Faster R-CNN) are mainly used for object detection tasks. They first generate a series of candidate regions that may contain objects, and then classify and regress bounding boxes for each candidate region to determine the location and category of the object.

[0066] In product defect detection, R-CNNs can be used to locate and identify specific defective areas on the product.

[0067] Generative Adversarial Networks (GANs):

[0068] GANs consist of two networks, a generator and a discriminator, which learn the distribution of data in an adversarial manner. Although GANs are commonly used for tasks such as generating images, they can also be used for anomaly detection in defect detection.

[0069] The generator is trained to learn the characteristics of normal products, and the discriminator is used to distinguish the generated product images from the real product images. When the discriminator can identify abnormal images, it means that a product defect has been detected.

[0070] Autoencoders:

[0071] An autoencoder is an unsupervised learning neural network that learns the compressed representation of data through the structure of an encoder and a decoder. In defect detection, an autoencoder can be used to learn the feature representation of normal products.

[0072] When a product image containing defects is input, since the decoder cannot completely reconstruct the original image, the defects can be detected by comparing the differences between the original image and the reconstructed image.

[0073] In some embodiments, the intelligent recognition and processing module uses a deep learning algorithm based on a convolutional neural network to perform real-time processing and analysis on the image data acquired by the multispectral camera to identify defects such as scratches, stains, dimensional deviations, cracks, dents, and bent pins on the product surface.

[0074] Automatic rejection module:

[0075] Used to remove defective products on the production line and place them in the defective product placement area;

[0076] In some embodiments, the automatic rejection module is a robotic arm and a pneumatic device that accurately rejects detected defective products and places them in a designated defective product placement area.

[0077] Early warning module:

[0078] Used to notify production personnel to inspect products in the defective product placement area to determine whether there is any false detection;

[0079] In some embodiments, when the number of products in the defective product placement area exceeds two-thirds, an early warning mechanism is automatically triggered, and a notification is sent to production personnel through an audible and visual alarm and a remote interactive control panel.

[0080] Remote interactive control panel module:

[0081] Provides a graphical user interface, supports touch and voice input, and displays system status, operation instructions and feedback information in real time. Production personnel can remotely monitor the status of the production line and send operation instructions through this module.

[0082] Implementation Process

[0083] System startup and initialization:

[0084] The remote interactive control panel starts the system, and each module performs self-check and initialization to ensure the normal operation of the system.

[0085] Environmental perception and lighting adjustment:

[0086] The environmental perception module continuously monitors the light intensity around the product and sends the data to the system control module. The system control module compares the real-time light intensity with the preset threshold. If it exceeds the range, it dynamically adjusts the lighting to the appropriate light intensity.

[0087] Image acquisition and defect recognition:

[0088] The multispectral camera module acquires product image data in real time under appropriate lighting conditions and sends it to the intelligent recognition and processing module. The intelligent recognition module uses deep learning algorithms to process and analyze image data and identify defects on the product surface.

[0089] Defective product removal and early warning:

[0090] If defects are detected on the product surface, the automatic rejection module will remove the defective product from the production line and place it in the defective product placement area. When the number of defective products reaches the warning threshold, the warning module will automatically trigger and notify the production staff to conduct a re-inspection.

[0091] Remote monitoring and operation:

[0092] Production personnel monitor the system status in real time through the remote interactive control panel and send operation instructions, such as adjusting detection parameters and viewing historical records.

[0093] The above embodiments have achieved the following: Improved yield rate: The yield rate of the production line has been significantly improved by accurately detecting and eliminating defective products. Improved production efficiency: The automated detection and elimination process reduces manual intervention and improves the overall efficiency of the production line. Reduced labor costs: The reliance on manual inspection personnel on the production line has been reduced, reducing labor costs. Enhanced management flexibility: The remote interactive control panel enables production managers to monitor the system status anytime and anywhere and adjust production strategies in a timely manner.

[0094] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A smart factory production line visual inspection system, characterized in that: include: Environmental perception module, system control module, dynamic adjustment illuminator, multi-spectral camera module, intelligent recognition and processing module, automatic rejection module, remote interactive control panel module and early warning module; The environment sensing module is used to sense and obtain the light intensity around the products on the production line in real time, and feed back the real-time obtained light intensity to the system control module; The system control module, as the center of the system, is responsible for communication management between devices and control of sending and receiving instructions, as well as receiving the light intensity of the environment perception module and comparing it with a preset threshold; The dynamic adjustment lighting device is used to receive the comparison result of the system control module and dynamically adjust the light intensity around the products on the production line according to the comparison result; The multispectral camera module is used for multispectral imaging to obtain product image data on the production line in real time; The intelligent recognition and processing module is used to receive and analyze the image data from the multispectral camera in real time, and process and analyze the acquired image data to identify defects on the surface of the product; The automatic rejection module is used to reject defective products on the production line and place the defective products in the defective product placement area; The early warning module is used to notify production personnel to inspect the products in the defective product placement area to determine whether there is a false detection; The remote interactive control panel module, as the main interactive interface for users, displays the system status and sends operation instructions to the system control module through a graphical interface, while displaying feedback information.

2. According to claim 1, a smart factory production line visual inspection system is characterized in that: The system control module communicates with the remote interactive control panel module via a wireless communication module.

3. The intelligent factory production line visual inspection system according to claim 1 is characterized in that: When the system control module compares the preset threshold and obtains the comparison result that the light intensity around the detected product is lower than the minimum light intensity of the multi-spectral camera module or higher than the maximum light intensity of the multi-spectral camera module, the system control module sends an instruction to the dynamic adjustment illuminator to adjust the light intensity around the product to a light intensity range between the minimum and maximum light intensities of the multi-spectral camera module through the dynamic adjustment illuminator.

4. The intelligent factory production line visual inspection system according to claim 3 is characterized in that: The product image data acquired in real time by the multi-spectral camera module is product image data having a light intensity that meets a threshold range of the system control module.

5. The intelligent factory production line visual inspection system according to claim 1, characterized in that: The intelligent recognition and processing module receives and analyzes the image data from the multispectral camera in real time, and uses a deep learning algorithm to perform real-time analysis on the acquired image data to identify defects on the product surface.

6. The intelligent factory production line visual inspection system according to claim 5, characterized in that: The deep learning algorithm is any one of: convolutional neural network, regional convolutional neural network, generative adversarial network and autoencoder.

7. The intelligent factory production line visual inspection system according to claim 6, characterized in that: The defects on the surface of the product include: scratches, stains, dimensional deviations, cracks, dents and bent pins.

8. The intelligent factory production line visual inspection system according to claim 1, characterized in that: The remote interactive control panel module supports touch or voice input.

9. The intelligent factory production line visual inspection system according to claim 2, characterized in that: The wireless communication module adopts Wi-Fi 6 or 5G communication technology to realize data transmission between the remote interactive control panel module and the system control module.

10. A smart factory production line visual inspection method based on a smart factory production line visual inspection system according to any one of claims 1 to 9, characterized in that: include: The environmental perception module obtains the light intensity around the products on the production line in real time; The system integrated control module receives the light intensity around the product in real time and compares the light intensity with a preset threshold; Dynamically adjust the lighting intensity around the product on the production line according to the threshold comparison results; The multi-spectral camera module acquires images of products that meet the light intensity in real time; Intelligent recognition and processing module, which acquires and analyzes image data from multispectral cameras in real time, processes and analyzes the acquired image data to identify defects on the product surface; If there are defects on the surface of the identified product, the defective products on the production line will be removed through the automatic rejection module and placed in the defective product placement area; If the defective product placement area contains more than two-thirds of the products, real-time feedback will be given to the remote interactive control panel module, and the early warning module will notify the production staff to inspect the products in the defective product placement area to determine whether there is any false detection.

Citation Information

Cited By

  • Production management method and system for single-phase cost control intelligent electric energy meter

    CN119294922A

  • Intelligent factory voice and vision cooperative control method and system

    CN120715909A

  • Intelligent factory voice and visual cooperative control method and system

    CN120715909B