Wire rod positioning detection device and electronic equipment

Abstract

The invention provides a wire rod positioning detection device and electronic equipment, the device comprises a light emitting unit, an acquisition unit and a calculation unit, the light emitting unit and the acquisition unit are oppositely arranged, a detection placement area is formed between the light emitting unit and the acquisition unit, and the detection placement area is configured to place a target wire rod to be detected; the light-emitting unit is used for responding to the light-emitting instruction of the calculation unit and irradiating the target wire rod at different positions for multiple times so as to present a plurality of projections of the target wire rod in the acquisition area of the acquisition unit; the acquisition unit is used for acquiring a plurality of projected images and sending the projected images to the calculation unit; and the calculation unit is used for calculating positioning parameters of the target wire rod based on the plurality of projected images, and the positioning parameters comprise position information and attitude information of the target wire rod. The multiple projected images are obtained through multiple times of irradiation, more effective areas can be obtained in the multiple projected images, the possibility of obtaining the effective areas is improved, and when the effective areas are enlarged, the effectiveness and accuracy of the wire rod posture detection result can be improved.

Description

Technical Field

[0001] The present application relates to the technical field of non-contact wire detection, and in particular to a wire positioning detection device and an electronic device. Background Art

[0002] With the rapid development of the machinery and electronics industries, the precision requirements for workpiece parameters are becoming higher and higher. Specifically, in the scenario of wire application, the posture detection of the wire is more important. In some wire posture detection methods, parallel light in different directions is generally used to illuminate the wire, and the diameter and posture of the wire in three directions are obtained by analyzing the projection of the wire. However, in practical applications, the cost of parallel light equipment is high, and the light always has reflection, refraction or diffraction, so parallel light is difficult to achieve accurately. In addition, the parallel light source installation position is highly dependent on the installation position of the light source, and the inaccurate installation position of the light source may affect the subsequent wire posture detection results. On the other hand, when irradiated with parallel light, it is limited to the overlapping part of the parallel light. Since the effective area of ​​the overlapping part is small, it is also difficult to obtain the light projection, which in turn affects the wire posture detection results. Summary of the invention

[0003] The embodiments of the present application provide a wire positioning detection device and an electronic device, which can illuminate the target wire multiple times through a light-emitting unit, thereby obtaining multiple projections to analyze the position information and posture information of the target wire. By obtaining multiple projection images through multiple illuminations, more effective areas can be obtained in the multiple projection images, thereby improving the possibility of obtaining the effective area. When the effective area is increased, the effectiveness and accuracy of the wire posture detection results can be improved.

[0004] In a first aspect, an embodiment of the present application provides a wire positioning detection device, which may include a light-emitting unit, a collection unit and a computing unit, wherein the light-emitting unit and the collection unit are arranged relative to each other, and a detection placement area is formed between the light-emitting unit and the collection unit, and the detection placement area is configured to place a target wire to be detected; the light-emitting unit is used to respond to the light-emitting instruction of the computing unit, and illuminate the target wire multiple times at different positions to present multiple projections of the target wire in the collection area of ​​the collection unit; the collection unit is used to acquire images of the multiple projections and send them to the computing unit; the computing unit is used to calculate the positioning parameters of the target wire based on the images of the multiple projections, and the positioning parameters include position information and posture information of the target wire.

[0005] In some possible implementations, the light emitting unit includes at least three point light sources arranged side by side; and irradiating the target wire material multiple times at different positions includes: the at least three point light sources illuminate the target wire material one by one. By using point light sources, the equipment cost can be reduced accordingly. By setting at least three point light sources, at least six tangent lines of the elliptical cross section of the wire material connected by the point light sources can be obtained through the illumination of the point light sources, so that the parameters of the elliptical interface can be calculated based on the corresponding functional relationship to ensure the reliability of the detection device.

[0006] In some possible implementations, an arrangement direction of the at least three point light sources is consistent with an extension direction of a collection area of ​​the collection unit; and / or a distance between any two adjacent ones of the at least three point light sources is the same.

[0007] In some possible implementations, the at least three point light sources emit light one by one to illuminate the target wire, including: the plurality of point light sources light up one by one based on a preset first light-emitting sequence, and illuminate the target wire.

[0008] In some possible implementations, the acquisition unit includes a linear CCD sensor; and acquiring the images of the plurality of projections includes acquiring the projections presented one by one in the acquisition area through the linear CCD sensor. By using a linear CCD sensor, a one-dimensional projection image, that is, a row of pixel information, can be acquired, which greatly reduces the amount of data compared to other sensors, thereby reducing the computing pressure of the device.

[0009] In some possible implementations, the calculation unit includes a preset neural network model and a control module; the calculation of the positioning parameters of the target wire based on the multiple projected images includes: the control module performs superposition preprocessing on the received multiple projected images to obtain a detection image, and inputs the detection image into the preset neural network model; the preset neural network model outputs the position information and the cross-sectional diameter of the target wire on the detection surface based on the multiple projections, wherein the position information includes coordinate information of the target wire relative to the detection surface, the detection surface is the horizontal plane where the point light source is located, and the cross-sectional diameter of the target wire includes a first diameter and a second diameter, wherein the first diameter is the longest cross-sectional diameter of the target wire, and the second diameter is the shortest cross-sectional diameter of the target wire; the preset neural network model calculates the posture information based on the cross-sectional diameter.

[0010] In some possible implementations, the training process of the preset neural network model includes: dividing the first part of the data in the historical detection data as a training set to train the neural network model, and obtaining the trained first neural network model; dividing the second part of the data in the historical detection data as a test set to test and calibrate the first neural network model until the mean square error loss is lower than the preset accuracy, and obtaining the preset neural network model; wherein the historical detection data includes detection images and data parameters of wires of different diameters placed at different positions and postures on the detection surface, and the data parameter types include the cross-sectional diameter, position information, and normal vector angle of the wire relative to the detection surface; the data of the position and posture of the wire in the historical detection data are randomly distributed. By randomly distributing the data of the position and posture of the wire in the historical detection data, the possibility of the neural network model falling into overfitting can be reduced.

[0011] In some possible implementations, the first portion of data accounts for a first proportion in the historical detection data, and the second portion of data accounts for a second proportion in the historical detection data, wherein the first proportion is greater than the second proportion.

[0012] In some possible implementations, before the neural network model is trained using the first part of the data in the divided historical detection data as a training set to obtain the trained first neural network model, it also includes: performing image recognition processing on each detection image in the historical detection data to obtain a recognized detection image; wherein the image recognition processing process includes: based on a preset grayscale range, identifying the effective projection area in the detection image.

[0013] In some possible implementations, the preset neural network model calculates the posture information based on the cross-sectional diameter, including: calculating the normal vector angle of the target wire relative to the detection surface based on the first diameter and the second diameter, wherein the posture information is the normal vector angle of the target wire relative to the detection surface.

[0014] In some possible implementations, calculating the normal vector angle of the target wire relative to the detection surface based on the first diameter and the second diameter includes: calculating a first ratio based on the second diameter and the first diameter; and calculating the arc cosine value of the first ratio to obtain the normal vector angle of the target wire relative to the detection surface.

[0015] In some possible implementations, the device further includes: a comparison unit, used to determine whether the normal vector angle is less than a preset angle; wherein, when the normal vector angle is less than the preset angle, it is determined that the posture of the target wire meets the preset conditions; when the normal vector angle is not less than the preset angle, it is determined that the posture of the target wire does not meet the preset conditions.

[0016] In a second aspect, an embodiment of the present application provides a wire positioning detection device, which is externally connected to an electronic device, and the device includes: a light-emitting unit and a collection unit, the light-emitting unit and the collection unit are arranged relative to each other, and a detection placement area is formed between the light-emitting unit and the collection unit, and the detection placement area is configured to place a target wire to be detected; the light-emitting unit is used to respond to the light-emitting instruction of the calculation unit, and illuminate the target wire multiple times at different positions to present multiple projections of the target wire in the collection area of ​​the collection unit; the collection unit is used to acquire images of the multiple projections and send them to the external electronic device, wherein the external electronic device is used to calculate the positioning parameters of the target wire based on the images of the multiple projections, and the positioning parameters include position information and posture information of the target wire.

[0017] In a third aspect, an embodiment of the present application further provides an electronic device, and the electronic device is connected to the wire positioning detection device provided in the second aspect, the electronic device comprising a preset neural network model and a control module; the electronic device calculates the positioning parameters of the target wire based on the multiple projections provided by the wire positioning detection device, specifically comprising: the control module performs superposition preprocessing on the received images of the multiple projections to obtain a detection image, and inputs the detection image into the preset neural network model; the preset neural network model outputs the position information and the cross-sectional diameter of the target wire on the detection surface based on the detection image, wherein the position information includes the coordinate information of the target wire relative to the detection surface, the detection surface is the horizontal plane where the point light source is located, and the cross-sectional diameter of the target wire includes a first diameter and a second diameter, wherein the first diameter is the longest cross-sectional diameter of the target wire, and the second diameter is the shortest cross-sectional diameter of the target wire; the preset neural network model calculates the posture information based on the cross-sectional diameter.

[0018] Through the above technical solution, when it is currently necessary to perform wire positioning detection on the target wire to be detected, the target wire is set between the light-emitting unit and the collection unit, and the light-emitting unit can respond to the light-emitting instruction to irradiate the target wire multiple times through an ordinary light source, so that the target wire presents a corresponding multiple projections in the collection area of ​​the collection unit, thereby obtaining multiple projected images to analyze the position information and posture information of the target wire. By obtaining multiple projected images through multiple irradiations of the light-emitting unit, more effective areas can be obtained in the multiple projected images, thereby improving the possibility of obtaining the effective area. When the effective area is increased, the effectiveness and accuracy of the results of the wire posture detection of the calculation unit can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0020] Figure 1a A simple structural diagram of a wire positioning detection device provided in one embodiment of the present application

[0021] Figure 1b A schematic diagram of the structure of a wire positioning detection device provided in one embodiment of the present application;

[0022] Figure 2a A schematic diagram of the arrangement of point light sources provided in one embodiment of the present application;

[0023] Figure 2b A schematic diagram of the arrangement of point light sources provided in another embodiment of the present application;

[0024] Figure 2c A schematic diagram of the arrangement of point light sources provided in another embodiment of the present application;

[0025] Figure 3 A schematic diagram of the lighting sequence of point light sources provided in one embodiment of the present application;

[0026] Figure 4 A schematic diagram of constructing a detection image provided by an embodiment of the present application;

[0027] Figure 5 A schematic diagram of a wire cross section in historical detection data provided by one embodiment of the present application;

[0028] Figure 6 A schematic diagram of effective shadow area identification provided by an embodiment of the present application;

[0029] Figure 7 A schematic diagram of a wire cross section provided in one embodiment of the present application;

[0030] Figure 8 A schematic diagram of the normal vector angle of a target wire provided in one embodiment of the present application. DETAILED DESCRIPTION

[0031] In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0032] Figure 1a A simple structural diagram of a wire positioning detection device provided in one embodiment of the present application.

[0033] Reference Figure 1a As shown, the wire positioning detection device may include a light emitting unit 10, a collection unit 20, and a calculation unit 30. The light emitting unit 10 and the collection unit 20 are arranged opposite to each other, and a detection placement area is formed between the light emitting unit 10 and the collection unit 20. The detection placement area can be configured to place the target wire to be detected, and then the target wire 50 to be detected can be placed in the detection placement area. The light emitting unit 10 responds to the light emitting instruction of the calculation unit 30, and the emitted light beam can illuminate the target wire 50 so as to present a corresponding projection in the collection area 20a of the collection unit 20.

[0034] In one embodiment, the light emitting unit 10 and the collection unit 20 may be arranged opposite to each other at the same horizontal height. In other embodiments, the light emitting unit 10 and the collection unit 20 may not be arranged at the same horizontal height, for example, the light emitting unit 10 and the collection unit 20 may be arranged opposite to each other at an angle. The present application does not limit the arrangement height of the light emitting unit 10 and the collection unit.

[0035] In some embodiments, the acquisition unit 20 is connected to the calculation unit 30. When the light emitting unit 10 illuminates the target wire 50 placed in the detection placement area in response to the light emitting instruction of the calculation unit 30, the acquisition unit 20 acquires the projection of the target wire 50 in the sampling area 20a, and sends the acquired projection image to the calculation unit 30, so that the calculation unit 30 calculates the positioning parameters of the target wire 50 based on the received projection image. Figure 1aAs shown, the computing unit 30 may be a unit device independent of the acquisition unit 20. The acquisition unit 20 is connected to the computing unit 30 by wire or wirelessly, that is, after the acquisition unit 20 acquires the projection image, the projection image is sent to the computing unit 30 by wire or wireless transmission. In other embodiments, the computing unit 30 may also be built in the acquisition unit 20. The acquisition unit 20 may transmit the acquired projection image to the computing unit 30 through an internal line.

[0036] In some embodiments, the light emitting unit 10 irradiates the target wire 50 in the detection placement area multiple times in response to the light emitting instruction of the computing unit 30. The light emitting unit may irradiate for a first duration each time, and a second duration may be spaced between two adjacent irradiations. The first duration and the second duration may be customized according to the application scenario and user needs. The present application does not limit the specific duration and setting method of the first duration and the second duration.

[0037] In some embodiments, the computing unit 30 can control the timing of the acquisition unit to acquire the projection image based on the first duration and the second duration currently set and applied, thereby improving the quality of the acquired projection image, for example, making the projection boundary of the acquired projection image clearer, so as to accurately determine the actual projection range. Based on this, the positioning parameters of the target wire calculated by the computing unit 30 based on the projection image are more accurate.

[0038] In some embodiments, the light-emitting unit 10 responds to the light-emitting instruction and irradiates the target wire 50 multiple times at different positions, so that multiple projections corresponding to the target wire 50 can be presented on the collection area 20a. Exemplarily, the light-emitting unit 10 irradiates the target wire 50 five times at different positions, and accordingly, at each irradiation, the projection of the target wire 50 can be presented on the collection area 20a. Further, the collection unit 20 can collect the projections presented on the collection area 20a based on the collection timing set by the calculation unit, and then can collect the projections corresponding to the five irradiations, and send the collected images of the multiple projections to the calculation unit 30.

[0039] In some embodiments, the calculation unit 30 may calculate the positioning parameters of the target wire 50 based on the acquired multiple projected images, wherein the calculated positioning parameters may include the position information and posture information of the target wire 50. The calculated positioning parameters of the target wire 50 may then be applied to the current application scenario of the target wire 50. For example, the posture of the target wire 50 may be adjusted in real time based on the positioning parameters acquired in real time.

[0040] With the wire positioning detection device provided by the above-mentioned embodiment, based on multiple irradiations at different positions, the acquisition unit can acquire multiple projected images corresponding to the multiple irradiations and send them to the calculation unit, and then the calculation unit can obtain more effective areas in the multiple projected images, thereby improving the possibility of obtaining the effective area. When the effective area increases, the larger the effective area obtained, the closer it is to the actual position and actual posture of the target wire, and then the position information and posture information of the target wire are calculated based on more effective areas, which can improve the effectiveness and accuracy of the target wire posture detection results.

[0041] In some embodiments, in order to facilitate the practical application of the wire positioning detection device, the light emitting unit and the collection unit in the wire positioning detection device can be set on the module shell, and the module shell can be configured into a corresponding shape according to the application requirements of the wire positioning detection.

[0042] Figure 1b A schematic structural diagram of a wire positioning detection device provided in one embodiment of the present application.

[0043] Reference Figure 1b As shown, in some embodiments, the wire positioning detection device may include a module housing S1, a light emitting unit 10, a collection unit 20, and a computing unit 30. The functions and connection settings of the light emitting unit 10, the collection unit 20, and the computing unit 30 may be the same as those of Figure 1b The same or similar embodiments are shown and will not be described in detail here.

[0044] Exemplarily, the wire positioning detection device can be used in conjunction with welding equipment, wherein the target wire can be the welding rod of the welding equipment, and the welding rod needs to pass through the wire positioning detection device to reach the working surface for welding. Based on the above scenario requirements, the module shell S1 can be a penetrating hollow module, and then the light-emitting unit 10 and the collection unit 20 can be arranged opposite to each other on both sides of the hollow part of the module shell S1, so as to facilitate the irradiation of the target wire (such as the above-mentioned welding rod) passing through the hollow part, and present a projection in the collection area.

[0045] It should be noted that, in other embodiments, the shape of the module housing S1 may be other shapes, and the present application does not impose any limitation on this.

[0046] In some embodiments, the light emitting unit 10 may include at least three point light sources arranged side by side. In some embodiments, the arrangement direction of at least three point light sources in the light emitting unit 10 is consistent with the extension direction of the collection area of ​​the collection unit; and / or the spacing between any two adjacent ones of the at least three point light sources is the same. Among them, a point light source refers to a light source that emits light uniformly from a point to the surrounding space. For example, a point light source can be an incandescent lamp, a fluorescent lamp, or an LED lamp. Point light sources do not require the use of parallel light beams for irradiation, which reduces equipment costs.

[0047] Figure 2a A schematic diagram of the arrangement of point light sources provided in one embodiment of the present application.

[0048] Reference Figure 2a As shown, the light emitting unit 10 may include five point light sources arranged side by side, namely, a point light source 10a, a point light source 10b, a point light source 10c, a point light source 10d, and a point light source 10e.

[0049] In some embodiments, the point light sources in the light emitting unit 10 are arranged at the same interval. Figure 2a As shown, the distance between two adjacent point light sources is L1, wherein the specific length of the distance L1 can be determined based on the actual size of the wire positioning detection device and the average diameter of the target wire. The embodiment of the present application does not limit the specific length of the distance L1.

[0050] In some embodiments, the point light sources in the light-emitting unit 10 are arranged at other regular intervals. In one embodiment, the regular interval arrangement may include that the intervals of at least three point light sources in the light-emitting unit 10 increase with a certain length d, and exemplarily, the intervals between two adjacent point light sources are L1, (L1+d), (L1+2d), (L1+3d)... In another embodiment, the regular interval arrangement may include that the intervals of at least three point light sources in the light-emitting unit 10 decrease with a certain length d, and exemplarily, the intervals between two adjacent point light sources are L1, (L1-d), (L1-2d), (L1-3d)... In another embodiment, the regular interval arrangement may include that the intervals of at least three point light sources in the light-emitting unit 10 satisfy a preset linear function, such as a waveform function. In a specific example, based on the waveform in the rectangular coordinate system, the waveform can be equally divided on the x-axis according to the number of point light sources (such as 5), and 5 x values ​​can be determined. Further, the y values ​​corresponding to the 5 x values ​​are determined, and the difference between two adjacent y values ​​can correspond to the spacing between two adjacent point light sources.

[0051] Figure 2b A schematic diagram of the arrangement of point light sources provided in another embodiment of the present application.

[0052] Reference Figure 2bAs shown, the light emitting unit 10 may include five point light sources arranged side by side, namely, a point light source 10a, a point light source 10b, a point light source 10c, a point light source 10d, and a point light source 10e. The spacing between the five point light sources is set to meet Figure 2b The sine wave band shown. Specifically, five evenly distributed points can be determined in the x-axis direction of the sine wave band based on the number of point light sources (5) to achieve equal division of the waveform on the x-axis. Five x values ​​(x1, x2, x3, x4, x5) can be determined by the equal division, and further, the y values ​​(y1, y2, y3, y4, y5) corresponding to the five x values ​​can be determined based on the linear function of the sine wave band. Among them, y1, y2, y3, y4, y5 can correspond to point light source 10a, point light source 10b, point light source 10c, point light source 10d, and point light source 10e, respectively. The difference between two adjacent y values ​​can correspond to the spacing between two corresponding adjacent point light sources. Specifically, (y2-y1) can be the spacing between point light source 10a and point light source 10b, (y3-y2) can be the spacing between point light source 10c and point light source 10b, (y4-y3) can be the spacing between point light source 10c and point light source 10d, and (y5-y4) can be the spacing between point light source 10d and point light source 10e.

[0053] In some embodiments, the point light sources of the light emitting unit 10 are arranged at irregular intervals.

[0054] Figure 2c A schematic diagram of the arrangement of point light sources provided in another embodiment of the present application.

[0055] Reference Figure 2c As shown, the light-emitting unit 10 may include 5 point light sources arranged side by side, namely, point light source 10a, point light source 10b, point light source 10c, point light source 10d, and point light source 10e, and the distance between two adjacent light sources in the 5 point light sources may be respectively: the distance between the point light source 10a and the point light source 10b is L1, the distance between the point light source 10b and the point light source 10c is L2, the distance between the point light source 10c and the point light source 10d is L3, and the distance between the point light source 10d and the point light source 10e is L4. In one embodiment, the distance L1 to L4 can be a distance set by the user, and the installation position of the point light source is set according to the distance. In other embodiments, the distance between the point light sources can also be set by other methods to achieve an irregular arrangement, and the present application does not limit the specific implementation method.

[0056] In some embodiments, the specific implementation of the light-emitting unit 10 irradiating the target wire 50 multiple times at different positions may include at least three point light sources in the light-emitting unit 10 lighting up one by one and irradiating the target wire 50. Exemplarily, the light-emitting unit 10 may include 5 point light sources arranged side by side, namely, point light source 10a, point light source 10b, point light source 10c, point light source 10d, and point light source 10e. In response to the light-emitting instruction, the point light source 10a, point light source 10b, point light source 10c, point light source 10d, and point light source 10e light up one by one, respectively irradiating the target wire 50, and each irradiation can present a projection of the target wire 50 on the collection area 20a of the collection unit 20. Then, the collection unit 20 can collect and obtain the projection image of the target wire 50 during the irradiation of the currently lit point light source. It should be noted that only one point light source is lit each time, and it is extinguished after the irradiation is completed, and the next point light source is lit and irradiated, and so on.

[0057] In some embodiments, the lighting order of the point light sources in the light-emitting unit 10 can be set by user needs, and in response to the lighting instruction, the point light sources in the light-emitting unit 10 light up one by one and illuminate the target wire 50 based on the lighting order set by the user.

[0058] Figure 3 A schematic diagram of the lighting sequence of point light sources provided in one embodiment of the present application.

[0059] Reference Figure 3 As shown, the light emitting unit 10 may include 5 point light sources arranged side by side, namely, point light source 10a, point light source 10b, point light source 10c, point light source 10d, and point light source 10e. When the lighting order of the point light sources pre-set by the user is: point light source 10a is the first order (T1), point light source 10b is the second order (T2), point light source 10c is the third order (T3), point light source 10d is the fourth order (T4), and point light source 10e is the fifth order (T5), the five point light sources light up one by one in the lighting order to illuminate the target wire 50. In the actual application scenario, when the target wire 50 is a cylindrical wire and the placement posture is not perpendicular to the horizontal plane, the horizontal cross-section of the target wire 50 is an elliptical shape. Based on the posture of the target wire 50, different point light sources will present different projections at different positions of the collection area 20a to be collected by the collection unit after irradiation.

[0060] In some embodiments, during the process of a point light source lighting up and irradiating the target wire 50, the acquisition unit may collect at least once the projection of the target wire 50 presented in the acquisition area 20a. During the process of the current point light source lighting up and irradiating the target wire 50, if the acquisition unit 20 collects multiple projection images, thereby obtaining the projection image set of the current point light source, the acquisition unit may filter out the projection image with the highest definition from the projection image set of the current point light source as the target projection image. In other embodiments, the acquisition unit 20 may also send the corresponding projection image set collected when each point light source lights up and irradiates the target wire 50 to the computing unit 30, so that the computing unit filters out the projection image with the highest definition from each projection image set as the target projection image corresponding to the corresponding point light source.

[0061] In some embodiments, the acquisition unit 20 may include a linear charge-coupled device (CCD) sensor (hereinafter referred to as a linear CCD sensor). Specifically, the linear CCD sensor includes a row of photoelectric sensors, each of which has the function of sensing light and converting it into a corresponding electrical signal. Exemplarily, when the resolution of the linear CCD sensor is 128, 128 photoelectric sensors are arranged in a row. When illuminated, each photodiode generates a current according to the perceived light intensity, and the current signal is then converted into a voltage signal through an integration circuit. Among them, each photodiode has its own integration circuit, and each integration circuit can be regarded as a pixel. The grayscale value of the image collected by each pixel is proportional to the intensity sensed by the photoelectric sensor at the position point. In other words, the greater the light intensity sensed by the photoelectric sensor, the closer the pixel grayscale at the position point is to full white. Conversely, the smaller the light intensity sensed by the photoelectric sensor, the closer the pixel grayscale at the position point is to full black. When the point light source is turned on to illuminate the target wire 50 and presents a projection on the acquisition area 20a, the linear CCD sensor can acquire a one-dimensional image with a resolution of 128, and the one-dimensional image contains grayscale values ​​of 128 pixels. It should be noted that in other embodiments, the resolution of the linear CCD sensor can be different, and this application does not limit the resolution.

[0062] In some embodiments, after the acquisition unit 20 completes the acquisition, that is, after acquiring the projection or projection image set of the target wire corresponding to each point light source when it is lit, the projection image or projection image set is sent to the calculation unit 30, so that the calculation unit 30 calculates the positioning parameters of the target wire 50 based on the received projection or projection image set. When the calculation unit 30 receives multiple projection image sets, the projection image with the highest definition can be selected from each projection image set as the target projection image, and then the positioning parameters of the target wire 50 are calculated based on the selected multiple target projection images.

[0063] In some embodiments, the computing unit 30 may include a control module and a preset neural network model.

[0064] In some embodiments, after the calculation unit 30 receives multiple projection images sent by the acquisition unit, the control module performs superposition preprocessing on the multiple projection images received to obtain a detection image, and inputs the detection image into the preset neural network model, thereby calculating the positioning parameters of the target wire through the preset neural network model.

[0065] Figure 4 A schematic diagram of constructing a detection image provided for one embodiment of the present application.

[0066] Reference Figure 4 As shown, five projections presented by five point light sources in the order of lighting up (T1 to T5) are collected, and from top to bottom are the projection images collected in the order of lighting up T1 to T5. The one-dimensional projection images collected when each point light source is irradiated can be sequentially superimposed according to the order of irradiation to obtain a detection image.

[0067] In some embodiments, the training process of the preset neural network model may include the following steps:

[0068] S101: Obtain historical detection data.

[0069] S102: Divide the first part of the historical detection data as a training set to train the neural network model, and obtain a trained first neural network model.

[0070] S103: Divide the second part of the historical detection data as a test set to test and calibrate the first neural network model until the mean square error loss (LOSS) is lower than the preset accuracy, thereby obtaining a preset neural network model.

[0071] About the specific implementation of S101:

[0072] In different scenarios, when wires of different diameters are placed at different positions and postures on the detection surface, the cross-sections of the wires on the detection surface are different. To improve the comprehensiveness of the data, the historical detection data includes detection images and data parameters of wires of different diameters placed at different positions and postures on the detection surface. The data parameter types include the cross-sectional diameter, position information, and normal vector angle of the wire relative to the detection surface.

[0073] Figure 5 A schematic diagram of a wire cross section in historical detection data provided by one embodiment of the present application.

[0074] Reference Figure 5As shown, the historical detection data may include actual detection data of wires of different diameters. Specifically, in the process of acquiring historical detection data, detection images and data parameters of wires of the same diameter placed in different positions and postures on the detection surface may be acquired. The detection images may be compared with Figure 4 The detection images provided by the illustrated embodiments are the same or similar. The types of data parameters may include the cross-sectional diameter, position information, and normal vector angle of the wire relative to the detection surface. Then, by arranging the data, a detection data list as shown in Table 1 below is obtained:

[0075] Table 1

[0076]

[0077] Referring to the test data list shown in Table 1, the actual diameter of the wire is d, and each line of test data includes the cross-sectional diameter, position coordinates, and normal vector angle of the wire on the test surface obtained when the wire is placed at a position on the test surface in a certain posture. And each line of data also corresponds to the test image of the wire in the current state. In one embodiment, the test data can be associated with the corresponding test image by image numbering.

[0078] In other embodiments, the type of data parameters may include the cross-sectional diameter and position information of the wire relative to the detection surface. Then, by sorting the data, a detection data list as shown in Table 2 below is obtained:

[0079] Table 2

[0080]

[0081] Referring to the test data list shown in Table 2, the actual diameter of the wire is d, and each line of test data contains the cross-sectional diameter and position coordinates of the wire on the test surface obtained when the wire is placed at a position on the test surface in a certain posture. And each line of data also corresponds to a test image of the wire in the current state. In one embodiment, the test data can be associated with the corresponding test image by means of image numbering.

[0082] In some embodiments, the detection data and corresponding detection images of wires with different diameters can be obtained in the above manner, and the detection data can be associated with the detection image to obtain historical detection data.

[0083] Since the historical detection data obtained may have a certain regular distribution, if the regularly distributed detection data is imported into the neural network model for training, the neural network model may fall into overfitting. In some embodiments, in order to reduce the possibility of the neural network model falling into overfitting, the historical detection data obtained may be shuffled. For example, the position and posture data of the wire in the historical detection data may be randomly distributed.

[0084] About the specific implementation of S102 and SS103:

[0085] In some embodiments, the first portion of data accounts for a first proportion in the historical detection data, and the second portion of data accounts for a second proportion in the historical detection data, wherein the first proportion is greater than the second proportion. Exemplarily, the historical detection data can be divided in a ratio of 9:1, thereby obtaining a first portion of data accounting for 90% of the historical detection data and a second portion of data accounting for 10% of the historical detection data. Then, the first portion of data is used as a training set for training, and the second portion of data is used as a test set for testing.

[0086] In some embodiments, before dividing the first part of the data in the historical detection data as a training set to train the neural network model and obtaining the trained first neural network model, it also includes: performing image recognition processing on each detection image in the historical detection data to obtain the recognized detection image; wherein the image recognition processing process includes: based on a preset grayscale range, identifying the effective projection area in the detection image.

[0087] Figure 6 A schematic diagram of effective shadow area identification provided for one embodiment of the present application.

[0088] Reference Figure 6 As shown in the figure, taking the one-dimensional projection image corresponding to the T1 order in the detection image as an example, the projection of the wire has a single connected dark area on the linear CCD sensor. The one-dimensional projection image can be divided into a shadow area (including the single connected dark area) and a non-shadow area, where the non-shadow area is a completely white area, that is, Figure 6 The grayscale value shown is the area of ​​G = 255. The shadow area contains the effective projection area and the invalid projection area. The effective projection area is the actual projection area of ​​the wire, that is, Figure 6 The region 601 shown in the figure has a pixel grayscale value G1 corresponding to the region 601, and the pixel grayscale value G1 falls within the preset grayscale range. The invalid projection area is when there are areas with higher brightness values ​​on both sides of the target wire that are not completely blocked, that is, Figure 6 As shown in the area 602 and the area 603, the grayscale value of the pixel corresponding to the area 602 is G2, and the grayscale value of the pixel corresponding to the area 603 is G3.

[0089] Identifying the effective projection area in the detection image can specifically be filtering out Figure 6 The area 601 shown. Specifically, the area falling within the preset grayscale range in each one-dimensional projection image in the detection image can be screened out. First, the all-white area with a grayscale value of G=255 can be eliminated. Further, the area in the shadow area whose grayscale value does not fall within the preset grayscale range is eliminated, that is, area 602 and area 603 are eliminated. Among them, the preset grayscale range may include one or more preset grayscale values. Exemplarily, when the preset grayscale range only contains the grayscale value G=0, the area with a non-zero grayscale value can be eliminated, and the remaining area that meets the conditions is the valid projection area, that is, the area 601 with a grayscale value of G1. Among them, it should be noted that the preset grayscale range can be customized according to the actual application scenario and user needs, and this application does not limit this.

[0090] After obtaining the effective projection area of ​​each one-dimensional projection image in the detection image, the endpoint position and length of each effective projection area can be obtained. Specifically, the projection length and the two end positions can be determined based on the pixel position corresponding to the effective projection area. Further, by connecting the position of the point light source to the two ends of the effective projection area, two tangents of the circular (elliptical or circular) cross section of the wire can be obtained. In the above manner, two tangents can be obtained according to each one-dimensional projection image. When the light-emitting unit includes at least three point light sources, a detection image includes at least three one-dimensional projection images, and at least 6 tangents can be obtained. The position information of the three point light sources and the tangent lengths of the three point light sources connecting to the same circle are currently known, and the parameters of the circular cross section of the wire can be calculated. The parameters can specifically include:

[0091] (x, y, d1, d2, θ)

[0092] Among them, the arrangement direction of the point light source on the collection area 20a can be taken as the X axis, and the direction from the midpoint of the linear CCD sensor to the point light source can be taken as the Y axis to establish a plane rectangular coordinate system. (x, y) represents the coordinate position of the elliptical cross section on the detection surface, (d1) represents the first diameter (i.e., the major axis) of the elliptical cross section, (d2) represents the second diameter (i.e., the minor axis) of the elliptical cross section, and (θ) represents the angle between the first diameter d1 and the Y axis.

[0093] After the above training, the neural network model can learn the relationship between the effective projection area in the detection image and the circular cross-section of the wire. Furthermore, the neural network model can be trained to learn the ability to calculate the normal vector angle between the wire and the detection surface based on the first diameter and the second diameter of the elliptical cross-section of the wire. Specifically, the corresponding calculation formula can be input into the model so that the model acquires the calculation ability, and finally the trained first neural network model is obtained.

[0094] After the first neural network model, the first neural network model can be further tested and calibrated through a test set until the mean square error loss (LOSS) is lower than a preset accuracy, thereby obtaining a preset neural network model.

[0095] In other embodiments, the preset neural network model can also output the parameters (x, y, d1, d2, θ) of the circular cross-section of the wire based on the input detection image. Furthermore, the control module of the calculation unit outputs the normal vector angle of the first diameter d1 and the second diameter d2 of the elliptical cross-section of the target wire relative to the detection surface based on the preset neural network model output data.

[0096] In some embodiments, the specific implementation of the calculation unit 30 calculating the positioning parameters of the target wire 50 based on the multiple projection images may include the following steps:

[0097] S101: Input multiple projection images into a preset neural network model.

[0098] In some embodiments, after receiving multiple projection images sent by the acquisition unit or screening out multiple target projection images based on the projection image set sent by the acquisition unit, the computing unit can input the multiple projection images or the multiple target projection images into a preset neural network model, so that the preset neural network model calculates the positioning parameters of the target wire 50 based on the input projection images. For the convenience of explanation, multiple projection images are used as an example for explanation below.

[0099] S102: The preset neural network model outputs position information and a cross-sectional diameter of the target wire on a detection surface based on a plurality of projection images, wherein the position information includes coordinate information of the target wire relative to the detection surface, the detection surface is a horizontal plane where the point light source is located, and the cross-sectional diameter of the target wire includes a first diameter and a second diameter, wherein the first diameter is the longest cross-sectional diameter of the target wire, and the second diameter is the shortest cross-sectional diameter of the target wire.

[0100] In some embodiments, the preset neural network model can calculate the position information of the target wire and the cross-sectional diameter on the detection surface based on the detection image. The detection surface can be the horizontal plane where the point light source is located, that is, the point light source emits light on the horizontal plane to illuminate the target wire 50. The position information may include the coordinate information of the target wire relative to the detection surface.

[0101] Figure 7 A schematic diagram of a wire cross-section provided for one embodiment of the present application.

[0102] Reference Figure 7As shown, in one embodiment, the arrangement direction of the point light source on the collection area 20a can be taken as the X-axis, and the direction from the midpoint of the linear CCD sensor to the point light source can be taken as the Y-axis to establish a plane rectangular coordinate system. The preset neural network model can output the coordinates (x, y) of the target wire relative to the detection surface based on the detection image, as well as the cross-sectional diameter of the target wire on the detection surface, and the cross-sectional diameter can include a first diameter d1 and a second diameter d2, wherein the first diameter d1 is the longest diameter of the cross section, and the second diameter d2 is the shortest diameter of the cross section. Furthermore, the preset neural network model can also output the angle θ between the first diameter d1 and the Y-axis. In one embodiment, the cross section of the target wire on the detection surface can be expressed as (x, y, d1, d2, θ).

[0103] S103: Preset the neural network model to calculate the posture information based on the cross-sectional diameter.

[0104] In some embodiments, the preset neural network model can also calculate and output the normal vector angle of the target wire relative to the detection surface based on the interface diameter in the obtained cross-sectional information (x, y, d1, d2, θ) of the target wire on the cross section, wherein the posture information can be the normal vector angle of the target wire relative to the detection surface.

[0105] In some embodiments, the normal vector angle of the target wire relative to the detection surface can be calculated based on the first diameter and the second diameter. Specifically, a first ratio is calculated based on the second diameter and the first diameter, and the arc cosine value of the first ratio is calculated. The arc cosine value of the first ratio is the normal vector angle of the target wire relative to the detection surface.

[0106] In one implementation, it can be calculated by the following formula:

[0107]

[0108] Wherein, γ represents the normal vector angle of the target wire relative to the detection surface, d1 represents the first diameter, and d2 represents the second diameter.

[0109] In some embodiments, the specific content of S103 may also be: the calculation unit calculates the posture information based on the cross-sectional diameter output by the preset neural network model.

[0110] The method of calculating the posture information based on the image by the calculation unit can be the same or similar to the method of calculating the posture information based on the cross-sectional diameter by the preset neural network model, which will not be described in detail here.

[0111] In some embodiments, a preset angle may be preset so that the calculated normal vector angle is compared with the preset angle to determine whether the posture of the target wire meets the preset conditions.

[0112] In some embodiments, the wire positioning detection device provided in the embodiments of the present application may further include a comparison unit for determining whether the normal vector angle of the target wire relative to the detection surface is less than a preset angle, wherein when the normal vector angle is less than the preset angle, it is determined that the posture of the target wire meets the preset conditions; when the normal vector angle is not less than the preset angle, it is determined that the posture of the target wire does not meet the preset conditions.

[0113] Figure 8 A schematic diagram of the normal vector angle of a target wire provided in one embodiment of the present application.

[0114] Reference Figure 8 As shown, the normal vector angle of the target wire 50 relative to the detection surface P1 is γ. In some application scenarios, the height distance H between the working surface P2 of the target wire 50 and the detection surface P1 is fixed. Therefore, after calculating the normal vector angle of the target wire 50 relative to the detection surface, it is possible to determine whether the endpoint of the target wire 50 on the working surface is within the preset working range based on the normal vector angle. In other words, it is determined whether the posture of the target wire 50 meets the preset conditions through the comparison result between the normal vector angle and the preset angle. If the preset conditions are met, it can be determined that the endpoint of the target wire 50 on the working surface is within the preset working range. Otherwise, if the preset conditions are not met, it can be determined that the endpoint of the target wire 50 on the working surface is not within the preset working range.

[0115] In some embodiments, the wire positioning detection device provided in the embodiments of the present application may further include an alarm unit, which is used to send a wire posture deviation alarm message when the comparison unit determines that the posture of the target wire does not meet the preset conditions. This allows the user to quickly understand whether the current posture of the wire meets the preset conditions. When an alarm occurs, the use of the wire can be stopped to reduce the impact of the wire posture deviation.

[0116] In some embodiments, the wire positioning detection device provided in the embodiments of the present application may further include a display unit and a button, wherein the display unit may be connected to the computing unit, and when the computing unit determines that the current posture of the target wire does not meet the preset conditions, a wire posture deviation alarm message is sent, and the wire posture deviation alarm message may be displayed on the display unit, thereby enabling the user to promptly discover the deviation of the target wire. In some embodiments, the display unit may also display corresponding menu information so that the user can click on the menu on the display unit to set parameters. Among them, the button may be a power button of the wire positioning detection device, which controls the turning on or off of the wire positioning detection device. In some embodiments, the button may also control the turning on and off of the light-emitting unit and the control mode switching, such as controlling the entry into the training mode to train the model.

[0117] In some embodiments, the present application also provides an external electronic device for connecting to a wire positioning detection device, the wire positioning detection device comprising: a light emitting unit, a collection unit, the light emitting unit and the collection unit are arranged relative to each other, a detection placement area is formed between the light emitting unit and the collection unit, and the detection placement area is configured to place the target wire to be detected; the light emitting unit is used to respond to the light emitting instruction of the calculation unit, and illuminate the target wire multiple times at different positions to present multiple projections of the target wire in the collection area of ​​the collection unit; the collection unit is used to obtain multiple projection images and send them to the external electronic device, wherein the external electronic device is used to calculate the positioning parameters of the target wire based on the multiple projection images, and the positioning parameters include the position information and posture information of the target wire. The device can provide a collection function, that is, the use image of the target wire is obtained through the light emitting unit and the collection unit, and the obtained multiple projection images are provided to the external electronic device for calculating the positioning parameters of the target wire.

[0118] In some embodiments, the present application also provides an electronic device, which (i.e., the aforementioned external electronic device) can be connected to the aforementioned wire positioning detection device, and the electronic device includes a preset neural network model and a control module; the electronic device calculates the positioning parameters of the target wire based on the multiple projection images provided by the wire positioning detection device, specifically including: the control module performs superposition preprocessing on the multiple projection images received to obtain a detection image, and inputs the detection image into the preset neural network model; the preset neural network model outputs position information and a cross-sectional diameter of the target wire on the detection surface based on the detection image, wherein the position information includes coordinate information of the target wire relative to the detection surface, the detection surface is a horizontal plane where the point light source is located, and the cross-sectional diameter of the target wire includes a first diameter and a second diameter, wherein the first diameter is the longest cross-sectional diameter of the target wire, and the second diameter is the shortest cross-sectional diameter of the target wire; the preset neural network model calculates posture information based on the cross-sectional diameter.

[0119] It should be noted that the terminals involved in the embodiments of the present application may include but are not limited to personal computers (PCs), personal digital assistants (PDAs), wireless handheld devices, tablet computers, mobile phones, MP3 players, MP4 players, etc.

[0120] It is understandable that the application may be an application program (naturalApp) installed on the terminal, or may also be a web page program (webApp) of a browser on the terminal, and this embodiment of the present application does not limit this.

[0121] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0122] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0123] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0124] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0125] The above-mentioned integrated unit implemented in the form of a software functional unit can be stored in a computer-readable storage medium. The above-mentioned software functional unit is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor (Processor) to perform some steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), disk or optical disk and other media that can store program codes.

[0126] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

[0127] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

Claims

1. A wire positioning detection device, characterized in that: The device comprises: A light emitting unit, a collection unit and a calculation unit, wherein the light emitting unit and the collection unit are arranged opposite to each other, a detection placement area is formed between the light emitting unit and the collection unit, and the detection placement area is configured to place a target wire to be detected; The light emitting unit is used to respond to the light emitting instruction of the computing unit and illuminate the target wire material multiple times at different positions, so as to present multiple projections of the target wire material in the collection area of ​​the collection unit; The acquisition unit is used to acquire the images of the multiple projections and send them to the calculation unit; The calculation unit is used to calculate the positioning parameters of the target wire based on the multiple projected images, and the positioning parameters include position information and posture information of the target wire.

2. The device according to claim 1, characterized in that The light emitting unit comprises at least three point light sources arranged side by side; The irradiating the target wire material multiple times at different positions includes: the at least three point light sources emit light one by one to irradiate the target wire material.

3. The device according to claim 2, characterized in that The arrangement direction of the at least three point light sources is consistent with the extension direction of the collection area of ​​the collection unit; and / or The distance between any two adjacent ones of the at least three point light sources is the same.

4. The device according to any one of claims 1 to 3, characterized in that The computing unit includes a preset neural network model and a control module; The calculating the positioning parameters of the target wire based on the multiple projected images includes: The control module performs superposition preprocessing on the received multiple projected images to obtain a detection image, and inputs the detection image into the preset neural network model; The preset neural network model outputs the position information and the cross-sectional diameter of the target wire on the detection surface based on the detection image, wherein the position information includes coordinate information of the target wire relative to the detection surface, the detection surface is the horizontal plane where the point light source is located, and the cross-sectional diameter of the target wire includes a first diameter and a second diameter, wherein the first diameter is the longest cross-sectional diameter of the target wire, and the second diameter is the shortest cross-sectional diameter of the target wire; The preset neural network model calculates the posture information based on the cross-sectional diameter.

5. The device according to claim 4, characterized in that The training process of the preset neural network model includes: Dividing the first part of the historical detection data as a training set to train the neural network model, thereby obtaining a first neural network model after training; Divide the second part of the historical detection data as a test set to test and calibrate the first neural network model until the mean square error loss is lower than the preset accuracy, thereby obtaining the preset neural network model; The historical detection data includes detection images and data parameters of wires of different diameters placed at different positions of the detection surface in different postures, and the data parameters include the cross-sectional diameter, position information and normal vector angle of the wire relative to the detection surface; The data of the position and posture of the wire in the historical detection data are randomly distributed.

6. The device according to claim 5, characterized in that Before the first part of the data in the divided historical detection data is used as a training set to train the neural network model to obtain the trained first neural network model, the method further includes: Performing image recognition processing on each detection image in the historical detection data to obtain a recognized detection image; The image recognition process includes: Based on a preset grayscale range, an effective projection area in the detection image is identified.

7. The device according to claim 5, characterized in that The preset neural network model calculates the posture information based on the cross-sectional diameter, including: The normal vector angle of the target wire relative to the detection surface is calculated based on the first diameter and the second diameter, wherein the posture information is the normal vector angle of the target wire relative to the detection surface.

8. The device according to claim 7, characterized in that The calculating the normal vector angle of the target wire relative to the detection surface based on the first diameter and the second diameter comprises: Calculating a first ratio based on the second diameter and the first diameter; The arc cosine value of the first ratio is calculated to obtain the normal vector angle of the target wire relative to the detection surface.

9. The device according to claim 7, characterized in that The device also includes: A comparison unit, used to determine whether the normal vector angle is less than a preset angle; Wherein, when the normal vector angle is less than the preset angle, it is determined that the posture of the target wire meets the preset condition; When the normal vector angle is not less than the preset angle, it is determined that the posture of the target wire does not meet the preset condition.

10. A wire positioning detection device, characterized in that: The device is externally connected to an electronic device, and the device comprises: A light emitting unit and a collection unit, wherein the light emitting unit and the collection unit are arranged opposite to each other, and a detection placement area is formed between the light emitting unit and the collection unit, and the detection placement area is configured to place a target wire to be detected; The light emitting unit is used to respond to the light emitting instruction of the computing unit and illuminate the target wire material multiple times at different positions, so as to present multiple projections of the target wire material in the collection area of ​​the collection unit; The acquisition unit is used to acquire the multiple projected images and send them to the external electronic device, wherein the external electronic device is used to calculate the positioning parameters of the target wire based on the multiple projected images, and the positioning parameters include position information and posture information of the target wire.

11. An electronic device, characterized in that: Connected to the wire positioning detection device of claim 10, the electronic device includes a preset neural network model and a control module; The electronic device calculates the positioning parameters of the target wire based on the multiple projected images provided by the wire positioning detection device, specifically including: The control module performs superposition preprocessing on the received multiple projected images to obtain a detection image, and inputs the detection image into the preset neural network model; The preset neural network model outputs the position information and the cross-sectional diameter of the target wire on the detection surface based on the detection image, wherein the position information includes coordinate information of the target wire relative to the detection surface, the detection surface is a horizontal plane where the point light source is located, and the cross-sectional diameter of the target wire includes a first diameter and a second diameter, wherein the first diameter is the longest cross-sectional diameter of the target wire, and the second diameter is the shortest cross-sectional diameter of the target wire; The preset neural network model calculates the posture information based on the cross-sectional diameter.

Images