Multi-type finished product automatic scanning system and sorting method
The automated system, which integrates multi-sensor identification modules and sorting execution modules, solves the problems of low efficiency and difficult data management in manual inspection, and achieves efficient and accurate finished product inspection and quality management, thereby improving production efficiency and data reliability.
Patent Information
- Application Number
- CN202511230218.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-08-29
- Publication Date
- 2026-01-23
AI Technical Summary
In existing technologies, finished product inspection relies on subjective human judgment, which is inefficient, unreliable, and difficult to achieve accurate quantification. Furthermore, manual inspection has become a bottleneck in production capacity, resulting in high rates of missed and false detections, difficulties in data management, and an inability to form a closed-loop data flow.
The system uses a multi-sensor identification module (spectral sensor, 3D laser scanner, high-speed industrial camera) to collect finished product information, compares it with a standard database through a central processing module, drives the sorting execution module to perform automated sorting, generates quality reports, and integrates with the enterprise management system.
It has achieved full automation of finished product testing, improved testing efficiency and result consistency, reduced missed and false detection rates, provided traceable quality data analysis, reduced labor and management costs, and improved production qualification rate and economic benefits.
Smart Images
Figure CN121372901A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The application is a multi-type finished product automatic scanning system and sorting method, which belongs to the field of industrial automation detection. BACKGROUND
[0002] In the modern industrial system, the consistency of the color, texture and pattern of products is a core element that determines brand reputation and customer satisfaction. In many industries such as textiles, printing, coatings, automotive interiors and high-end consumer goods, the product control link in the production process relies heavily on accurate comparison between factory-finished products and standard products.
[0003] The traditional comparison method mainly relies on the subjective visual judgment of quality inspectors, which is inefficient, labor-intensive, and the judgment standard is easily affected by factors such as people, time, and environmental light, resulting in poor consistency and low reliability of test results. In addition, manual inspection is difficult to accurately quantify indicators such as color deviation or texture characteristics, making it difficult to provide objective data for tracing and analysis in quality disputes.
[0004] In addition, in today's mass production lines, manual inspection speed has become a bottleneck in production capacity, and fatigue can lead to increased missed and false detection rates, allowing substandard products to enter the market, increasing after-sales costs and brand risks. Moreover, the inspection results are usually recorded on paper or in simple electronic spreadsheets, making it difficult to integrate with production execution systems (MES) and enterprise resource planning systems (ERP), and unable to form a closed-loop data flow from production to quality, making it difficult to optimize production processes using data. SUMMARY
[0005] To address the deficiencies in the prior art, the present application aims to provide a multi-type finished product automatic scanning system and sorting method to solve the technical problems of existing comparison methods that rely heavily on human subjectivity, are prone to misjudgment, have low efficiency, and have difficulty in data management.
[0006] To achieve the above-mentioned purpose, the present application is implemented by the following technical solution: a multi-type finished product automatic scanning system and sorting method, comprising: An electromechanical control end, comprising a multi-sensing identification module for collecting multi-dimensional information of finished products and a sorting execution module for executing sorting actions; An upper computer end, comprising a central processing module in communication with the electromechanical control end and a standard database connected to the central processing module; The central processing module is configured to receive finished product data output by the multi-sensing identification module, compare the data with standard finished product data in the standard database, generate control instructions based on the comparison results to drive the sorting execution module to perform corresponding sorting actions, and record the execution results of the sorting actions to generate a quality report.
[0007] Further, the multi-sensing identification module comprises a spectral sensor for collecting color data of the finished product, a 3D laser scanner for collecting texture characteristic parameters of the finished product, and a high-speed industrial camera for identifying pattern characteristic parameters of the finished product.
[0008] Further, the standard finished product data stored in the standard database comprises standard color data, texture characteristic parameters, and pattern characteristic parameters.
[0009] Further, the sorting mechanism in the sorting execution module is one or more of a mechanical arm, a pneumatic push rod, or a rotary sample divider, and the central processing module controls different sorting mechanisms to sort according to the comparison result of the finished product data output by the multi-sensing identification module and the standard finished product data in the standard database.
[0010] Further, the upper computer end further comprises a quality report generation module, which is used to generate a quality report according to the processing data of the central processing module.
[0011] Further, the system further comprises an enterprise management module, which is one of a manufacturing execution system or an enterprise resource planning system, and is in communication connection with the quality report generation module.
[0012] Further, the central processing module is in bidirectional communication connection with the enterprise management module, the central processing module generates a production report of the batch, and automatically uploads the production report to the enterprise management module in a structured data format, and receives a new production instruction issued by the enterprise management module.
[0013] Further, the quality report comprises real-time sorting quantity statistics, a qualified rate, a defective product type distribution, and characteristic deviation data thereof from the standard finished product.
[0014] Further, the electromechanical control end further comprises an automatic conveying path for placing and conveying the to-be-tested finished product, the multi-sensing identification module is located above the automatic conveying path, and the sorting execution module is located at the end of the automatic conveying path.
[0015] A sorting method of a multi-type finished product scanning automation system, comprising the following steps: S1, collecting and comparing step: a multi-sensing identification module collects multi-dimensional data of a finished product, and a central processing module compares the multi-dimensional data with standard finished product data in a standard database; S2, sorting and feedback step: the central processing module generates a control instruction according to the comparison result, drives a sorting execution module to perform a corresponding sorting action, and generates a quality report of the sorting result back to the central processing module, for subsequent adjustment and optimization.
[0016] The beneficial effects of the present application are: The present application realizes the full-process automation operation from product information collection, data comparison to physical sorting through the integration of high-precision multi-sensing identification modules (such as spectrum sensors, 3D laser scanners, industrial cameras) and sorting execution mechanisms on the automatic conveying path, greatly improves the detection efficiency, and through machine vision and spectrum analysis technology, the color, texture, pattern and other characteristics of the finished product are accurately quantified, collected and analyzed, eliminating the influence of artificial subjectivity, making the test results have high consistency, and reducing the misjudgment of the finished product.
[0017] The quality report generation module of the present application can automatically generate a quality report containing detailed deviation data and statistical charts, and can be directly connected to MES / ERP and other enterprise management systems, providing comprehensive and traceable quality data analysis for enterprises.
[0018] Through the efficient closed-loop quality control loop, not only can the unqualified products be found in time, but also through the data feedback mechanism, the quality problems (such as specific color deviation trend, texture defect) can be fed back to the upstream production link in real time, so as to guide the adjustment of process parameters and reduce the generation of unqualified products from the source.
[0019] In summary, while reducing a large amount of labor cost and management cost, the present application greatly reduces the missed detection and misjudgment rate, effectively avoids the rework and after-sales cost caused by batch quality accidents, improves the production qualified rate and overall economic benefit, and enhances the market competitiveness of enterprises. BRIEF DESCRIPTION OF DRAWINGS
[0020] Other features, objects and advantages of the present application will become more apparent through reading the detailed description of the non-limiting embodiments with reference to the following drawings: Figure 1 It is a structural schematic diagram of a multi-type finished product scanning automation system of the present application; Figure 2 It is a flowchart of a multi-type finished product scanning automation system of the present application; Figure 3 It is a step schematic diagram of a sorting method of a multi-type finished product scanning automation system of the present application.
[0021] The reference signs are: 1, multi-sensing identification module; 2, sorting execution module; 3, central processing module; 4, standard database; 5, quality report generation module; 6, enterprise management module; 7, automatic conveying path. DETAILED DESCRIPTION
[0022] In order to make the technical means, creative features, purposes and effects of the present application easy to understand, the present application is further described below in conjunction with specific embodiments.
[0023] A specific embodiment of a multi-type finished product automatic scanning system: Embodiment 1: As shown in Figure 1 and Figure 2 , the present embodiment provides a basic structure of a multi-type finished product automatic scanning system, and details the specific workflow for completing a basic scanning, comparison and sorting task of the system.
[0024] The present application provides a multi-type finished product automatic scanning system, which comprises: an electromechanical control end, an upper computer end and an enterprise management module 6. I. Software and hardware structure: 1. The electromechanical control end comprises: an automatic conveying path 7 for placing and conveying the to-be-tested finished product; a multi-sensing identification module 1 arranged above the automatic conveying path 7 for collecting multi-dimensional information of the to-be-tested finished product; and a sorting execution module 2 arranged at the end of the automatic conveying path 7 for performing sorting actions; the automatic conveying path 7 comprises a fixing device or a buffer belt for ensuring the stability of the product in the detection area.
[0025] Preferably, the automatic conveying path 7 further comprises a light control assembly for providing a constant light source to ensure the stability and accuracy of visual monitoring.
[0026] Preferably, the automatic conveying path 7 further comprises a shielding assembly for reducing the influence of external electromagnetic interference on the system.
[0027] 2. The upper computer end comprises: a central processing module 3 for receiving the finished product data output by the multi-sensing identification module 1 and issuing sorting instructions for the sorting execution module 2; a standard database 4 connected with the central processing module 3; and a quality report generation module 5 for generating a quality report according to the processing data of the central processing module 3.
[0028] 3. The enterprise management module 6 is one of a manufacturing execution system (MES) or an enterprise resource planning system (ERP), which is communicatively connected with the quality report generation module 5 and used for transmitting the quality report, so as to facilitate the subsequent traceability of the finished product quality and the production optimization of the previous production line equipment.
[0029] II. Specific workflow: 1. Data entry: The operator places a standard finished product on the automated conveying path 7, the system starts scanning, the multi-sensing identification module 1 collects the color, texture, pattern data of the standard finished product, and uploads it to the central processing module 3, which takes these data as a reference, with "standard product ID: S001" as the identifier, as standard finished product data (standard color data, texture characteristic parameters and pattern characteristic parameters) stored in the standard database 4, for subsequent comparison.
[0030] 2. Online detection and sorting: After the standard finished product data in the standard database 4 is entered, the operator places the finished product to be detected on the automated conveying path 7 for online detection. The specific detection process is as follows: Spectrum sensor: used to collect color data of the product to be tested, including color, spectral reflectance, color uniformity, etc. 3D laser scanner: used to collect texture characteristic parameters of the product to be tested, including texture direction, texture roughness, texture depth, etc. High-speed industrial camera: used to identify pattern characteristic parameters of the product to be tested, including scratches, labels, shapes or sizes on the finished product, etc. According to the detection results, the finished products are classified, and the system will start the corresponding sorting mechanism (mechanical arm, pneumatic push rod, rotary sample separator, etc.), which will guide the products to different classification areas according to the categories. The specific sorting process is as follows: Qualified products: the detection data is basically consistent with the standard finished product, and the qualified products are sorted into the packaging or warehousing area by the mechanical arm. Defective products: the detection data has slight deviation from the standard finished product, but does not exceed the scrap range, and the defective products are sorted into the repair or rework area by the pneumatic push rod for further processing. Scrap products: the detection data has obvious deviation from the standard finished product, and the scrap products are sorted into the scrap area by the rotary sample separator for centralized processing or recycling, and the reasons for the scrap products are recorded for understanding.
[0031] For example: calculate the ΔE value of the color of the current product to be tested and the color of the standard product (i.e. calculate the deviation of the texture roughness from the standard value): the central processing module 3 makes a judgment according to the preset rules (such as ΔE less than 1.0 is qualified, ΔE between 1.0-2.5 is defective, and ΔE greater than or equal to 2.5 is scrap), and generates a control instruction to drive the corresponding sorting mechanism to act, and sort the detected finished products into the corresponding area.
[0032] Wherein, the sorting mechanism of the present application can be adjusted according to the sorting equipment on the automatic conveying path 7, and when the sorting equipment type is less, two kinds of finished products can be classified by one sorting equipment; and the sorting equipment of the present application has no fixed relationship with the finished product type and the to-be-tested finished product type, and the present embodiment only provides one pairing mode, and the operator can adjust it according to the needs.
[0033] 3. Report generation and optimization: The quality report generation module 5 records in detail the detection results, deviation data and final sorting state of each finished product, and automatically generates a quality report containing real-time sorting quantity statistics, qualified rate, defective product type distribution and characteristic deviation data of the standard finished product, which is displayed on the human-computer interaction interface and transmitted to the external enterprise management module 6; The enterprise management module 6 analyzes and records the quality report and feeds it back to the central processing module 3, so as to facilitate the operator to adjust the process parameters of the previous production line and optimize each equipment.
[0034] In summary, the present application realizes the full-process automation operation from finished product information collection, data comparison to physical sorting by integrating high-precision multi-sensing identification modules 1 (such as spectrum sensors, 3D laser scanners, industrial cameras) and multiple sorting execution mechanisms on the automatic conveying path 7, greatly improves the finished product detection efficiency, and through machine vision and spectrum analysis technology, the color, texture, pattern and other characteristics of the product are accurately quantified and collected and analyzed, eliminating the influence of human subjective factors, making the test results have high consistency and reducing the misjudgment of finished products.
[0035] The present application can automatically generate a quality report containing detailed deviation data and statistical charts through the quality report generation module 5, and can be directly connected to MES / ERP and other enterprise management systems, providing comprehensive and traceable quality data analysis for enterprises.
[0036] Through the efficient closed-loop quality control loop, not only can the unqualified products be found in time, but also through the data feedback mechanism, the quality problems (such as specific color deviation trend, texture defect) can be fed back to the upstream production link in real time, so as to guide the adjustment of process parameters and reduce the generation of unqualified products from the source.
[0037] Embodiment 2: This embodiment further describes the adaptive optimization function of the system based on embodiment 1: After the system runs for a week, the central processing module 3 performs unified big data analysis on the historical quality reports, further investigates the causes of the recent appearance of defective and scrap products, and if there is a problem with the raw materials, instructs the operator to replace them; if there is a problem with the equipment, the system will automatically adjust the classification interval of the finished products during the waiting period for the operator to debug, and after calibration, it will return to the default value.
[0038] For example: after big data analysis, it can be found that the proportion of “defective products” due to “color deviation” has increased by 15% in recent times, and the deviation value is mostly concentrated between ΔE = 1.2-1.4. The operator can gradually investigate whether it is a raw material problem or an equipment problem.
[0039] After excluding the possibility of raw material problems, the system can perform adaptive adjustment while waiting for manual calibration of the sensor. The specific adjustment process is as follows: the central processing module 3 will automatically generate a device maintenance prompt and send it to the terminal of the maintenance personnel; and during the waiting period for calibration, in order to reduce system misjudgment, the color tolerance threshold of “defective products” is adjusted adaptively, i.e. ΔE = 1.0 is widened to ΔE = 1.1, and finished products with ΔE values between 1.0 and 1.1 will be reclassified as “qualified products”, ensuring the continuity of production. After the sensor is calibrated, the system automatically restores the tolerance threshold to the default value.
[0040] Embodiment 3 This embodiment further describes the integration of the central processing module 3 of the system with the external enterprise management module 6 based on embodiment 1: The host computer end of the present application also has a communication interface module for establishing a bidirectional connection with the external enterprise management module 6, i.e. after completing the detection of each production batch, the central processing module 3 not only generates a production report, but also automatically uploads the summary data of the batch (such as batch number, total quantity, qualified quantity, quantity of various defective products, and main defect types) in a structured data format to the enterprise management module 6, so that the manufacturing execution system or enterprise resource planning system in the enterprise management module 6 updates the production quality status of the batch according to the above data.
[0041] When the factory switches to a new product model, the enterprise management module 6 will issue a new “production instruction” to the central processing module 3 of the present application, which contains the standard finished product ID (such as “S002”) corresponding to the new model product. After receiving the instruction, the central processing module 3 automatically calls the data of “S002” from the standard database 4 as the new comparison reference, without the need for manual switching, and automatically realizes seamless linkage of production line model switching.
[0042] A specific implementation of a sorting method of a multi-type finished product scanning automation system For example Figure 3As shown, the present application provides a technical solution of a sorting method of a multi-type finished product scanning automation system: which includes the following detailed steps: S1, scanning and data acquisition step: the inspected finished product is carried by the automatic conveying path and transmitted to the scanning station; each component in the multi-sensing identification module 1 is started, and the same inspected finished product is synchronously triggered and collected, and each sensor uploads the collected original data to the central processing module 3 in real time through the high-speed industrial bus; S2, data fusion and comparison step: the central processing module 3 receives multi-source sensing data and starts parallel processing threads (image processing thread, color analysis thread, texture analysis thread), the central processing module 3 integrates the comparison results of the three threads, and makes a comprehensive judgment (qualified product, defective product or scrap product) on the inspected finished product according to the preset and self-defined quality tolerance threshold; Among them, the image processing thread: using machine learning algorithm (such as convolutional neural network CNN), extracting image features, matching them with the pattern template of standard finished product in standard database 4, calculating the pattern coincidence degree and detecting whether there are scratches, stains and other defects; the color analysis thread: converting the measured spectrum data into standard color space value, and comparing with the standard color value in the database to calculate the color difference value; the texture analysis thread: processing the three-dimensional point cloud data of the finished product, calculating the characteristic parameters of the surface texture, and comparing with the standard texture parameters in the database.
[0043] S3, execution and physical sorting step: the central processing module 3 generates corresponding control instructions (such as "qualified mechanical hip action", "defective pneumatic push rod action", "scrap rotating sample separator steering") according to the judgment result, which can be sent to the sorting execution module 2 through I / O interface or field bus; the sorting mechanism in the sorting execution module 2 receives the instruction and executes the sorting action under the accurate timing control to guide the finished product to the corresponding classification area: S4, data recording and optimization step: the central processing module 3 synchronously records the production report of this batch of sorting and stores it into the historical database, and at the same time, the quality report generation module 5 generates the quality report according to the sorting data, which is automatically returned to the external enterprise management module 6 (such as MES) through the communication interface.
[0044] Among them, the central processing module 3 can analyze the trend of long-term historical data and send device maintenance warning (such as "spectrum sensor reading drift, suggest calibration") to the administrator in time, and can automatically fine-tune the tolerance threshold in the standard database 4 during the calibration waiting process, realize the self-adaptive optimization of the system, and form a continuous improvement closed-loop quality control system.
[0045] The foregoing merely illustrates the principles of the application and application of its leading features. This application is not limited to the illustrative embodiments shown and described herein. Rather, the scope of the present application is defined by the appended claims, and other embodiments of this application will readily occur to those skilled in the art. Accordingly, the application is not limited to that described in the foregoing description or illustrated in the accompanying drawings. It is intended to cover any adaptations or variations of the present application and to encompass within the scope of the patent the appropriate scope of equivalents. Any and all embodiments of the present application can be practiced alone or in combination with one another. Thus, individual features of embodiments of this application can be used in combinations other than the combinations explicitly described herein. It is intended that each of the claims is defined not only by the elements embodied in the claim, but also by the alternative embodiments of the elements embodied in the claim. Furthermore, it is intended that means-plus-function claims be construed such that the functions recited in the claims should not be limited to the corresponding structures that are presently known to be the best techniques for performing the functions, but should be understood to include any future techniques of performing the functions described by the functions.
[0046] In addition, it should be understood that although the description herein is made on the basis of the embodiments, not every embodiment contains only one independent technical solution, and the description herein is made in this way only for the sake of clarity, and those skilled in the art should understand the description as a whole, and the technical solutions in each embodiment can also be properly combined to form other embodiments that those skilled in the art can understand.
Claims
1. A multi-type finished product automated scanning system, characterized by: It comprises: An electromechanical control end comprising a multi-sensing identification module (1) for collecting multi-dimensional information of finished products and a sorting execution module (2) for performing sorting actions; A host computer end comprising a central processing module (3) in communication connection with the electromechanical control end and a standard database (4) connected with the central processing module (3); The central processing module (3) is used for receiving finished product data output by the multi-sensing identification module (1), comparing the data with standard finished product data in the standard database (4), generating a control instruction according to the comparison result to drive the sorting execution module (2) to perform a corresponding sorting action, and recording the execution result of the sorting action to generate a quality report.
2. The multi-type finished product automated scanning system according to claim 1, wherein: The multi-sensing identification module (1) comprises a spectral sensor for collecting color data of finished products, a 3D laser scanner for collecting texture characteristic parameters of finished products, and a high-speed industrial camera for identifying pattern characteristic parameters of finished products.
3. The multi-type finished product automated scanning system according to claim 1, wherein: The standard finished product data stored in the standard database (4) includes color data, texture characteristic parameters and pattern characteristic parameters.
4. The multi-type finished product automated scanning system according to claim 1, wherein: The sorting mechanism in the sorting execution module (2) is one or more of a mechanical arm, a pneumatic push rod or a rotary sample divider, and the central processing module (3) controls different sorting mechanisms to sort according to the comparison result of the finished product data output by the multi-sensing identification module (1) and the standard finished product data in the standard database (4).
5. The multi-type finished product automated scanning system according to claim 1, wherein: The host computer end further comprises a quality report generation module (5) for generating a quality report according to the processing data of the central processing module (3).
6. A multi-type finished product automated scanning system according to claim 5, characterized in that: The system further comprises an enterprise management module (6) which is one of a manufacturing execution system or an enterprise resource planning system and is in communication connection with the quality report generation module (5).
7. A multi-type finished product automated scanning system according to claim 6, characterized in that: The central processing module (3) is in bidirectional communication connection with the enterprise management module (6), the central processing module (3) generates a production report of the batch, automatically uploads the production report to the enterprise management module (6) in a structured data format, and receives a new production instruction issued by the enterprise management module (6).
8. The multi-type finished product automated scanning system according to claim 5, wherein: The quality report includes real-time sorting quantity statistics, qualified rate, defective product type distribution and characteristic deviation data from standard finished products.
9. The multi-type finished product automated scanning system according to claim 1, wherein: The electromechanical control end further comprises an automatic conveying path (7) for placing and conveying the to-be-tested finished products, the multi-sensing identification module (1) is located above the automatic conveying path (7), and the sorting execution module (2) is located at the end of the automatic conveying path (7).
10. A sorting method of a multi-type finished product automated scanning system, using the multi-type finished product scanning automated system according to any one of claims 1 to 9, characterized in that: It comprises the following steps: S1, collecting and comparing step: the multi-sensing identification module (1) collects multi-dimensional data of finished products, and the central processing module (3) compares the multi-dimensional data with standard finished product data in the standard database (4); S2, sorting and feedback step: the central processing module (3) generates a control instruction according to the comparison result, drives the sorting execution module (2) to perform a corresponding sorting action, and returns the sorting result to the central processing module (3) to generate a quality report, which is convenient for subsequent adjustment and optimization.