Checking and rejecting device and checking and rejecting method under high-speed condition

Abstract

The invention relates to the technical field of intelligent manufacturing, in particular to a checking and rejecting device and method under the high-speed condition, and the device comprises a conveying belt, a code reading triggering photoelectric sensor, a depth camera, a code reader, a rejecting assembly, a rejecting detection photoelectric sensor, an encoder and a PLC. The accurate position of the bottle cap is obtained through the depth camera, and the code reader is controlled to be aligned with the bottle cap, so that the accuracy of two-dimensional code reading is ensured; the vibration sensor and the light sensor are introduced to detect the change of environment vibration and illumination intensity, and the sensitivity of the photoelectric sensor and the trigger delay of the rejecting device are dynamically adjusted through the PLC, so that high-precision two-dimensional code verification and accurate rejecting of unqualified products under a high-speed condition are realized, and the production efficiency and the product quality are remarkably improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent manufacturing technology, and in particular to a verification and rejection device and a verification and rejection method under high-speed conditions. Background Art

[0002] With the rapid development of industrial automation, high-speed production lines have been widely used in the packaging, food and beverage industries. On these production lines, product quality inspection and rejection of defective products are key links to ensure product qualification. Existing inspection and rejection systems are usually based on photoelectric sensors and PLC control, using trigger signals to verify identification information such as QR codes, and remove defective products from the conveyor belt through rejection devices.

[0003] At present, the accuracy of QR code verification and the success rate of rejection are affected by many factors under high-speed conditions, such as vibration of the conveyor belt, changes in ambient light, delays in sensor signals, etc. In addition, existing systems usually have difficulty in real-time adjusting the sensitivity of photoelectric sensors or the trigger logic of rejection devices to adapt to complex environmental changes. This lack of intelligent dynamic adjustment capabilities may result in the inability to correctly read the QR code or the failure of rejection, which in turn affects production efficiency and product quality. Summary of the invention

[0004] The present invention aims to solve the problems of missed rejection, instability, poor environmental adaptability, etc. when processing defective products on a high-speed production line. The purpose is to provide a verification and rejection device and a verification and rejection method under high-speed conditions, which realizes the functions of two-dimensional code verification and accurate rejection of defective products for bottle caps conveyed at high speed under complex environmental conditions. At the same time, by dynamically adjusting the sensitivity of the photoelectric sensor and the trigger delay of the rejection device, the adaptability, accuracy and reliability of the verification and rejection system are improved.

[0005] The present invention is achieved through the following technical solutions: A high-speed inspection and rejection device, comprising: a conveyor belt, a code reading trigger photoelectric sensor, a depth camera, a code reader, a rejection component, a rejection detection photoelectric sensor, an encoder and a PLC; The code reading trigger photoelectric sensor, the code reader, the rejection component and the rejection detection photoelectric sensor are sequentially arranged along the conveying direction of the conveyor belt, and the conveyor belt is used to convey bottle caps with QR codes; The code reading trigger photoelectric sensor and the rejection detection photoelectric sensor are both arranged on one side of the conveyor belt, and the detection direction is perpendicular to the conveying direction of the bottle cap; the depth camera and the code reader are arranged in parallel and located above the two-dimensional code reading area of ​​the bottle cap; The rejecting assembly applies a lateral force to the bottle caps on the conveyor belt; The conveyor belt, the code reading trigger photoelectric sensor, the depth camera, the code reader, the rejection component, the rejection detection photoelectric sensor and the encoder are all electrically connected to the PLC.

[0006] Furthermore, the device also includes a vibration sensor and a light sensor. The vibration sensor is arranged on the conveyor belt, and the light sensor is arranged near the code reading trigger photoelectric sensor and the rejection detection photoelectric sensor. The vibration sensor and the light sensor are both electrically connected to the PLC.

[0007] A high-speed verification and rejection method is based on the high-speed verification and rejection device as described above, and the verification and rejection method comprises: The code reading trigger photoelectric sensor generates a trigger signal when detecting that the bottle cap passes through the trigger point; According to the trigger signal, the PLC triggers the code reader to read the data of the QR code on the bottle cap; The PLC receives the reading result of the code reader and compares it with the pre-stored data to generate the corresponding verification signal; Determine whether the bottle cap is qualified according to the verification signal. If it is qualified, it will not be rejected; if it is unqualified, a control signal will be applied to the rejection component; The rejection component applies lateral force to the defective products to remove the caps from the conveyor belt; The reject detection photoelectric sensor verifies the reject result and issues an alarm signal or stops the conveyor belt if the rejects are not successfully rejected.

[0008] Specifically, the method of triggering the code reader to read the data of the two-dimensional code on the bottle cap includes: The method for the code reader to read the data of the QR code on the bottle cap includes: Determine the distance from the code reading trigger photoelectric sensor to the code reading area , and convert it into a pulse value ,in, is the pulse resolution of the encoder; After receiving the trigger signal, the PLC records the current encoder count value and uses it as the code reading reference value ; Calculate the difference between the current encoder count value and the target position , ,in, It is the count value fed back by the encoder in real time; Calculate dynamic compensation error parameters , ,in, is the conveyor belt running speed, is the signal delay time of PLC, is the correction of the pulse value by the environmental error, and is the experience weight coefficient; like , the PLC sends a code reading control signal to the code reader.

[0009] Specifically, the precise position of the bottle cap is obtained by using a depth camera, and the code reader is controlled to aim at the bottle cap. The method includes: The depth camera collects the initial depth map of the code reading area ; Perform mean filtering and denoising on the depth values ​​in the depth map to obtain a depth map ; Setting Dynamic Depth Threshold and , ,in, is the median of the depth values ​​of the depth map, is the interquartile range of the depth values ​​of the depth map, is the dynamic adjustment coefficient; Binarize the depth map to obtain the mask map , ,in, Depth map Medium pixel coordinates The depth value of The mask image Medium pixel coordinates The binary depth value of Determine the center of mass of the bottle cap in pixel coordinates , ,in, is the area of ​​the bottle cap in pixel coordinates ; Get the two-dimensional physical coordinates of the bottle cap , ; Get the actual center position of the barcode reader , and calculate the error between the bottle cap position and the actual center position ,like Then read the QR code; otherwise adjust the actual center position of the code reader until it meets ,in, and is the allowable horizontal error range and vertical error range.

[0010] Optionally, the allowed horizontal error range , allowable vertical error range ,in, and is the static basic error range, is the conveyor belt speed, is the depth camera frame rate, is the speed influence coefficient; The method to adjust the actual center position of the code reader is: ,in, is the actual center position of the code reader after adjustment. To adjust the step size factor.

[0011] Specifically, the vibration intensity of the conveyor belt is detected by a vibration sensor, the light intensity near the code reading trigger photoelectric sensor and the rejection detection photoelectric sensor is detected by a light sensor, and when the vibration intensity or light change exceeds a preset threshold, the sensitivity of the photoelectric sensor and the trigger delay of the rejection device are adjusted.

[0012] Optionally, the adjustment method includes: Collect environmental signals and record them in time series ,in, for The vibration intensity at the moment, for Light intensity at the time; Determine the quantitative index of vibration intensity ,in, is the sampling time window; Calculate the relative rate of change of light intensity ,in, and For the past time window The maximum and minimum values ​​of the internal light intensity, is the average light intensity; Dynamic adjustment function to determine the sensitivity of photosensors ,in, is the initial sensitivity of the photoelectric sensor, A coefficient used to control the effect of vibration and light changes on sensitivity; Dynamic adjustment function that determines the triggering delay of the reject device ,in, Delay for initial triggering of the reject device, is used to control the influence of vibration and light changes on the trigger delay. A dynamic prediction model based on long short-term memory network is constructed and and As input, we can get the prediction result ; Adjust sensitivity and trigger delay in advance based on prediction results; .

[0013] Specifically, the method of verifying the rejection result by the rejection detection photoelectric sensor includes: Determine the distance from the reject zone to the reject detection photoelectric sensor , and convert it into a pulse value ,in, is the pulse resolution of the encoder; PLC records the encoder count value when the rejection action occurs , calculate the target count value of the rejection detection position ; Get When the feedback signal of the photoelectric sensor is detected, if no trigger signal is detected, the rejection is judged to be successful; if a trigger signal is detected, the rejection is judged to be a failure.

[0014] Compared with the prior art, the present invention has the following advantages and beneficial effects: The device of the present invention includes a conveyor belt, a code reading trigger photoelectric sensor, a depth camera, a code reader, a rejection component, a rejection detection photoelectric sensor, an encoder and a PLC, and combines a variety of sensing devices to realize the two-dimensional code verification and rejection of high-speed conveyed bottle caps.

[0015] The present invention obtains the precise position of the bottle cap through a depth camera and controls the code reader to align with the bottle cap to ensure the accuracy of QR code reading; by introducing vibration sensors and light sensors to detect changes in environmental vibration and light intensity, and dynamically adjusting the sensitivity of the photoelectric sensor and the trigger delay of the rejection device through PLC, high-precision QR code verification and accurate rejection of defective products under high-speed conditions are achieved, which significantly improves production efficiency and product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings illustrate exemplary embodiments of the present invention and, together with the description thereof, are used to explain the principles of the present invention. These drawings are included to provide a further understanding of the present invention, and the accompanying drawings are included in and constitute a part of this specification and do not constitute a limitation of the embodiments of the present invention.

[0017] Figure 1 It is a schematic structural diagram of a high-speed checking and rejecting device according to the present invention.

[0018] Figure 2 It is a flow chart of a high-speed verification and rejection method according to the present invention.

[0019] Figure numerals: 1- conveyor belt, 2- code reading trigger photoelectric sensor, 3- code reader, 4- rejection component, 5- rejection detection photoelectric sensor, 6- encoder. DETAILED DESCRIPTION

[0020] To make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and implementation methods. It is understood that the specific implementation methods described herein are only used to explain the relevant content, rather than to limit the present invention.

[0021] It should also be noted that, for the convenience of description, only the parts related to the present invention are shown in the drawings.

[0022] In the absence of conflict, the embodiments and features of the embodiments of the present invention may be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0023] Embodiment 1

[0024] like Figure 1 As shown, a high-speed inspection and rejection device includes: a conveyor belt 1, a code reading trigger photoelectric sensor 2, a depth camera, a code reader 3, a rejection component 4, a rejection detection photoelectric sensor 5, an encoder 6 and a PLC; The code reading trigger photoelectric sensor 2, the code reader 3, the rejection component 4 and the rejection detection photoelectric sensor 5 are arranged in sequence along the conveying direction of the conveyor belt 1. The conveyor belt 1 is used to convey the bottle caps with QR codes on the conveyor belt 1; the bottle caps with QR codes are transported to various functional areas in sequence, including the QR code reading area, the rejection area and the rejection detection area.

[0025] The code reading trigger photoelectric sensor 2 and the rejection detection photoelectric sensor 5 are both arranged on one side of the conveyor belt 1, and the detection direction is perpendicular to the conveying direction of the bottle cap; the depth camera and the code reader 3 are arranged in parallel and located above the two-dimensional code reading area of ​​the bottle cap; The rejecting component 4 applies a lateral force to the bottle caps on the conveyor belt 1 to reject them from the conveyor belt 1 .

[0026] The conveyor belt 1, the code reading trigger photoelectric sensor 2, the depth camera, the code reader 3, the rejection component 4, the rejection detection photoelectric sensor 5 and the encoder 6 are all electrically connected to the PLC.

[0027] The code reading trigger photoelectric sensor 2 is installed on one side of the conveyor belt 1 to detect the passage of the bottle cap vertically. When the bottle cap reaches the QR code reading area, the photoelectric sensor generates a trigger signal to notify the code reader 3 to read the QR code. The depth camera is used to obtain the three-dimensional position of the bottle cap in real time, assist the code reader 3 to align the bottle cap position, and improve the accuracy of QR code reading. The code reader 3 is used to read the QR code information on the bottle cap and send the data to the PLC for comparison and verification.

[0028] The rejection detection photoelectric sensor 5 is arranged behind the rejection component 4 to detect whether there are still bottle caps in the rejection area: if no signal is detected, it indicates that the rejection is successful. If a signal is detected, it indicates that the rejection fails, and the PLC will issue an alarm signal or stop the conveyor belt 1.

[0029] The encoder 6 is installed on the conveyor belt 1 to record the moving position of the conveyor belt 1 in real time and generate a pulse signal for the PLC to calculate the precise timing of triggering and rejecting.

[0030] The PLC serves as the core control system, connected to all sensors, cameras, barcode readers 3 and rejection components 4, coordinating the entire verification and rejection process.

[0031] The device also includes a vibration sensor and a light sensor. The vibration sensor is arranged on the conveyor belt 1 to detect the running vibration of the conveyor belt 1. The light sensor is arranged near the code reading trigger photoelectric sensor 2 and the rejection detection photoelectric sensor 5 to monitor the change of the ambient light intensity. Both the vibration sensor and the light sensor are electrically connected to the PLC.

[0032] The workflow is briefly described as follows: The bottle cap is transported to the detection range of the code reading trigger photoelectric sensor 2 through the conveyor belt 1. After the trigger signal is generated, the depth camera and the code reader 3 work together to complete the QR code reading.

[0033] The PLC compares the two-dimensional code information of the code reader 3 with the pre-stored data to determine whether the bottle cap is a defective product.

[0034] For unqualified bottle caps, the PLC controls the rejection component 4 to apply a lateral force to remove the bottle caps from the conveyor belt 1.

[0035] The rejection detection photoelectric sensor 5 verifies the rejection result. If the rejection fails (cap signal is detected), the PLC triggers an alarm or stops the conveyor belt 1.

[0036] The vibration sensor and light sensor detect the vibration and light intensity of the conveyor belt 1 in real time. The PLC adjusts the sensitivity of the photoelectric sensor and the trigger delay of the rejection device according to the detection results to ensure stable operation of the equipment.

[0037] Embodiment 2

[0038] like Figure 2 As shown, a verification and rejection method under high-speed conditions is provided. Based on the above verification and rejection device under high-speed conditions, the verification and rejection method includes: The code reading trigger photoelectric sensor generates a trigger signal when it detects that the bottle cap passes the trigger point; this signal is received by the PLC and used as the trigger condition for subsequent control logic.

[0039] According to the trigger signal, the PLC triggers the code reader to read the data of the QR code on the bottle cap; reads the QR code information on the bottle cap and transmits the reading result back to the PLC.

[0040] The PLC receives the reading result of the code reader and compares it with the pre-stored data (qualified information) to generate the corresponding verification signal; Determine whether the bottle cap is qualified according to the verification signal. If it is qualified, it will not be rejected; if it is unqualified, a control signal will be applied to the rejection component; The rejection component applies lateral force to the defective products to remove the caps from the conveyor belt; The rejection result is verified by the rejection detection photoelectric sensor, and an alarm signal is issued or the conveyor belt is stopped when the defective products are not successfully rejected. If no bottle cap signal is detected, it indicates that the rejection is successful; if the bottle cap signal is still detected, it is judged as a failure.

[0041] Embodiment 3

[0042] The trigger signal is generated by the photoelectric sensor through code reading, and the position information and dynamic compensation error parameters fed back by the encoder are combined to accurately determine the time when the bottle cap reaches the QR code reading area, and control the PLC to send a signal to the code reader to ensure the accuracy and timeliness of the QR code reading. The method of triggering the code reader to read the QR code on the bottle cap includes: Determine the distance from the code reading trigger photoelectric sensor to the code reading area , and convert the distance into pulse value ,in, is the pulse resolution of the encoder; When the code reading trigger photoelectric sensor detects that the bottle cap passes the trigger point, the PLC receives the trigger signal and records the current encoder count value, which is used as the code reading reference value. , which means the position of the bottle cap when it passes the trigger point, and is used to calculate the target position.

[0043] The target position is the count value of the code reading area, calculated as follows: : Calculate the difference between the current encoder count value and the target position , ,in, It is the count value fed back by the encoder in real time; Calculate dynamic compensation error parameters to adapt to the effects of conveyor speed fluctuations, signal delays and environmental interference in high-speed production environments , ,in, is the conveyor belt running speed, is the signal delay time of PLC, Correction of pulse value by environmental error (such as error caused by vibration, temperature change, etc.). and is the empirical weight coefficient used to adjust the contribution of speed and delay to the error.

[0044] judge Whether to dynamically compensate error parameters Within the range, if , the PLC sends a code reading control signal to the code reader.

[0045] Embodiment 4

[0046] This embodiment uses a depth camera to obtain the precise position of the bottle cap, calculates the two-dimensional physical coordinates of the bottle cap, compares and adjusts them with the actual center position of the code reader, ensures that the code reader is aligned with the bottle cap, and realizes accurate reading of the two-dimensional code. The method includes key steps such as depth camera calibration, depth map processing, coordinate calculation and error adjustment. The specific method includes: Use the calibration board to calibrate the depth camera's intrinsic parameters and obtain the camera's intrinsic parameter matrix , ,in, is the focal length of the depth camera, in pixels, is the main offset of the depth camera, which is used to correct the offset between the optical axis and the image center.

[0047] The depth camera collects the initial depth map of the code reading area .

[0048] Perform mean filtering and denoising on the depth values ​​in the depth map to obtain a depth map ,in, is the pixel coordinate Neighborhood window of Calculate the dynamic depth threshold based on the median and interquartile range of the denoised depth map and , ,in, is the median of the depth values ​​of the depth map, is the interquartile range of the depth values ​​of the depth map, It is a dynamic adjustment coefficient.

[0049] Binarize the depth map to obtain the mask map , ,in, Depth map Medium pixel coordinates The depth value of The mask image Medium pixel coordinates The binary depth value of Determine the center of mass of the bottle cap in pixel coordinates Determine the center of mass of the bottle cap in pixel coordinates , ,in, is the area of ​​the bottle cap in pixel coordinates ; Get the two-dimensional physical coordinates of the bottle cap , ; Get the actual center position of the barcode reader , and calculate the error between the bottle cap position and the actual center position ,like Then read the QR code; otherwise adjust the actual center position of the code reader until it meets ,in, and is the allowable horizontal error range and vertical error range.

[0050] Allowable horizontal error range , allowable vertical error range ,in, and is the static basic error range, is the conveyor belt speed, is the depth camera frame rate, is the speed influence coefficient; as the conveyor belt speed increases, the horizontal error range increases accordingly to adapt to the dynamic offset of the bottle cap when running at high speed, thereby improving the alignment tolerance of the barcode reader to the bottle cap.

[0051] When the error between the bottle cap position and the barcode reader position exceeds the allowable range, the actual center position of the barcode reader needs to be adjusted. The method for adjusting the actual center position of the barcode reader is as follows: ,in, is the actual center position of the code reader after adjustment. To adjust the step size coefficient, it is used to control the amplitude of each adjustment to ensure a smooth and accurate adjustment process.

[0052] By adjusting the position of the barcode reader, the alignment error caused by the dynamic offset of the bottle cap is compensated to ensure that the barcode reader can accurately cover the QR code area on the bottle cap.

[0053] Embodiment 5

[0054] The vibration intensity of the conveyor belt is detected by the vibration sensor, and the light intensity near the code reading trigger photoelectric sensor and the rejection detection photoelectric sensor is detected by the light sensor. When the vibration intensity or light change exceeds the preset threshold, the sensitivity of the photoelectric sensor and the trigger delay of the rejection device are adjusted.

[0055] Specific adjustment methods include: Collect environmental signals and record them in time series ,in, for The vibration intensity at the moment, for Light intensity at the time; Quantify vibration intensity using root mean square (RMS) ,in, is the sampling time window; Used to dynamically adjust the sensitivity of photoelectric sensors.

[0056] In order to describe the magnitude of the light change, the relative rate of change of light intensity is calculated ,in, and For the past time window The maximum and minimum values ​​of the internal light intensity, is the average light intensity.

[0057] Dynamic adjustment function to determine the sensitivity of photosensors ,in, is the initial sensitivity of the photoelectric sensor, It is used to control the influence of vibration and light changes on sensitivity. The higher the vibration intensity, the lower the sensitivity exponentially, thus reducing the possibility of false triggering.

[0058] Dynamic adjustment function that determines the triggering delay of the reject device ,in, Delay for initial triggering of the reject device, It is used to control the influence of vibration and light changes on the trigger delay. The greater the light fluctuation, the longer the trigger delay will be, thus avoiding false elimination caused by changes in ambient light.

[0059] Embodiment 6

[0060] This method constructs a dynamic prediction model based on the long short-term memory network (LSTM) to quantify the vibration intensity index. and the relative rate of change of light intensity The system can learn and predict historical data to obtain the predicted vibration intensity and light intensity at future moments. According to the prediction results, the sensitivity of the photoelectric sensor and the trigger delay of the rejection device are adjusted in advance to improve the system's adaptability and real-time performance to environmental changes.

[0061] Construct a dynamic prediction model based on long short-term memory network and and The historical values ​​of are used as input.

[0062] The network structure includes input layer, LSTM hidden layer and output layer. and The LSTM hidden layer learns the temporal correlation of historical data through memory cells; the output layer predicts the future moment and .

[0063] The LSTM network is trained using historical vibration intensity and light change rate data to minimize the predicted value. With the true value of error.

[0064] Prediction model based on time series As input, get the prediction result ; According to the predicted and Adjust sensitivity and trigger delay in advance; .

[0065] As time goes by, vibration and light data are continuously updated, allowing the LSTM network to learn new environmental patterns in real time, ensuring the accuracy of the prediction results and the robustness of the model.

[0066] Compare the error between the predicted value and the true value, and use the back propagation algorithm to update the network parameters to further improve the prediction accuracy.

[0067] Embodiment 7

[0068] The method of verifying the rejection result through the rejection detection photoelectric sensor determines whether the bottle cap is successfully rejected from the conveyor belt based on the count value of the encoder and the feedback signal of the rejection detection photoelectric sensor. It can accurately determine the rejection effect and trigger an alarm or further processing when the rejection fails, ensuring the stability and accuracy of the system operation. The methods of verifying the rejection result through the rejection detection photoelectric sensor include: Determine the distance from the reject zone to the reject detection photoelectric sensor , which is a fixed physical parameter between the rejection component and the detection area. And convert it into a pulse value ,in, is the pulse resolution of the encoder; At the moment when the rejection component applies lateral force to the bottle cap, the PLC records the encoder count value when the rejection action occurs. , calculate the target count value of the rejection detection position ; When the encoder count value Reach target count value When the PLC removes the photoelectric sensor, it obtains the feedback signal. When the feedback signal of the photoelectric sensor is detected, the bottle cap is rejected. If no trigger signal is detected, it means that the bottle cap has been successfully rejected from the conveyor belt. If a trigger signal is detected, it is judged that the rejection has failed, indicating that the bottle cap is still on the conveyor belt and the rejection has failed.

[0069] When rejection fails, the PLC immediately triggers corresponding measures: triggering an alarm signal to alert the operator; or stopping the conveyor belt to prevent unqualified products from entering subsequent production links.

[0070] In the description of this specification, the description with reference to the terms "one embodiment / method", "some embodiments / methods", "example", "specific example", or "some examples" etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment / method or example are included in at least one embodiment / method or example of the present application. In this specification, the schematic representations of the above terms do not necessarily refer to the same embodiment / method or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments / methods or examples in a suitable manner. In addition, those skilled in the art may combine and combine the different embodiments / methods or examples described in this specification and the features of the different embodiments / methods or examples, unless they are contradictory.

[0071] In addition, the terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" and "second" may explicitly or implicitly include at least one of the features. In the description of this application, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.

[0072] It should be understood by those skilled in the art that the above embodiments are only for the purpose of clearly illustrating the present invention, and are not intended to limit the scope of the present invention. For those skilled in the art, other changes or modifications may be made based on the above invention, and these changes or modifications are still within the scope of the present invention.

Claims

1. A high-speed inspection and rejection device, characterized in that: include: Conveyor belt (1), code reading trigger photoelectric sensor (2), depth camera, code reader (3), rejection component (4), rejection detection photoelectric sensor (5), encoder (6) and PLC; The code reading trigger photoelectric sensor (2), the code reader (3), the rejection component (4) and the rejection detection photoelectric sensor (5) are arranged in sequence along the conveying direction of the conveyor belt (1), and the conveyor belt (1) is used to convey bottle caps with two-dimensional codes on the conveyor belt (1); The code reading trigger photoelectric sensor (2) and the rejection detection photoelectric sensor (5) are both arranged on one side of the conveyor belt (1), and the detection direction is perpendicular to the conveying direction of the bottle caps; the depth camera and the code reader (3) are arranged in parallel and are located above the two-dimensional code reading area of ​​the bottle caps; The rejection component (4) applies a lateral force to the bottle caps located on the conveyor belt (1); The conveyor belt (1), the code reading trigger photoelectric sensor (2), the depth camera, the code reader (3), the rejection component (4), the rejection detection photoelectric sensor (5) and the encoder (6) are all electrically connected to the PLC.

2. A high-speed inspection and rejection device according to claim 1, characterized in that: It also includes a vibration sensor and a light sensor, wherein the vibration sensor is arranged on the conveyor belt (1), and the light sensor is arranged near the code reading trigger photoelectric sensor (2) and the rejection detection photoelectric sensor (5), and the vibration sensor and the light sensor are both electrically connected to the PLC.

3. A verification and rejection method under high-speed conditions, characterized in that: Based on the high-speed verification and rejection device as claimed in claim 2, the verification and rejection method includes: The code reading trigger photoelectric sensor generates a trigger signal when detecting that the bottle cap passes through the trigger point; According to the trigger signal, the PLC triggers the code reader to read the data of the QR code on the bottle cap; The PLC receives the reading result of the code reader and compares it with the pre-stored data to generate the corresponding verification signal; Determine whether the bottle cap is qualified according to the verification signal. If it is qualified, it will not be rejected; if it is unqualified, a control signal will be applied to the rejection component; The rejection component applies lateral force to the defective products to remove the caps from the conveyor belt; The reject detection photoelectric sensor verifies the reject result and issues an alarm signal or stops the conveyor belt if the rejects are not successfully rejected.

4. The high-speed verification and rejection method according to claim 3, characterized in that: The method of triggering the code reader to read the data of the QR code on the bottle cap includes: Determine the distance from the code reading trigger photoelectric sensor to the code reading area , and convert it into a pulse value ,in, is the pulse resolution of the encoder; After receiving the trigger signal, the PLC records the current encoder count value and uses it as the code reading reference value ; Calculate the difference between the current encoder count value and the target position , ,in, It is the count value fed back by the encoder in real time; Calculate dynamic compensation error parameters , ,in, is the conveyor belt running speed, is the signal delay time of PLC, is the correction of the pulse value by the environmental error, and is the experience weight coefficient; like , the PLC sends a code reading control signal to the code reader.

5. A high-speed verification and rejection method according to claim 4, characterized in that: The depth camera is used to obtain the precise position of the bottle cap and the barcode reader is controlled to align with the bottle cap. The method includes: The depth camera collects the initial depth map of the code reading area ; Perform mean filtering and denoising on the depth values ​​in the depth map to obtain a depth map ; Setting Dynamic Depth Threshold and , ,in, is the median of the depth values ​​of the depth map, is the interquartile range of the depth values ​​of the depth map, is the dynamic adjustment coefficient; Binarize the depth map to obtain the mask map , ,in, Depth map Medium pixel coordinates The depth value of The mask image Medium pixel coordinates The binary depth value of Determine the center of mass of the bottle cap in pixel coordinates , ,in, is the area of ​​the bottle cap in pixel coordinates ; Get the two-dimensional physical coordinates of the bottle cap , ; Get the actual center position of the barcode reader , and calculate the error between the bottle cap position and the actual center position ,like Then read the QR code; otherwise adjust the actual center position of the code reader until it meets ,in, and is the allowable horizontal error range and vertical error range.

6. A high-speed verification and rejection method according to claim 5, characterized in that: Allowable horizontal error range , allowable vertical error range ,in, and is the static basic error range, is the conveyor belt speed, is the depth camera frame rate, is the speed influence coefficient; The method to adjust the actual center position of the code reader is: ,in, is the actual center position of the code reader after adjustment. To adjust the step size factor.

7. The high-speed verification and rejection method according to claim 3, characterized in that: The vibration intensity of the conveyor belt is detected by the vibration sensor, and the light intensity near the code reading trigger photoelectric sensor and the rejection detection photoelectric sensor is detected by the light sensor. When the vibration intensity or light change exceeds the preset threshold, the sensitivity of the photoelectric sensor and the trigger delay of the rejection device are adjusted.

8. The high-speed verification and rejection method according to claim 7, characterized in that: Adjustment methods include: Collect environmental signals and record them in time series ,in, for The vibration intensity at the moment, for Light intensity at the time; Determine the quantitative index of vibration intensity ,in, is the sampling time window; Calculate the relative rate of change of light intensity ,in, and For the past time window The maximum and minimum values ​​of the internal light intensity, is the average light intensity; Dynamic adjustment function to determine the sensitivity of photosensors ,in, is the initial sensitivity of the photoelectric sensor, A coefficient used to control the effect of vibration and light changes on sensitivity; Dynamic adjustment function that determines the triggering delay of the reject device ,in, Delay for initial triggering of the reject device, A coefficient used to control the effect of vibration and light changes on the trigger delay.

9. A high-speed verification and rejection method according to claim 8, characterized in that: Construct a dynamic prediction model based on long short-term memory network and and As input, we can get the prediction result ; Adjust sensitivity and trigger delay in advance based on prediction results; .

10. The high-speed verification and rejection method according to claim 3, characterized in that: Methods for verifying reject results with reject detection photoelectric sensors include: Determine the distance from the reject zone to the reject detection photoelectric sensor , and convert it into a pulse value ,in, is the pulse resolution of the encoder; PLC records the encoder count value when the rejection action occurs , calculate the target count value of the rejection detection position ; Get When the feedback signal of the photoelectric sensor is detected, if no trigger signal is detected, the rejection is judged to be successful; if a trigger signal is detected, the rejection is judged to be a failure.

Images