Connector spring height dimension measurement method, device and computer readable storage medium

Abstract

The present application mainly relates to the technical field of component size measurement. In order to solve the problem that the appearance size measurement of the current micro device is mainly based on the caliper measurement or the secondary dimension measurement, the operation is inconvenient, there are many error factors, the precision of measurement is controlled by manual operation, and the efficiency is low; the present application provides a connector spring height size measurement method, device and computer readable storage medium, the connector spring feature and the reference component feature are extracted by establishing a feature extraction model, and the spring height size of the connector is calculated according to the obtained connector spring feature and reference component feature. Uncertain error can be effectively avoided in the process of measuring the spring height size of the connector, the measurement result is influenced, and the measurement efficiency can be greatly improved.

Description

Technical Field

[0001] This invention relates primarily to the field of electronic component measurement technology, and in particular to a method, apparatus, and computer-readable storage medium for measuring the spring height of a connector. Background Technology

[0002] With the advancement of technology, the manufacturing requirements for many tiny parts are becoming increasingly precise, and the quality requirements for appearance are also becoming higher in the actual production process, especially in terms of appearance dimensions. Even small dimensional differences can cause significant losses. Traditional appearance dimension measurement mainly relies on calipers and two-dimensional measurement, but both of these methods have problems such as inconvenience and low efficiency. Inconsistent operating techniques can also lead to inaccurate measurements.

[0003] Connectors are widely used general-purpose electronic components in the electronics industry. In various electronic systems, connectors are essential basic components for connecting devices to components, components to individual units, individual units to individual units, and systems to systems. To enable the connector interface to connect with the interface of digital products, the conductive terminals inside the connector typically have a spring height for connection between the connection end and the digital product interface. This spring height is used to elastically engage with the interface of the digital product, making the connection more compact. The accuracy of the spring height measurement of the connector greatly affects the working performance of various devices. Summary of the Invention

[0004] The technical problem to be solved by the present invention

[0005] This invention provides a method, apparatus, and computer-readable storage medium for measuring the spring height of connectors, which solves the problems of current methods for measuring the spring height of connectors, which mainly rely on calipers or two-dimensional measuring instruments. These methods are inconvenient to operate, have many error factors, and rely on manual control of the measurement precision, resulting in low efficiency.

[0006] The technical solution adopted by the present invention to solve the above problems

[0007] A method for measuring the spring height of a connector, comprising,

[0008] Acquire multiple sets of image data of different connectors and connector component annotation information;

[0009] A feature extraction model is trained based on the acquired connector image data and connector component annotation information;

[0010] The spring features and reference component features of the connector to be detected are extracted using a trained feature extraction model;

[0011] The spring height dimension of the connector is calculated based on the obtained connector spring features and reference component features.

[0012] Furthermore, the mask image of the connector is obtained, and a feature extraction model is established based on the connector mask image.

[0013] Furthermore, the step of training the feature extraction model based on the acquired connector image data and connector component annotation information specifically includes:

[0014] Connectivity analysis is performed on the acquired mask image of the connector component. Each channel of the mask image after connectivity analysis is transformed using Euclidean transformation to obtain the transformed mask image.

[0015] The acquired connector image is input into a predefined image segmentation network to obtain the predicted mask image;

[0016] Calculate the loss value between the predicted mask image and the Euclidean transformed mask image, and train a predefined image segmentation network to convergence based on the loss value to obtain the feature extraction model.

[0017] Furthermore, the feature extraction model includes N output channels, with the channel corresponding to the category of the spring piece outputting the spring piece features and the channel corresponding to the category of the reference component outputting the reference component features.

[0018] Furthermore, the calculation of the connector spring height dimension based on the acquired connector spring characteristics and reference component characteristics specifically includes:

[0019] Select corresponding feature points on the spring features according to the set geometric constraints;

[0020] Select the corresponding reference component feature points on the reference component features according to the set geometric constraints;

[0021] Calculate the distance from the feature point of the spring piece to the feature point of the reference component to obtain the spring height dimension of the connector.

[0022] Furthermore, before inputting the image of the connector to be detected into the feature extraction model, the image of the connector to be detected is preprocessed. The preprocessing includes calculating the mean and then the variance of the image of the connector to be detected.

[0023] Furthermore, annotation tools are used to perform pixel-level edge annotations on each component of the connector to obtain the annotation information of the connector components.

[0024] Based on the aforementioned method for measuring the spring height of a connector, the present invention further provides a device for measuring the spring height of a connector, specifically comprising:

[0025] Data acquisition module: used to acquire multiple sets of image data and annotation information of the target connector;

[0026] Training module: Used to train a feature extraction model based on the acquired connector image data and annotation information and a predefined image segmentation network;

[0027] Feature extraction module: used to extract the features of the spring contacts and reference components of the connector to be detected based on the feature extraction model;

[0028] Calculation module: used to calculate the connector spring height dimension based on the acquired connector spring features and reference component features.

[0029] Furthermore, the data acquisition module is a CCD camera.

[0030] The present invention also provides a computer-readable storage medium storing program code for implementing the connector spring height measurement method.

[0031] Beneficial effects of the present invention

[0032] This invention provides a method for measuring the spring height of a connector. It involves establishing a feature extraction model to extract the features of the connector spring contacts and the features of a reference component, and then calculating the spring height of the connector based on these features. This method effectively avoids uncertainties in the measured spring height and significantly improves the efficiency of connector spring height measurement.

[0033] The data acquisition module of the connector spring height measurement device provided in this invention uses a CCD camera, and the parameters can be set according to the specific situation of the connector. It can measure the spring height of various types of connectors, thereby improving the applicability of the measurement device. Attached Figure Description

[0034] Figure 1 This is a flowchart of a connector spring height measurement method according to the present invention;

[0035] Figure 2 This invention relates to a connector spring height measurement device. Detailed Implementation

[0036] like Figure 2 As shown, the connector spring height measurement device of the present invention specifically includes:

[0037] Data acquisition module: used to acquire multiple sets of image data and annotation information of the target connector; the data acquisition module may specifically be a CCD camera;

[0038] Training module: Used to train a feature extraction model based on image data and annotation information;

[0039] Feature extraction module: used to extract features of spring contacts and reference components based on the image of the connector under test using a feature extraction model;

[0040] Calculation module: Used to calculate the projectile height based on the characteristics of the shrapnel and the reference characteristics.

[0041] like Figure 1 As shown, based on the connector spring height measurement device described above, the connector spring height measurement method of the present invention specifically includes the following steps.

[0042] S1. Obtain images of multiple different connectors and annotation information of connector components;

[0043] The specific method for acquiring connector image data is to capture connector images using a CCD camera. The selection of the CCD camera is based on the actual size of the target connector, the working distance of the CCD, and the measurement accuracy requirements of the manufacturing process. The CCD and lens parameters are calculated and determined. Then, multiple appearance images of the target connectors are captured using the selected hardware. Labelme and other annotation tools are used to perform pixel-level edge annotation on each component of the connector, and the annotation information is converted into a mask image.

[0044] S2. Train a feature extraction model based on the acquired connector images and connector component annotation information;

[0045] Connectivity analysis is performed on the mask image described in step S1 to obtain a preprocessed mask image. Each channel of the preprocessed mask image consists of a binary image of the same category of object, for a total of N channels, where N is the number of connector components. Simultaneously, Euclidean distance transformation is performed on individual instances of each channel of the preprocessed mask image to obtain a transformed mask image. The image data described in S1 is then input into a predefined image network segmentation model to obtain a predicted mask image. Based on the predicted mask image and the transformed mask image, a loss value is calculated, and backpropagation is performed to update the segmentation network parameters until convergence, resulting in a trained feature extraction model. The convergence state is defined as the loss being less than a preset value, or the loss no longer decreasing, or the test metric on the test set no longer improving.

[0046] S3. Use the trained feature extraction model to extract the spring features and reference component features of the connector to be detected;

[0047] The image of the connector under test is normalized to obtain a normalized image. The normalization process includes mean and variance processing of the image. The normalized image is then input into a trained feature extraction model to obtain an output with N channels; the channel corresponding to the category of the spring piece outputs the spring piece features, and the channel corresponding to the category of the reference component outputs the reference component features.

[0048] S4. Calculate the spring height dimension of the connector based on the obtained connector spring characteristics and reference component characteristics.

[0049] Select corresponding feature points on the spring features according to the set geometric constraints;

[0050] Select the corresponding reference component feature points on the reference component features according to the set geometric constraints;

[0051] Calculate the distance from the feature point of the spring piece to the feature point of the reference component to obtain the spring height dimension of the connector.

[0052] Based on the above-described method for measuring the spring height of a connector, the present invention also provides a computer-readable storage medium storing program code for implementing the aforementioned method for measuring the spring height of a connector.

Claims

1. A method for measuring the spring height of a connector, characterized in that, include, Acquire multiple sets of different connector image data and connector component annotation information, wherein the connector image data is a mask image; A feature extraction model is trained based on the acquired connector image data and connector component annotation information; The feature extraction model includes N output channels. The channel corresponding to the category of the spring piece outputs the spring piece features, and the channel corresponding to the category of the reference component outputs the reference component features. The training of the feature extraction model based on the acquired connector image data and connector component annotation information specifically includes: Connectivity analysis is performed on the acquired mask image of the connector component. Each channel of the mask image after connectivity analysis is transformed using Euclidean transformation to obtain the transformed mask image. The acquired connector image is input into a predefined image segmentation network to obtain the predicted mask image; Calculate the loss value between the predicted mask image and the Euclidean transformed mask image, and train a predefined image segmentation network to convergence based on the loss value to obtain the feature extraction model; The spring features and reference component features of the connector to be detected are extracted using a trained feature extraction model; The spring height dimension of the connector is calculated based on the obtained connector spring features and reference component features.

2. The connector spring height measurement method according to claim 1, characterized in that, The spring height dimensions of the connector are calculated based on the obtained characteristics of the connector spring contacts and the characteristics of the reference components. Specifically, this includes... Select corresponding feature points on the spring features according to the set geometric constraints; Select the corresponding reference component feature points on the reference component features according to the set geometric constraints; Calculate the distance from the feature point of the spring piece to the feature point of the reference component to obtain the spring height dimension of the connector.

3. The connector spring height measurement method according to claim 1, characterized in that, Before inputting the image of the connector to be detected into the feature extraction model, the image of the connector to be detected is preprocessed. The preprocessing includes calculating the mean and then the variance of the image of the connector to be detected.

4. The connector spring height measurement method according to claim 1, characterized in that, Use annotation tools to perform pixel-level edge annotations on each component of the connector to obtain the annotation information of the connector components.

5. A connector spring height measurement device, used to implement the connector spring height measurement method according to any one of claims 1-4, characterized in that, include: Data acquisition module: used to acquire multiple sets of image data and annotation information of the target connector; Training module: Used to train a feature extraction model based on the acquired connector image data and annotation information and a predefined image segmentation network; Feature extraction module: used to extract the features of the spring contacts and reference components of the connector to be detected based on the feature extraction model; Calculation module: used to calculate the connector spring height dimension based on the acquired connector spring features and reference component features.

6. The connector spring height measuring device according to claim 5, characterized in that, The data acquisition module is a CCD camera.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores program code for implementing the connector spring height measurement method according to any one of claims 1-4.

Images