Binding detection method of intelligent ring sensor, medium and electronic equipment

Abstract

The invention relates to the technical field of intelligent ring detection, in particular to a binding detection method for an intelligent ring sensor, a medium and electronic equipment, and the method comprises the following steps: a, carrying out the binding detection of a sensor based on an electric signal, and collecting and analyzing electric signal data through simulating a wearing scene; b, performing single-channel stability detection and multi-channel consistency detection on the electric signal data of each channel; c, performing defect detection on the binding area by using machine vision and a deep learning algorithm; and d, comprehensively judging whether binding is qualified or not according to an electric signal detection result and a machine vision detection result. Whether binding meets the requirement or not is judged by analyzing and comparing the difference between an electric signal of a single channel and a standard electric signal and the difference between electric signals of multiple channels; binding defects which are difficult to distinguish visually are detected by using machine vision and a deep learning algorithm, so that the manpower demand is greatly reduced, and the detection efficiency and accuracy are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of smart ring detection, and in particular to a binding detection method, medium and electronic equipment for a smart ring sensor. Background Art

[0002] In recent years, smart rings, as a highly anticipated smart wearable device, have successfully integrated the elegant aesthetics of traditional rings with the powerful functions of modern technology. As the problem of China's aging population gradually becomes prominent, the number of empty-nest elderly and elderly people living alone continues to rise, and people are paying more and more attention to health and lifestyle. The smart ring industry has made great progress in the development of health monitoring technology, and continues to promote the innovative development of the industry. The continuous upgrading of sensor technology has made the functions of smart rings increasingly powerful. As the core component of smart rings, sensors can accurately monitor and analyze the user's physiological data. At present, smart rings have realized many functions such as heart rate monitoring, sleep quality monitoring, and exercise step counting. However, the sensors of smart rings are thin and precise, and are easily damaged during welding or binding processes.

[0003] The existing smart ring sensor binding detection has the following defects:

[0004] 1. Slow detection speed: The traditional detection method requires manual spot check of the sensor binding area under a microscope, which is time-consuming and labor-intensive, has a slow detection speed, and is prone to missed detection, which cannot meet the needs of the production line.

[0005] 2. Poor reliability: Since visual inspection by human eyes can only detect and judge the binding effect based on the surface features of the binding area, it is difficult to accurately distinguish between ACF attachment marks and sensor line tearing and other adverse conditions, and there is a risk of misjudgment.

[0006] 3. Lack of comprehensive judgment based on multiple factors: Traditional detection methods use electrical signal detection as the main detection and judgment basis, that is, if the electrical signal is normal, it is judged as a good product. Sensors with defects after binding cannot be detected, resulting in defective products in the later stage. Summary of the invention

[0007] In view of the problems of the prior art, the present invention provides a binding detection method, medium and electronic device for a smart ring sensor, which determines whether the binding meets the requirements by analyzing and comparing the difference between the electrical signal of a single channel and the standard electrical signal and the difference between the electrical signals of multiple channels; uses machine vision and deep learning algorithms to detect binding defects that are difficult to distinguish visually, greatly reducing manpower requirements and improving detection efficiency and accuracy.

[0008] In order to solve the above technical problems, the present invention adopts the following technical solutions:

[0009] The present invention provides a binding detection method for a smart ring sensor, which comprises the following steps:

[0010] Step a: Sensor binding detection based on electrical signals, collecting and analyzing electrical signal data by simulating wearing scenarios;

[0011] Step b, performing single-channel stability detection and multi-channel consistency detection on the electrical signal data of each channel;

[0012] Step c: Use machine vision combined with deep learning algorithms to detect defects in the binding area; Step d: Comprehensively determine whether the binding is qualified based on the electrical signal detection results and the machine vision detection results.

[0013] Wherein, in the step a, the electrical signal collection is carried out under specific lighting conditions, and the returned electrical signal data is continuously collected and recorded, and the electrical signal data is saved in an array form for subsequent analysis.

[0014] Wherein, in the step b, the stability of the electrical signal is evaluated by calculating the standard deviation of the electrical signal data collected by the same channel, and whether there is signal instability in the bound sensor channel is determined according to a preset threshold.

[0015] Wherein, in the step b, by comparing the difference ratio between the electrical signals of multiple channels with the difference of theoretical signal values, possible binding connection problems are identified.

[0016] Wherein, in the step c, the sensor binding defect is detected using the YOLOv10-S model, and the sensor binding defect detection by the YOLOv10-S model includes the following steps:

[0017] Step c1, collection and annotation of defect data set;

[0018] Step c2: Model selection of YOLOv10.

[0019] Wherein, step c includes the process of training and evaluating the YOLOv10-S model, and the process of training and evaluating the YOLOv10-S model is used to optimize the hyperparameter selection.

[0020] Among them, in the step d, the specific defect type detected and its confidence level are output, and defects with a confidence level less than 70% are ignored; defects with a confidence level between 70-85% are manually reviewed; and defects with a confidence level higher than 85% are directly used. The detection results of the algorithm model are used.

[0021] The present invention also provides a computer storage medium, which stores computer instructions. When the computer instructions are called, they are used to execute the binding detection method of the smart ring sensor.

[0022] The present invention also provides an electronic device, wherein the electronic device includes: a processor; and a memory arranged to store computer executable instructions, wherein the executable instructions, when executed, cause the processor to execute the binding detection method of the smart ring sensor.

[0023] Beneficial effects of the present invention:

[0024] The present invention significantly improves the detection efficiency and accuracy of the smart ring sensor after binding by combining electrical signal detection technology and machine vision defect detection algorithm; introduces an automated detection process based on electrical signals and machine vision, replaces traditional manual visual inspection, greatly reduces manual intervention, and improves the detection speed; can achieve real-time detection and immediate feedback of detection results, meeting the production line's demand for rapid response; is not limited to electrical signal detection, but also adds machine vision detection, which can identify subtle defects such as ACF attachment marks, sensor line tearing, etc. that are difficult to detect with the naked eye, improving the comprehensiveness and accuracy of detection; through the use of machine learning models (such as YOLOv10), defects can be more accurately identified and classified, reducing the possibility of misjudgment; integrates electrical signal stability and consistency detection with machine vision defect analysis, realizes a comprehensive evaluation of binding effects, and ensures that only high-quality products can enter the market; especially for those products that have physical damage or binding problems even if the electrical signal is normal, effective detection is carried out to prevent defective products from flowing into subsequent links; reduces dependence on manual inspectors and reduces labor costs in long-term operations; by early detection and interception of defective products, reduces the rework and repair costs that may occur in the later stage; improves the automation level and efficiency of production. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 The present invention is a flowchart of a method for detecting binding of a smart ring sensor.

[0026] Figure 2 The figure is a flow chart of the electric signal detection method of the present invention.

[0027] Figure 3 This is a flow chart of the sensor binding detection method based on YOLOv10 of the present invention. DETAILED DESCRIPTION

[0028] In order to facilitate the understanding of those skilled in the art, the present invention is further described below in conjunction with the embodiments and drawings, and the contents mentioned in the embodiments are not intended to limit the present invention. The present invention is described in detail below in conjunction with the drawings.

[0029] Embodiment 1

[0030] In the first embodiment of the present application, a binding detection method for a smart ring sensor is provided, which includes the following steps:

[0031] Step a: Sensor binding detection based on electrical signals, collecting and analyzing electrical signal data by simulating wearing scenarios;

[0032] Step b, performing single-channel stability detection and multi-channel consistency detection on the electrical signal data of each channel;

[0033] Step c: Use machine vision combined with deep learning algorithms to detect defects in the binding area; Step d: Comprehensively determine whether the binding is qualified based on the electrical signal detection results and the machine vision detection results.

[0034] The embodiment of the present application significantly improves the detection efficiency and accuracy of the smart ring sensor after binding by combining electrical signal detection technology and machine vision defect detection algorithm; introduces an automated detection process based on electrical signals and machine vision, replaces traditional manual visual inspection, greatly reduces manual intervention, and improves the detection speed; can achieve real-time detection and immediate feedback of detection results, meeting the production line's demand for rapid response; is not limited to electrical signal detection, but also adds machine vision detection, which can identify subtle defects such as ACF attachment marks and sensor line tearing that are difficult to detect with the naked eye, improving the comprehensiveness and accuracy of detection; through the use of machine learning models (such as YOLOv10), defects can be more accurately identified and classified, reducing the possibility of misjudgment; integrates electrical signal stability and consistency detection with machine vision defect analysis, realizes a comprehensive evaluation of binding effects, and ensures that only high-quality products can enter the market; especially for those products that have physical damage or binding problems even if the electrical signal is normal, effective detection is carried out to prevent defective products from flowing into subsequent links; reduces dependence on manual inspectors and reduces labor costs in long-term operations; reduces rework and repair costs that may occur later by early detection and interception of defective products; improves the automation level and efficiency of production.

[0035] In the embodiment of the present application, in the step a, the electrical signal acquisition is performed under specific lighting conditions, and the returned electrical signal data is continuously acquired and recorded, and the electrical signal data is saved in the form of an array for subsequent analysis; under specific lighting conditions, the scene of wearing a smart ring is simulated, and the photoelectric data received by the sensor terminal is acquired through the PCBA board on the smart ring; within a continuous 2000 milliseconds, the returned electrical signal data is acquired and recorded for each channel every 100 milliseconds, and finally saved in the form of an array, and the acquired data is as follows:

[0036]

[0037] In the embodiment of the present application, in the step b, the stability of the electrical signal is evaluated by calculating the standard deviation of the electrical signal data collected by the same channel, and whether the bound sensor channel has signal instability is determined according to a preset threshold. Specifically, after the sensor is bound, the electrical signal data of each channel is analyzed and compared through single-channel electrical signal detection, so as to better understand the stability and reliability of the signal acquisition of the detection sensor. Theoretically, under the same conditions, the 20 electrical signal data collected by each individual channel are the same. Therefore, the stability of the electrical signal can be evaluated by calculating the standard deviation of the electrical signal data collected by each channel. The principle is shown in Formula 1. When S exceeds the set threshold H S , then the signal of the channel bound to the sensor is unstable, otherwise, the test is passed;

[0038]

[0039] Among them, x i Represents the electrical signal data collected by the same channel. It is the average value of the electrical signal data collected by the same channel.

[0040] In the embodiment of the present application, in the step b, by comparing the difference ratio between the electrical signals of multiple channels and the difference between the theoretical signal values, possible binding connection problems are identified. Specifically, the main detection principle of multi-channel electrical signal detection is: theoretically, the electrical signal values ​​returned by the same photoelectric sensors under the same conditions are indifferent. Based on this characteristic, a corresponding algorithm can be designed to intercept bound products whose signal value differences exceed a threshold; the embodiment of the present application uses the difference ratio between signal channels and the difference between the actual signal and the theoretical signal to design a corresponding detection method based on the sensor characteristics.

[0041] The difference ratio between signal channels is based on the same optical signal input condition of the built-in photoelectric sensor of the same sensor. Theoretically, the electrical signal data collected each time is the same. Considering the interference of resistance, the difference ratio threshold H can be set for the difference ratio between signal channels. φ When the difference ratio φ between the signal channels exceeds H φ When , the difference between the signal channels is too large, that is, there is a problem with the sensor binding connection. Otherwise, this step can be used for detection; where H φ The calculation method of is shown in formula (2):

[0042] right x1≠x2, there is Among them, I standard It is the theoretical electrical signal value output under the current optical signal conditions.

[0043] In the embodiment of the present application, in step c, the sensor binding defect is detected using the YOLOv10-S model. Step c includes the process of training and evaluating the YOLOv10-S model, and the process of training and evaluating the YOLOv10-S model is used to optimize the hyperparameter selection; in step d, the specific defect type detected and its confidence are output, and defects with a confidence of less than 70% are ignored; defects with a confidence of 70-85% are manually re-judged; for defects with a confidence of more than 85%, the detection results of the algorithm model are directly used. Specifically, under the above settings, sensor binding detection based on YOLOv10 is a detection technology that integrates deep learning and machine vision to achieve accurate detection of sensor binding; YOLOv10 is a highly efficient target detection algorithm that can quickly and accurately identify multiple targets in an image and has a good balance between speed and accuracy; by applying the YOLOv10 algorithm to the field of sensor binding detection, the accuracy and efficiency of sensor binding can be improved, thereby better meeting the needs of sensor binding in fields such as industrial production and smart devices; the sensor binding detection method based on YOLOv10 is specifically as follows:

[0044] Collection and annotation of defect data sets: For defective data, collect and organize them into defect data sets and annotate them: Deep learning plays an important role in the training of sensor binding defect detection models, especially in detecting partial tearing of sensors, misaligned binding, and too few binding particles. The existence of these problems may cause quality problems in the production process of smart rings, affecting the performance and reliability of the product. Therefore, it is crucial to collect defect detection data sets, and the quality and diversity of the data sets need to be considered to ensure that the model can accurately capture the characteristics of various defect types.

[0045] Model selection of YOLOv10: For the problem of smart ring sensor binding defect detection, the selection of the YOLOv10 model needs to comprehensively consider the actual application scenarios. First of all, as a target detection model, YOLOv10 has six different sizes of N / S / M / B / L / X. Therefore, it is necessary to select according to the specific application. For the specific task of sensor binding defect detection, it is necessary to conduct detailed detection and analysis of the sensor binding situation, so the accuracy and robustness of the model need to be considered. In addition, considering that sensor binding defects are relatively typical and limited defect types, the generalization ability of the model is low, and the efficiency of the model is required to be high. In addition, according to the data characteristics and actual needs of specific application scenarios, it is also necessary to comprehensively consider factors such as the training and deployment costs of the model, as well as the requirements for hardware equipment. Therefore, according to the characteristics of the smart ring sensor binding defect, the embodiment of the present application weighs the specific detection situation and selects YOLOv10-S as the model for sensor binding detection of YOLOv10.

[0046] Model training and evaluation: The selection of appropriate hyperparameters also has an important impact on model performance. By continuously optimizing the model training process, its effectiveness and accuracy in detecting partial tearing of sensors, misaligned binding, and a small number of bound particles can be improved; the training of deep learning defect detection models is a complex and critical task. It is necessary to adjust the ratio of training sets and validation sets according to the actual data set, and continuously adjust and optimize to adapt to the ever-changing production environment and needs. In addition, during the training process, the use of validation sets is very important for evaluating models. By using evaluation indicators such as accuracy, recall rate, and F1 score, the performance of the model can be effectively measured; by analyzing the evaluation results, the model parameters and training strategies can be adjusted in a targeted manner, thereby further optimizing the performance of the model and improving the accuracy and generalization ability of the model; the selection of hyperparameters for model training is shown in Table 1.

[0047] Table 1 Hyperparameters for model training

[0048]

[0049]

[0050] Output defect type: Output pass or fail information based on the test results. If the product passes the test, the output result is "no defects detected", which means that the product meets the quality standards and does not have any defects. However, if the product fails the test, the specific defect type detected and the defect confidence μ will be output in detail; if the output defect confidence is less than 70%, the defect will be ignored directly; for defects with defect confidence in the range of 70-85%, manual re-judgment will be performed to reduce the possibility of misjudgment; if the defect confidence is higher than 85%, the detection result of the algorithm model will be used directly.

[0051]

[0052] In summary, it can be seen that there are three key points in the embodiment of the present application: Key point 1: In view of the problems of difficult detection and low detection efficiency after the smart ring sensor is bound, a new sensor binding detection method based on electrical signals is proposed. By analyzing and comparing the difference between the electrical signal of a single channel and the standard electrical signal and the difference between the electrical signals of multiple channels, it is determined whether the binding meets the requirements. Key point 2: The use of machine vision and deep learning algorithms to detect binding defects that are difficult to distinguish visually greatly reduces manpower requirements and improves detection efficiency and accuracy. Key point 3: Based on the above two points, it is possible to detect products that can pass electrical signal detection but have defects, such as partial tearing of the sensor binding line and too few binding particles.

[0053] Compared with the existing technology, the embodiments of the present application propose a new electrical signal detection method based on the characteristics and principles of the sensor, which can efficiently and accurately perform electrical signal detection on the bound sensor; in addition, on the basis of electrical signal detection, the binding area is simultaneously subjected to binding defect detection using the improved machine vision and deep learning algorithms of the embodiments of the present application to detect invisible defects that cannot be detected by the electrical signal detection method, such as crushing of the sensor wire, partial tearing, and insufficient number of binding particles; the embodiments of the present application can be applied to the actual production of smart rings of different models, and only requires flexible adjustment of hardware equipment such as cameras, light sources, fixed bases, and electrical signal connecting cables, thereby realizing fast and real-time smart ring sensor binding detection, while ensuring the accuracy of the judgment, greatly improving the production efficiency of smart ring binding detection.

[0054] Embodiment 2

[0055] In the second embodiment of the present application, a computer storage medium is provided, wherein the computer storage medium stores computer instructions, and when the computer instructions are called, they are used to execute the binding detection method of the smart ring sensor.

[0056] Embodiment 3

[0057] In the third embodiment of the present application, an electronic device is provided, wherein the electronic device includes: a processor; and a memory arranged to store computer executable instructions, wherein the executable instructions, when executed, cause the processor to execute the binding detection method of the smart ring sensor.

[0058] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0059] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0060] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0061] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0062] Although the preferred embodiments of the present invention have been described, those skilled in the art may make other changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present invention.

[0063] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A binding detection method for a smart ring sensor, characterized in that: The method comprises the following steps: step a, sensor binding detection based on electrical signals, collecting and analyzing electrical signal data by simulating wearing scenarios; Step b, performing single-channel stability detection and multi-channel consistency detection on the electrical signal data of each channel; Step c: Use machine vision combined with deep learning algorithms to detect defects in the binding area; Step d: Comprehensively determine whether the binding is qualified based on the electrical signal detection results and the machine vision detection results.

2. The binding detection method of a smart ring sensor according to claim 1, characterized in that: In the step a, the electrical signal collection is performed under specific lighting conditions, and the returned electrical signal data is continuously collected and recorded, and the electrical signal data is saved in an array form for subsequent analysis.

3. The binding detection method of a smart ring sensor according to claim 1, characterized in that: In the step b, the stability of the electrical signal is evaluated by calculating the standard deviation of the electrical signal data collected by the same channel, and it is determined whether the bound sensor channel has signal instability according to a preset threshold.

4. The binding detection method of a smart ring sensor according to claim 1, characterized in that: In the step b, possible binding connection problems are identified by comparing the difference ratios between the electrical signals of multiple channels with the theoretical signal value differences.

5. The binding detection method of a smart ring sensor according to claim 1, characterized in that: In the step c, the sensor binding defect is detected using the YOLOv10-S model, and the sensor binding defect detection by the YOLOv10-S model includes the following steps: Step c1, collection and annotation of defect data set; Step c2: Model selection of YOLOv10.

6. The method for detecting binding of a smart ring sensor according to claim 1, characterized in that: The step c includes a process of training and evaluating the YOLOv10-S model, and the process of training and evaluating the YOLOv10-S model is used to optimize the hyperparameter selection.

7. The method for detecting binding of a smart ring sensor according to claim 1, characterized in that: In the step d, the specific defect type detected and its confidence level are output, defects with a confidence level less than 70% are ignored; defects with a confidence level between 70-85% are manually re-judged; and defects with a confidence level higher than 85% are directly detected using the detection results of the algorithm model.

8. A computer storage medium, wherein the computer storage medium stores computer instructions, and when the computer instructions are called, they are used to execute the binding detection method of the smart ring sensor as described in any one of claims 1-7.

9. An electronic device, wherein: The electronic device comprises: a processor; and a memory arranged to store computer executable instructions, wherein when the executable instructions are executed, the processor executes the binding detection method of the smart ring sensor as described in any one of claims 1-7.

Images