A detection and sorting device and method suitable for high-precision aluminum templates

Abstract

This invention relates to the field of aluminum formwork inspection technology, solving the problem of insufficient accuracy and precision in the inspection and sorting of high-precision aluminum formwork for aluminum buildings due to the difficulty in comprehensively covering complex surfaces and nodes. Specifically, it discloses an inspection and sorting device and method suitable for high-precision aluminum formwork. The device includes a feeding input mechanism, an inspection and sorting mechanism, an output mechanism, and a temporary storage mechanism. The feeding input mechanism is located between the output mechanism and the temporary storage mechanism, and the inspection and sorting mechanism is located above the feeding input mechanism, output mechanism, and temporary storage mechanism. The method includes S100, feeding positioning and preprocessing; S200, multi-dimensional inspection data acquisition; S300, algorithm inspection data processing; S400, sorting execution; and S500, inspection data recording and scheduling. This invention is used for the inspection and sorting of high-precision aluminum formwork in aluminum building structures, improving the inspection effect by increasing the inspection coverage and accuracy of complex surfaces.

Description

Technical Field

[0001] This invention relates to the field of aluminum template inspection technology, and in particular to an inspection and sorting device and method suitable for high-precision aluminum templates. Background Technology

[0002] The inspection and sorting of aluminum formwork for aluminum building structures involves loading and centering the formwork using a conveyor system. High-pressure cleaning and drying remove surface contaminants such as oil and dust. The inspection phase utilizes a vision recognition system, employing multi-directional image acquisition units to achieve three-dimensional scanning of the formwork surface and key areas. This is combined with auxiliary inspection technologies to cover dimensional verification and surface morphology identification, capturing potential defects and dimensional deviations. In the judgment stage, image processing technology extracts feature information, which is compared with the standard requirements for aluminum formwork used in construction to complete defect classification and pass / fail determination. Finally, handling equipment moves the formwork to the corresponding workstation based on the judgment results, simultaneously recording inspection data to support quality traceability.

[0003] In scenarios with extremely high requirements for construction precision and structural safety, such as super high-rise buildings, prefabricated buildings, and landmark buildings, the insufficient surface inspection precision of existing inspection systems is particularly prominent. In these applications, the assembly precision of aluminum formwork directly affects the stability of the building structure, and the requirements for surface flatness and defect control are more stringent. However, existing inspection technologies have blind spots in the inspection of complex nodes, and cannot fully cover the surface condition of critical areas such as the corners of the formwork and the area around reserved holes. At the same time, existing systems are prone to missing subtle defects that may cause hidden dangers during long-term use of the building. Although these minor defects do not affect construction in the short term, they will reduce the durability and reusability of the formwork. Due to interference from environmental factors and the reflective properties of the formwork surface during the inspection process, errors can occur in the determination of surface flatness and defects, which in turn affect the construction precision after the formwork is assembled, making it difficult to meet the stringent standards for structural safety and construction quality in high-requirement building scenarios. Summary of the Invention

[0004] To address the problem of insufficient accuracy and precision in the existing detection and sorting technology for high-precision aluminum formwork used in aluminum buildings due to the difficulty in fully covering complex surfaces and nodes, this invention provides a detection and sorting device and method suitable for high-precision aluminum formwork.

[0005] The technical solution adopted in this invention is:

[0006] A detection and sorting device suitable for high-precision aluminum templates includes a feeding input mechanism, a detection and sorting mechanism, an output mechanism, and a temporary storage mechanism. The feeding input mechanism is disposed between the output mechanism and the temporary storage mechanism, and the detection and sorting mechanism is disposed above the feeding input mechanism, the output mechanism, and the temporary storage mechanism.

[0007] The feeding input mechanism is equipped with a feeding conveyor belt; the output mechanism is equipped with an output guide rail and a support frame that can move along the output guide rail; the detection and sorting mechanism includes a detection transfer component and a three-axis motion component. The detection transfer component is connected to the three-axis motion component. The detection transfer component is equipped with a visual recognition unit for locating detection points and a detection sensor unit for detecting index data. The detection transfer component is also equipped with a workpiece gripping unit. The visual recognition unit and the detection sensor unit are respectively connected to the input end of the device control processing center through signal transmission. The output end of the device control processing center is electrically connected to the three-axis motion component.

[0008] Furthermore, the three-axis motion assembly includes an X-axis motion assembly, a Y-axis motion assembly, and a Z-axis motion assembly. The Z-axis motion assembly is slidably disposed on the Y-axis motion assembly, the X-axis motion assembly is slidably disposed on the Z-axis motion assembly, and the detection and transfer assembly is slidably disposed on the X-axis motion assembly.

[0009] The X-axis motion component, Y-axis motion component, Z-axis motion component and detection transfer component are all connected by motion slide rails and motion sliders.

[0010] Furthermore, the detection and transfer assembly is provided with a spacing adjustment slide rail, and multiple spacing adjustment slider frames are connected to the spacing adjustment slide rail by sliding engagement. The workpiece gripping unit is disposed on the spacing adjustment slider frame and corresponds to the number of spacing adjustment slider frames. Multiple vision recognition units and detection sensor units are respectively disposed on the spacing adjustment slider frame.

[0011] The workpiece gripping unit is equipped with a gripping suction cup, which is used to generate negative pressure suction force by an external vacuum drive mechanism to pick up the aluminum template workpiece.

[0012] Furthermore, the detection sensor unit includes:

[0013] Laser contour sensors are used to capture the three-dimensional contour information of complex surfaces on aluminum template workpieces.

[0014] Eddy current sensors are used to identify near-surface micro-cracks and material uniformity deviations.

[0015] Laser thickness sensors are used to detect the thickness uniformity and local deformation of aluminum template workpieces.

[0016] A high-resolution area array vision sensor is used in conjunction with a vision recognition unit to detect minute surface scratches and pinhole defects.

[0017] The pressure sensing module, integrated into the workpiece gripping unit, is used to monitor the gripping pressure.

[0018] Tilt sensors are used to detect posture deviations in aluminum template workpieces during loading and inspection in real time.

[0019] A detection and sorting method suitable for high-precision aluminum templates includes the following steps:

[0020] S100, feeding, positioning and pretreatment, transports the aluminum template workpiece to be inspected to the preset position, performs posture calibration and surface cleaning and drying;

[0021] S200, multi-dimensional detection data acquisition, through detection area positioning, detection unit spacing adjustment and multiple sensor detection, collects detection index data of aluminum template workpieces;

[0022] S300 algorithm detection data processing: preprocesses the collected raw detection data, optimizes the detection coverage of complex surfaces and enhances the identification of minor defects, and generates detection judgment results and sorting decision instructions;

[0023] S400, Sorting Execution: Based on the sorting decision instructions, qualified and unqualified aluminum template workpieces are transferred to the output mechanism and temporary storage mechanism respectively through grabbing, transferring and classifying placement operations.

[0024] S500: Detection data recording and scheduling; storage of historical detection and sorting data; recording of non-conforming aluminum template workpiece data in the scheduling temporary storage mechanism; and dynamic adjustment of detection and sorting strategies by the device control and processing center based on the non-conforming aluminum template workpiece data.

[0025] Furthermore, S100 specifically includes: S101, feeding and conveying, starting the feeding conveyor belt of the feeding input mechanism to transport the aluminum template workpiece to be inspected to the preset initial inspection position;

[0026] S102, Attitude calibration: The three-axis motion component of the detection and sorting mechanism drives the detection and transfer component to move. Real-time attitude data is collected by the tilt sensor on the detection and transfer component and transmitted to the device control and processing center. The device control and processing center drives the three-axis motion component to calibrate the horizontal attitude of the aluminum template workpiece with the detection benchmark based on the attitude deviation data.

[0027] S103. Surface pretreatment: The cleaning module sprays gas to remove dust and oil stains from the surface of the aluminum template workpiece, and the hot air drying module is started to dry the moisture.

[0028] Furthermore, S200 specifically includes: S201, detection area positioning, the device control and processing center generates preset detection point coordinates according to the model parameters of the aluminum template workpiece to be detected, drives the three-axis motion component to drive the detection transfer component to move along the X, Y, and Z axes, captures the edge and key feature points of the aluminum template workpiece through the visual recognition unit, and performs secondary correction on the preset detection coordinates;

[0029] S202, Spacing adjustment adaptation: Based on the size of the aluminum template workpiece and the distribution of detection points, control the spacing adjustment slide rail to drive multiple spacing adjustment slider frames to slide, adjust the spacing of the vision recognition unit, detection sensor unit and workpiece gripping unit, and detect multiple different areas of the aluminum template workpiece.

[0030] S203, multi-sensor detection: A laser contour sensor scans the complex surface and key nodes of the aluminum template workpiece to collect three-dimensional contour data; an eddy current sensor moves along the surface of the aluminum template workpiece to detect near-surface micro-cracks and material uniformity; a laser thickness sensor moves along a preset path on the aluminum template workpiece to collect thickness data at different locations in real time; and a high-resolution area array vision sensor captures images of surface micro-scratches and pinhole defects; the detection data from multiple sensors is transmitted to the device control and processing center in real time.

[0031] S204. Pre-grabbing status confirmation: The gripping pressure value of the workpiece gripping unit is detected by the pressure sensing module, and at the same time, the tilt sensor collects the posture data of the aluminum template workpiece again to confirm that there is no posture deviation of the aluminum template workpiece during the detection process.

[0032] Furthermore, S300 specifically includes: S301, data preprocessing, where the device control and processing center integrates the raw data collected by each sensor, removes environmental interference and sensor error data through data noise reduction processing, and establishes a complete detection dataset for a single aluminum template workpiece.

[0033] S302. Complex surface detection coverage optimization: The surface contour matching algorithm is used to analyze the complex area data collected by the laser contour sensor, fill in the detection blind area data caused by surface structure occlusion, and generate complete aluminum template workpiece surface model data.

[0034] S303. Enhanced identification of minor defects: The defect feature enhancement algorithm is used to process visual images and eddy current detection data, extract the grayscale features, morphological features and material anomaly features of minor defects, compare them with the preset standard defect threshold, and identify and classify minor defects.

[0035] S304. Detection Result Judgment: The device control and processing center will comprehensively compare the processed detection data with the preset parameters to determine whether the aluminum template workpiece meets the requirements, mark the qualified level and the unqualified defect type, and generate the corresponding sorting decision instruction.

[0036] Furthermore, S400 specifically includes: S401, aluminum template workpiece gripping, according to the sorting decision instruction, the three-axis motion component drives the detection and transfer component to move to the aluminum template workpiece gripping position, the workpiece gripping unit grips the aluminum template workpiece, and the pressure sensing module monitors the pressure value in real time.

[0037] S402, Transfer Positioning: The control and processing center drives the three-axis motion assembly to move the gripped aluminum template workpiece according to the sorting decision instructions. The vision recognition unit assists in positioning the output guide rail and carrier frame of the output mechanism, and at the same time confirms the idle station information of the temporary storage mechanism.

[0038] S403. Classify and place the workpieces. If the aluminum template workpieces are deemed qualified, they are transferred to the support frame of the output mechanism. The support frame moves along the output guide rail to the corresponding qualified product storage area. If the aluminum template workpieces are deemed unqualified, they are transferred to the designated workstation of the temporary storage mechanism for storage. After the sorting operation is completed, the workpiece gripping unit releases pressure and resets to the initial detection position.

[0039] Furthermore, S500 specifically includes:

[0040] S501. Data storage: The device control and processing center classifies and stores the inspection data, defect identification results, and sorting results of each aluminum template workpiece to establish a complete quality traceability database.

[0041] S502, Temporary storage and scheduling: Mark the information of unqualified aluminum template workpieces in the temporary storage mechanism, record the defect type and location, generate a non-conforming product handling strategy, and update the occupancy status of the temporary storage mechanism workstations in real time.

[0042] S503, Equipment linkage optimization: Based on historical detection data and sorting efficiency, the device control and processing center dynamically adjusts the feeding and conveying speed, detection point distribution, and sorting execution rhythm.

[0043] The beneficial effects of this invention are:

[0044] In the operation of this invention, the feeding conveyor belt of the feeding input mechanism transports the aluminum template to be inspected to a preset position, first completing posture calibration and surface cleaning and drying to eliminate interference from impurities and posture deviations, thereby improving the accuracy of the inspection. The three-axis motion component of the inspection and sorting mechanism drives the inspection transfer component to achieve flexible displacement. The vision recognition unit can first locate the inspection point, and combined with the spacing adjustment function of the inspection unit, the inspection sensor unit can more comprehensively cover complex nodes such as the inside and outside corners of the aluminum template and the periphery of reserved holes. By using multiple sensors to collect multi-dimensional index data such as size, flatness, and surface defects, the problem of large inspection blind spots in the prior art is compensated. After the collected raw data is preprocessed by the device control and processing center, the data integrity is improved by the complex surface inspection coverage optimization algorithm, and the sensitivity of identifying minor defects is improved by the micro-defect recognition enhancement algorithm, effectively solving the problems of missed detection of minor defects and judgment errors caused by environmental reflection and surface characteristics. Based on the processed inspection results, the device control and processing center generates sorting decision instructions to execute the output and temporary storage of the corresponding workpiece. This solution achieves more comprehensive coverage detection of complex surfaces and key nodes by combining positioning recognition and motion coordination with detection algorithm optimization and dynamic data control. It significantly improves the accuracy of identifying minute defects and detection precision, effectively reduces judgment errors, ensures the construction precision after aluminum formwork assembly, meets the stringent requirements for structural safety and construction quality of super high-rise buildings, and improves the durability of aluminum formwork, ensuring the integrity of quality traceability and the stability of detection and sorting efficiency. Attached Figure Description

[0045] Figure 1 This is a top view of the device of the present invention.

[0046] Figure 2 This is a front view of the device of the present invention.

[0047] Figure 3 This is a schematic diagram of the detection and sorting mechanism of the present invention.

[0048] Figure 4 This is a flowchart of the method of the present invention.

[0049] Reference numerals: 1-Feeding input mechanism, 11-Feeding conveyor belt, 2-Detection and sorting mechanism, 21-Detection and transfer assembly, 211-Workpiece gripping unit, 212-Gap adjustment slide rail, 213-Gap adjustment slider frame, 22-Three-axis motion assembly, 221-X-axis motion assembly, 222-Y-axis motion assembly, 223-Z-axis motion assembly, 224-Motion slide rail, 3-Output mechanism, 31-Output guide rail, 32-Bearing frame, 4-Temporary storage mechanism. Detailed Implementation

[0050] The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0051] Example 1: As Figures 1-3 As shown, when the device of the present invention is working, the feeding conveyor belt 11 of the feeding input mechanism 1 transports the aluminum template workpiece to be inspected to the designated position. Since the feeding input mechanism 1 is located between the output mechanism 3 and the temporary storage mechanism 4, and the inspection and sorting mechanism 2 is located above the three, the three-axis motion component 22 of the inspection and sorting mechanism 2 can drive the inspection transfer component 21 to move flexibly in three-dimensional space. The vision recognition unit on the inspection transfer component 21 first completes the positioning of the inspection point, and then the inspection sensor unit starts to collect various inspection index data of the aluminum template. These data are sent to the device control and processing center in real time through signal transmission. After processing the data, the device control and processing center outputs control signals to drive the three-axis motion component 22 to move, and at the same time controls the workpiece gripping unit 211 on the inspection transfer component 21 to perform corresponding actions. If the aluminum template is determined to be qualified, the workpiece gripping unit 211 grips it and transfers it to the carrier frame 32 of the output mechanism 3. The carrier frame 32 can move along the output guide rail 31 to the qualified product storage area; if it is determined to be unqualified, it is transferred to the temporary storage mechanism 4 for storage.

[0052] This solution achieves automated linkage between detection and sorting through the coordinated operation of various mechanisms and the processing and control of the device control center. The combination of a visual recognition unit and a detection sensor unit enhances detection accuracy, while a three-axis motion assembly 22 ensures flexibility in detection and transfer. This invention effectively solves the problem of incomplete detection coverage of complex surfaces and nodes in existing technologies, achieving automated and high-precision detection and sorting of aluminum templates, improving detection efficiency, and providing data for quality traceability, facilitating the tracking of the causes of problems and missed detections.

[0053] Example 2: This example is based on the foregoing examples. In this example, as... Figure 2 As shown, the three-axis motion assembly 22 consists of an X-axis motion assembly 221, a Y-axis motion assembly 222, and a Z-axis motion assembly 223. These motion assemblies are connected via motion rails 224 and motion sliders. The Z-axis motion assembly 223 is slidably mounted on the Y-axis motion assembly 222, the X-axis motion assembly 221 is slidably mounted on the Z-axis motion assembly 223, and the detection and transfer assembly 21 is slidably mounted on the X-axis motion assembly 221. During operation, control signals output from the device's control processing center drive the X, Y, and Z-axis motion assemblies 223 to move. The relative movement between the motion assemblies is achieved through the sliding contact between the motion rails 224 and the motion sliders, thereby displacing the detection and transfer assembly 21 in the X, Y, and Z directions. The working principle of this scheme is to utilize the three-axis linkage structure to expand the motion range of the detection and transfer assembly 21. Simultaneously, the contact structure between the rails and the sliders reduces friction and deviation during movement, ensuring motion accuracy.

[0054] This embodiment further improves the movement flexibility and positioning accuracy of the detection transfer component 21, ensuring that the visual recognition unit and the detection sensor unit can accurately reach the preset detection point, effectively covering all key parts of the aluminum template, avoiding detection blind spots caused by movement deviation, and significantly improving the overall detection accuracy.

[0055] Example 3: This example is based on the foregoing examples. In this example, as... Figure 2 , Figure 3 As shown, the detection transfer assembly 21 is equipped with a spacing adjustment slide rail 212. Multiple spacing adjustment slider frames 213 are slidably connected to the spacing adjustment slide rail 212. Workpiece gripping units 211 are corresponding in number to the spacing adjustment slider frames 213 and are mounted on them. Multiple vision recognition units and detection sensor units are also respectively mounted on the spacing adjustment slider frames 213. The workpiece gripping unit 211 is equipped with a gripping suction cup, which is used to connect to an external vacuum drive mechanism. During operation, according to the size of the aluminum template to be inspected and the distribution requirements of the detection points, the device control processing center controls the spacing adjustment slider frames 213 to slide along the spacing adjustment slide rail 212, thereby adjusting the spacing between the vision recognition unit, the detection sensor unit, and the workpiece gripping unit 211 to adapt to the inspection of aluminum templates of different specifications. When gripping the aluminum template, the vacuum drive mechanism is activated to generate negative pressure suction, achieving stable suction of the aluminum template workpiece through the gripping suction cup. Among them, the spacing adjustment slide rail 212 and the spacing adjustment slider frame 213 are conventional sliding adjustment structures. The smooth sliding is achieved by the cooperation of the slide rail and the slider, thereby adjusting the spacing. Vacuum adsorption uses the principle of negative pressure to make the suction cup and the surface of the aluminum template fit tightly to achieve gripping.

[0056] The solution in this embodiment enhances the versatility of the device, making it adaptable to the detection of aluminum templates of different sizes. The setup of multiple visual recognition units and detection sensor units further improves the detection coverage and efficiency. The vacuum adsorption method of the gripping suction cup can avoid damage to the surface of the aluminum template during the gripping process, ensuring the quality of the aluminum template and avoiding losses caused by damage to the aluminum template during the detection and sorting process.

[0057] Example 4: This example is based on the previous examples. In this example, the detection sensor unit includes multiple sensors, each performing different detection functions. During operation, the laser contour sensor scans the complex surface and key nodes of the aluminum template workpiece to capture its three-dimensional contour information; the eddy current sensor moves along the surface of the aluminum template to identify defects such as near-surface micro-cracks and material uniformity deviations; the laser thickness sensor moves along a preset path to collect thickness data at different positions of the aluminum template in real time, detecting thickness uniformity and local deformation; the high-resolution area array vision sensor, in conjunction with the vision recognition unit, captures image information of minor scratches, pinholes, and other defects on the surface of the aluminum template; the pressure sensing module is integrated on the workpiece gripping unit 211 to monitor gripping pressure in real time to ensure gripping stability; and the tilt sensor collects posture data in real time during the aluminum template loading and inspection process for posture calibration. This solution utilizes the collaborative work of multiple sensors to comprehensively collect multi-dimensional detection indicators such as aluminum template size, surface morphology, and material properties. The detection data from each sensor complement each other to form a complete detection dataset, effectively solving the problem that single detection methods in existing technologies cannot fully cover complex detection needs. It can improve the accuracy of detecting minute defects and dimensional deviations, significantly enhancing the comprehensiveness and precision of detection. This provides more accurate data for subsequent detection data processing and result judgment, thereby improving the accuracy of sorting and judgment.

[0058] Laser profilometry sensor: GOCATOR 2520, which achieves optimal performance on both dark and mirror-like targets, supports multi-sensor measurement networks, and is suitable for 3D profilometry scanning of complex surfaces on aluminum templates.

[0059] Eddy current sensor: SBBR-026-300-032, can be used to detect micro-cracks and material uniformity near the surface of aluminum templates. It can detect through thin coatings and has a frequency range of 0.5-50kHz.

[0060] Laser thickness sensor: optoNCDT 1420, based on the triangulation principle, non-contact measurement, high accuracy, fast response speed, suitable for detecting thickness uniformity and local deformation of aluminum templates, with flexible measurement range.

[0061] High-resolution area array vision sensor: MV-CA003-20GC: suitable for detecting minute scratches and pinhole defects on aluminum template surfaces, and can be used with a vision recognition unit to capture high-precision image information.

[0062] Pressure sensing module: BOTA MiniONE, which can monitor gripping pressure in real time, ensuring stable gripping of aluminum templates and preventing overpressure damage, suitable for precision assembly and testing scenarios.

[0063] Tilt sensor: ENX4461, dual-axis measurement, high precision, fast response, can detect posture deviations in aluminum template loading and inspection processes in real time, providing accurate posture calibration data for three-axis motion components and ensuring consistency of inspection benchmarks.

[0064] Example 5: This example describes a detection and sorting method suitable for high-precision aluminum templates, such as... Figure 4 As shown, the process begins with loading, positioning, and pre-processing of the aluminum template. After the workpiece is transported to a preset position, attitude calibration and surface cleaning and drying are completed. Next, a multi-dimensional data acquisition step is initiated. Through detection area positioning, detection unit spacing adjustment, and detection by multiple sensors, various detection index data of the aluminum template are collected from multiple perspectives. Subsequently, algorithmic data processing is performed, including preprocessing of the collected raw data, optimization of complex surface detection coverage, and enhancement of micro-defect identification, generating detection judgment results and sorting decision instructions. Based on the sorting decision instructions, the sorting execution steps are carried out. Through grasping, transferring, and classifying placement operations, qualified and unqualified aluminum templates are transferred to the output mechanism and temporary storage mechanism, respectively. In the detection data recording and scheduling step, historical detection and sorting data are stored, and unqualified aluminum template data from the temporary storage mechanism is scheduled and recorded. The device control and processing center dynamically adjusts the detection and sorting strategy based on this data.

[0065] This embodiment's method constructs a fully automated mechanism encompassing preprocessing, detection, processing, sorting, and feedback optimization. Through process design, it ensures the orderly execution of detection and sorting, and through data feedback, it achieves dynamic optimization of the detection and sorting strategy. This implementation achieves full automation of aluminum template detection and sorting, effectively solving the problems of insufficient detection accuracy and precision in existing technologies, improving detection and sorting efficiency, ensuring the integrity of quality traceability, and meeting the detection requirements of high-precision aluminum templates.

[0066] Example 6: This example is based on the previous examples. In this example, the loading, positioning, and pretreatment steps are specifically divided into three stages: loading and conveying, attitude calibration, and surface pretreatment. During operation, the loading conveyor belt of the loading input mechanism is first activated to transport the aluminum template workpiece to be inspected to the preset initial inspection position, completing the loading and conveying. Subsequently, the three-axis motion component of the inspection and sorting mechanism drives the inspection transfer component to move. The tilt sensor on the inspection transfer component collects the attitude data of the aluminum template workpiece in real time and transmits the data to the device control and processing center. Based on the attitude deviation data, the device control and processing center drives the three-axis motion component to move the relevant components to calibrate the horizontal attitude of the aluminum template workpiece against the inspection benchmark, completing the attitude calibration. Finally, the cleaning module sprays gas to remove dust, oil, and other impurities from the surface of the aluminum template workpiece, and then the hot air drying module is activated to dry the moisture in the aluminum template workpiece, completing the surface pretreatment.

[0067] This embodiment achieves orderly supply of aluminum templates through feeding and conveying, ensures the consistency of detection benchmarks through posture calibration, and eliminates the interference of impurities and moisture on subsequent detection through surface pretreatment. This effectively improves the accuracy of subsequent detection steps, avoids detection errors caused by aluminum template posture deviations or surface impurities, improves the accuracy of multi-dimensional detection data collection, and further ensures the overall quality of detection and sorting.

[0068] Example 7: This example is based on the previous examples. In this example, the multi-dimensional detection data acquisition steps specifically include four stages: detection area positioning, spacing adjustment and adaptation, multi-sensor detection, and pre-grabbing state confirmation. During operation, the device control and processing center first generates preset detection point coordinates based on the model parameters of the aluminum template workpiece to be inspected. Then, it drives the three-axis motion assembly to move the detection transfer assembly along the X, Y, and Z axes. The vision recognition unit on the detection transfer assembly captures the edges and key feature points of the aluminum template workpiece, and performs secondary correction on the preset detection coordinates to complete the detection area positioning. According to the size of the aluminum template workpiece and the distribution of detection points, the spacing adjustment slide rail is controlled to drive multiple spacing adjustment sliders to slide, adjusting the spacing between the vision recognition unit, the detection sensor unit, and the workpiece gripping unit to adapt to the detection requirements of different areas, thus completing the spacing adjustment and adaptation. Afterward, each sensor starts detection synchronously. The laser contour sensor scans complex surfaces and key nodes to collect three-dimensional contour data, the eddy current sensor detects near-surface defects and material uniformity, the laser thickness sensor collects thickness data, and the high-resolution area array vision sensor captures surface defect image information. All detection data is transmitted to the device control and processing center in real time to complete multi-sensor detection. Finally, the gripping pressure value of the workpiece gripping unit is detected by the pressure sensing module, and at the same time, the tilt sensor collects the posture data of the aluminum template workpiece again to confirm that there is no posture deviation of the aluminum template workpiece during the detection process, thus completing the pre-grip state confirmation.

[0069] This embodiment ensures comprehensive detection coverage through more precise detection area positioning and spacing adjustment, achieves multi-dimensional and complete detection data through multi-sensor collaborative acquisition, and ensures the stability of subsequent grasping operations through pre-grabbing state confirmation. It effectively solves the problem of blind spots in the detection of complex nodes in the prior art, improves the integrity and reliability of detection data, and provides more accurate and reliable data for subsequent data processing and sorting execution.

[0070] Example: This example is based on the previous example. In this example, the algorithm detection data processing steps specifically include four stages: data preprocessing, complex surface detection coverage optimization, micro-defect identification enhancement, and detection result judgment. During operation, the device control and processing center first integrates the raw data collected by each sensor, removes invalid data caused by environmental interference and sensor errors through data noise reduction processing, and establishes a complete detection dataset for a single aluminum template workpiece, completing data preprocessing. Then, a surface contour matching algorithm is used to analyze the complex area data collected by the laser contour sensor, fill in the detection blind zone data caused by surface structure occlusion, and generate complete aluminum template workpiece surface model data, completing complex surface detection coverage optimization. Next, a defect feature enhancement algorithm is used to process the visual image and eddy current detection data, extract the grayscale features, morphological features, and material anomaly features of micro-defects, compare them with the preset standard defect threshold, identify and classify micro-defects, and complete micro-defect identification enhancement. Finally, the device control and processing center comprehensively compares the processed detection data with preset parameters to determine whether the aluminum template workpiece meets the requirements, marks the qualified level and unqualified defect type, generates corresponding sorting decision instructions, and completes the detection result judgment.

[0071] Specifically, the surface contour matching and completion algorithm can be implemented as follows:

[0072] Z'(x,y) = Z(x,y) + λ·∇²Z(x,y);

[0073] in:

[0074] Z'(x,y): The height value of the completed aluminum template surface contour;

[0075] Z(x,y): The original contour height value acquired by the laser contour sensor;

[0076] λ: Contour smoothing coefficient, set according to the surface roughness of the aluminum template, with a value range of 0.02-0.05;

[0077] ∇²: Laplacian operator, used to calculate the second derivative of the original contour data, identify the contour trend of the detection blind zone, and realize the completion of blind zone data.

[0078] The surface contour matching and completion algorithm calculates the second derivative of the original contour height value Z(x,y) using the Laplacian operator. This allows for more accurate identification of contour variation trends on complex surfaces and key nodes of aluminum templates. Combined with a contour smoothing coefficient λ (0.02-0.05) set according to surface roughness, it completes the detection blind zone data. During the algorithm's calculation, the Laplacian operator's second derivative analysis of the original data captures contour curvature changes and accurately determines the contour extension direction of occluded areas. The adaptive setting of the λ coefficient avoids contour distortion caused by excessive smoothing while completing the data. This process completes the incomplete original data collected by the laser contour sensor into a complete surface contour height value Z'(x,y), effectively covering traditional detection blind zones such as the corners of aluminum templates and the periphery of reserved holes, achieving an upgrade from single-point detection to multi-point coverage in three-dimensional detection. This solution can acquire complete three-dimensional contour information of the entire surface of the aluminum template, solving the problem of incomplete detection coverage of complex surfaces and key nodes, avoiding omissions in three-dimensional detection, and providing comprehensive three-dimensional data support for subsequent judgment and sorting. It significantly reduces the risk of unqualified templates flowing into the next stage due to insufficient detection coverage.

[0079] Specifically, the enhanced algorithm for identifying minute defects can be implemented as follows:

[0080] F(x,y)=[(I(x,y)-I_avg) / σ]·K;

[0081] in:

[0082] F(x,y): Feature value after defect feature enhancement;

[0083] I(x,y): The pixel grayscale value of the detection point;

[0084] I_avg: Average pixel grayscale value within the detection area;

[0085] σ: Standard deviation of pixel grayscale values ​​within the detection area;

[0086] K: Defect contrast enhancement coefficient, set according to the defect type. The value is 1.8-2.2 for scratch detection and 2.0-2.5 for pinhole detection.

[0087] The micro-defect recognition enhancement algorithm first calculates the difference between the grayscale value I(x,y) of the detected pixel and the average grayscale value I_avg of the detection area, then divides it by the standard deviation σ of the grayscale value to normalize the local pixel grayscale features. Finally, it multiplies by a contrast enhancement coefficient K adapted according to the defect type (1.8-2.2 for scratch detection, 2.0-2.5 for pinhole detection) to highlight the feature value F(x,y) of the micro-defect. In this calculation process, the statistical calculation of I_avg and σ can effectively eliminate overall illumination interference and background noise in the detection area, while the differentiated setting of the K coefficient can specifically enhance the grayscale feature differences of different types of micro-defects, making the feature values ​​of micro-defects such as scratches and pinholes significantly distinguishable from normal areas. At the same time, the complete 3D contour data output by the surface contour matching and completion algorithm and the defect feature data output by the micro-defect recognition enhancement algorithm form multimodal detection information, which are combined for comprehensive evaluation. This algorithm, through multimodal detection information processing, has higher recognition and reliability. It can more accurately distinguish between qualified templates, templates with minor defects that can be repaired, and templates with serious defects that are unqualified. It avoids misjudgment caused by single detection information, greatly improves the accuracy of defect classification and qualification judgment, makes sorting decision instructions more targeted, achieves high-precision classification and sorting, and meets the stringent quality requirements of aluminum templates in high-demand building scenarios.

[0088] In conjunction with the technical solution of this embodiment, the surface contour matching and completion algorithm and the micro-defect recognition and enhancement algorithm achieve effective identification and correction of error factors through parameter adaptation and data processing logic. In the surface contour completion calculation, the λ coefficient is dynamically adjusted according to the surface roughness of the aluminum template, which can smooth the fluctuations of the original data caused by surface reflection and sensor noise. The calculation of the second derivative by the Laplacian operator can identify and correct the misjudgment of the contour trend caused by the sensor angle deviation, ensuring that the completed Z'(x,y) is closer to the actual surface state of the template. In the micro-defect recognition and enhancement calculation, the calculation process of I_avg and σ can identify and eliminate the abnormal grayscale values ​​caused by interference factors such as changes in ambient light and residual oil on the surface. The typological setting of the K coefficient corrects the recognition deviation caused by the inherent differences in grayscale features of different defect types, so that F(x,y) can truly reflect the actual situation of micro-defects. The computational implementation of the two algorithms forms a dual error correction mechanism, which effectively reduces detection deviations caused by environmental interference, surface characteristics, and sensor errors. This significantly improves the accuracy of detection indicators such as surface flatness and defect size, making the detection results more closely reflect the actual quality state of the aluminum formwork. Based on the corrected detection data, the sorting process can achieve more precise separation of qualified and unqualified formwork. Simultaneously, the device control and processing center dynamically adjusts the sorting strategy based on accurate detection data, further enhancing the stability and reliability of the detection and sorting system. This ensures that the output aluminum formwork meets the requirements for construction accuracy and structural safety in scenarios such as high-rise buildings and prefabricated buildings.

[0089] Example 9: This example is based on the previous example. In this example, the sorting execution steps specifically include three stages: aluminum template workpiece gripping, transfer positioning, and classification placement. During operation, according to the sorting decision command, the three-axis motion assembly drives the detection and transfer assembly to move to the gripping position of the aluminum template workpiece. The workpiece gripping unit starts to grip the aluminum template workpiece, and the pressure sensing module monitors the gripping pressure value in real time to ensure that the gripping force is moderate, thus completing the gripping of the aluminum template workpiece. Subsequently, according to the sorting decision command, the device control and processing center drives the three-axis motion assembly to move the gripped aluminum template workpiece. The vision recognition unit assists in positioning the output guide rail and the carrier frame of the output mechanism, and at the same time confirms the idle station information of the temporary storage mechanism to complete the transfer positioning. If the aluminum template workpiece is determined to be qualified, it is transferred to the carrier frame of the output mechanism, and the carrier frame moves along the output guide rail to the corresponding qualified product storage area; if the aluminum template workpiece is determined to be unqualified, it is transferred to the designated station of the temporary storage mechanism for storage. After completing the sorting operation, the workpiece gripping unit releases pressure and resets to the initial detection position, completing the classification placement.

[0090] This solution improves the accuracy and stability of aluminum template grasping and transfer by driving a three-axis motion component, ensures positioning accuracy with the help of a vision recognition unit, and classifies and places qualified and unqualified products according to sorting decision instructions. This achieves automation and precision in aluminum template sorting, avoiding the problems of low efficiency and high error rate caused by manual sorting, and improving sorting efficiency and stability.

[0091] Example 10: This example is based on the previous examples. In this example, the detection data recording and scheduling steps specifically include three stages: data storage, temporary storage scheduling, and equipment linkage optimization. During operation, the device control and processing center classifies and stores the detection data, defect identification results, and sorting results of each aluminum template workpiece, establishing a complete quality traceability database and completing data storage. Then, non-conforming aluminum template workpieces in the temporary storage mechanism are marked with information, and the defect type and location are recorded in detail. A non-conforming product handling strategy is generated, and the occupancy status of the temporary storage mechanism is updated in real time to facilitate subsequent non-conforming product handling and scheduling, completing temporary storage scheduling. Finally, based on historical detection data and sorting efficiency, the device control and processing center dynamically adjusts the feeding conveyor speed, detection point distribution, and sorting execution rhythm to achieve equipment linkage optimization. This solution establishes a quality traceability system through data storage, achieves orderly management of non-conforming products through temporary storage scheduling, and achieves dynamic optimization of equipment operating parameters through historical data analysis. This implementation plan ensures the integrity of quality traceability, facilitates subsequent tracking and analysis of product quality, improves production management efficiency through the orderly scheduling of non-conforming products, and further enhances the efficiency and accuracy of overall testing and sorting by optimizing equipment linkage, thereby achieving continuous optimization of equipment operating status.

[0092] The embodiments described above are merely illustrative of specific implementations of the present invention, and while the descriptions are detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention.

Claims

1. A method for detecting and sorting a high-precision aluminum template, characterized in that, Includes the following steps: S100, feeding, positioning and pretreatment, transports the aluminum template workpiece to be inspected to the preset position, performs posture calibration and surface cleaning and drying; S200, multi-dimensional detection data acquisition, through detection area positioning, detection unit spacing adjustment and multiple sensor detection, acquires detection index data of aluminum template workpiece; S200 specifically includes: S201. Detection area positioning: The device control and processing center generates preset detection point coordinates based on the model parameters of the aluminum template workpiece to be detected, drives the three-axis motion component to move the detection transfer component along the X, Y, and Z axes, and captures the edge and key feature points of the aluminum template workpiece through the vision recognition unit to perform secondary correction on the preset detection coordinates. S202, Spacing adjustment adaptation: Based on the size of the aluminum template workpiece and the distribution of detection points, control the spacing adjustment slide rail to drive multiple spacing adjustment slider frames to slide, adjust the spacing of the vision recognition unit, detection sensor unit and workpiece gripping unit, and detect multiple different areas of the aluminum template workpiece. S203, multi-sensor detection: A laser contour sensor scans the complex surface and key nodes of the aluminum template workpiece to collect three-dimensional contour data; an eddy current sensor moves along the surface of the aluminum template workpiece to detect near-surface micro-cracks and material uniformity; a laser thickness sensor moves along a preset path on the aluminum template workpiece to collect thickness data at different locations in real time; and a high-resolution area array vision sensor captures images of surface micro-scratches and pinhole defects; the detection data from multiple sensors is transmitted to the device control and processing center in real time. S204. Pre-grabbing status confirmation: The gripping pressure value of the workpiece gripping unit is detected by the pressure sensing module, and at the same time, the tilt sensor collects the aluminum template workpiece posture data again to confirm that the aluminum template workpiece has no posture deviation during the detection process. S300 algorithm detection data processing: preprocesses the collected raw detection data, optimizes the detection coverage of complex surfaces and enhances the identification of minor defects, and generates detection judgment results and sorting decision instructions; S400, Sorting Execution: Based on the sorting decision instructions, qualified and unqualified aluminum template workpieces are transferred to the output mechanism and temporary storage mechanism respectively through grabbing, transferring and classifying placement operations. S500: Detection data recording and scheduling; storage of historical detection and sorting data; recording of non-conforming aluminum template workpiece data in the scheduling temporary storage mechanism; and dynamic adjustment of detection and sorting strategies by the device control and processing center based on the non-conforming aluminum template workpiece data.

2. The detection and sorting method for high-precision aluminum templates according to claim 1, characterized in that, Specifically, S100 includes: S101. Feeding and conveying: Start the feeding conveyor belt of the feeding input mechanism to transport the aluminum template workpiece to be inspected to the preset initial inspection position; S102, Attitude calibration: The three-axis motion component of the detection and sorting mechanism drives the detection and transfer component to move. Real-time attitude data is collected by the tilt sensor on the detection and transfer component and transmitted to the device control and processing center. The device control and processing center drives the three-axis motion component to calibrate the horizontal attitude of the aluminum template workpiece with the detection benchmark based on the attitude deviation data. S103. Surface pretreatment: The cleaning module sprays gas to remove dust and oil stains from the surface of the aluminum template workpiece, and the hot air drying module is started to dry the moisture.

3. The detection and sorting method for high-precision aluminum templates according to claim 1, characterized in that, Specifically, S300 includes: S301. Data preprocessing: The device control and processing center integrates the raw data collected by each sensor, removes environmental interference and sensor error data through data noise reduction processing, and establishes a complete detection dataset for a single aluminum template workpiece. S302. Complex surface detection coverage optimization: The surface contour matching algorithm is used to analyze the complex area data collected by the laser contour sensor, fill in the detection blind area data caused by surface structure occlusion, and generate complete aluminum template workpiece surface model data. S303. Enhanced identification of minor defects: The defect feature enhancement algorithm is used to process visual images and eddy current detection data, extract the grayscale features, morphological features and material anomaly features of minor defects, compare them with the preset standard defect threshold, and identify and classify minor defects. S304. Detection Result Judgment: The device control and processing center will comprehensively compare the processed detection data with the preset parameters to determine whether the aluminum template workpiece meets the requirements, mark the qualified level and the unqualified defect type, and generate the corresponding sorting decision instruction.

4. The detection and sorting method for high-precision aluminum templates according to claim 1, characterized in that, Specifically, S400 includes: S401, Aluminum template workpiece gripping: According to the sorting decision instruction, the three-axis motion component drives the detection and transfer component to move to the aluminum template workpiece gripping position, the workpiece gripping unit grips the aluminum template workpiece, and the pressure sensing module monitors the pressure value in real time. S402, Transfer Positioning: The control and processing center drives the three-axis motion assembly to move the gripped aluminum template workpiece according to the sorting decision instructions. The vision recognition unit assists in positioning the output guide rail and carrier frame of the output mechanism, and at the same time confirms the idle station information of the temporary storage mechanism. S403. Classify and place the workpieces. If the aluminum template workpieces are deemed qualified, they are transferred to the support frame of the output mechanism. The support frame moves along the output guide rail to the corresponding qualified product storage area. If the aluminum template workpieces are deemed unqualified, they are transferred to the designated workstation of the temporary storage mechanism for storage. After the sorting operation is completed, the workpiece gripping unit releases pressure and resets to the initial detection position.

5. The detection and sorting method for high-precision aluminum templates according to claim 1, characterized in that, The S500 specifically includes: S501. Data storage: The device control and processing center classifies and stores the inspection data, defect identification results, and sorting results of each aluminum template workpiece to establish a complete quality traceability database. S502, Temporary storage and scheduling: Mark the information of unqualified aluminum template workpieces in the temporary storage mechanism, record the defect type and location, generate a non-conforming product handling strategy, and update the occupancy status of the temporary storage mechanism workstations in real time. S503, Equipment linkage optimization: Based on historical detection data and sorting efficiency, the device control and processing center dynamically adjusts the feeding and conveying speed, detection point distribution, and sorting execution rhythm.

Images