Non-magnetic metal classification device and method for reducing upper surface flatness influence
Through image recognition technology equivalent to flat upper surface and using the phases of eddy current sensors and mutual inductance signals for classification, the problem of uneven upper surface in non-magnetic metal classification technology affecting classification accuracy is solved, and a high-accuracy classification effect is achieved.
Patent Information
- Application Number
- CN202510445087.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-04-10
- Publication Date
- 2025-05-23
AI Technical Summary
When the existing non-magnetic metal classification technology deals with uneven sample blocks on the upper surface, the classification accuracy is low and is greatly affected by the flatness of the upper surface.
Image recognition technology is used to equivalently equilibrate the uneven upper surface of the non-magnetic metal sample block into a flat upper surface, and classify it through the phases of the eddy current sensor and mutual inductance signal to reduce the influence of the flatness of the upper surface on the classification results.
It effectively improves the accuracy of non-magnetic metal classification, especially when the upper surface of the sample block is uneven, the classification accuracy can reach 96.67%.
Smart Images

Figure CN120023111A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a non-magnetic metal classification technology, in particular to a non-magnetic metal classification device and method that reduce the influence of upper surface flatness. Background Art
[0002] In the process of sorting scrap metals, magnetic metals can be sorted out relatively easily. The remaining non-magnetic metals, mainly composed of copper, aluminum, zinc, tin, lead, titanium, etc., usually have higher recycling value. At present, in the recycling process of non-magnetic metals, the eddy current method is gradually being used in metal classification due to its fast response speed and non-destructive characteristics. However, in practical applications, the classification results of this method are greatly affected by the flatness of the upper surface of the non-magnetic metal sample block. Specifically, when the upper surface of the non-magnetic metal sample block is uneven, the distance and angle between the sensor and the measured surface (i.e., the upper surface of the non-magnetic metal sample block) are constantly changing, which will seriously affect the accuracy of classification. Based on this, it is necessary to invent a non-magnetic metal classification device and method that reduces the influence of the upper surface flatness to solve the problem of low classification accuracy of the existing non-magnetic metal classification technology when the upper surface of the non-magnetic metal sample block is uneven. Summary of the invention
[0003] In order to solve the problem that the classification accuracy of the existing non-magnetic metal classification technology is low when the upper surface of the non-magnetic metal sample block is uneven, the present invention provides a non-magnetic metal classification device and method that reduce the influence of the upper surface flatness.
[0004] The present invention is achieved by adopting the following technical solutions:
[0005] A non-magnetic metal sorting device for reducing the influence of upper surface flatness, comprising a belt conveyor, two support rods, a mounting plate, an eddy current sensor, a beam-type photoelectric switch, a camera, a signal generator, a signal conditioner, and a host computer;
[0006] The belt conveyor is fixed horizontally on the ground; each support rod is L-shaped, and each support rod includes a horizontal section and a vertical section arranged downward; the vertical sections of the two support rods are respectively fixed on the ground on both sides of the belt conveyor; the horizontal sections of the two support rods are both located above the belt conveyor; the mounting plate is fixed horizontally between the horizontal ends of the two support rods, and the mounting plate is located above the belt conveyor;
[0007] The eddy current sensor includes an excitation coil and a receiving coil which are coaxially arranged; the excitation coil is horizontally fixed to the lower surface of the mounting plate; the receiving coil is horizontally fixed to the upper surface of the mounting plate; the transmitter and receiver of the opposing photoelectric switch are respectively fixed to the vertical section sides of two support rods; the camera is fixed to the vertical section side of one of the support rods; the host computer is electrically connected to the excitation coil through a signal generator; the receiving coil is electrically connected to the host computer through a signal conditioner; the receiver and camera of the opposing photoelectric switch are both electrically connected to the host computer.
[0008] Furthermore, the vertical sections of the two support rods are telescopic structures with adjustable lengths; the mounting plate is an acrylic plate with a thickness of 2 mm, and its height can be adjusted with the lengths of the vertical sections of the two support rods; the excitation coil and the receiving coil are cylindrical hollow coils of the same size, with an inner diameter of 20 mm, an outer diameter of 20.2 mm, a height of 3.5 mm, and 31 turns.
[0009] A non-magnetic metal classification method for reducing the influence of upper surface flatness is implemented based on a non-magnetic metal classification device for reducing the influence of upper surface flatness according to the present invention. The method is implemented by the following steps:
[0010] Step 1: Start the through-beam photoelectric switch, and the transmitter of the through-beam photoelectric switch emits a light beam to the receiver;
[0011] Step 2: placing the non-magnetic metal sample block with an uneven upper surface to be classified on the upper surface of the conveyor belt of the belt conveyor, and the non-magnetic metal sample block moves along the conveyor belt of the belt conveyor;
[0012] Step 3: When the non-magnetic metal sample block moves between the vertical sections of the two support rods, the light beam is blocked by the non-magnetic metal sample block, and the receiver of the opposing photoelectric switch outputs a switch signal. The host computer controls the camera and the signal generator to start according to the switch signal; then, the camera collects the image of the side of the non-magnetic metal sample block and transmits the collected image to the host computer; at the same time, the signal generator outputs a single-frequency excitation signal; the single-frequency excitation signal is transmitted to the excitation coil, so that an eddy current is induced in the non-magnetic metal sample block, thereby inducing a mutual inductance signal in the receiving coil; the mutual inductance signal is conditioned by the signal conditioner and transmitted to the host computer; the host computer demodulates the mutual inductance signal, thereby obtaining the real part M of the mutual inductance signal. R and the imaginary part M I ;
[0013] Step 4: Perform Canny edge detection on the image of the side of the non-magnetic metal sample block to obtain the pixel area of the side of the non-magnetic metal sample block, and convert the pixel area into the actual area; then, divide the actual area of the side of the non-magnetic metal sample block by the length of the bottom side to obtain the equivalent height of the side of the non-magnetic metal sample block, thereby converting the uneven upper surface of the non-magnetic metal sample block into a flat upper surface;
[0014] Step 5: According to the real part M of the mutual inductance signal R and the imaginary part M I , calculate the phase p of the mutual inductance signal and use it as the characteristic value of the non-magnetic metal sample block; the specific calculation formula is as follows:
[0015]
[0016] Step six: using the equivalent height of the side of the non-magnetic metal sample block as the horizontal coordinate and the characteristic value of the non-magnetic metal sample block as the vertical coordinate, determine the characteristic coordinate point of the non-magnetic metal sample block; then, respectively calculate the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block to the first calibration characteristic curve, the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block to the second calibration characteristic curve, the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block to the third calibration characteristic curve, the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block to the fourth calibration characteristic curve, the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block to the fifth calibration characteristic curve, and the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block to the sixth calibration characteristic curve;
[0017] Step 7: According to the calculation results of step 6, the material of the non-magnetic metal sample block is determined, and the non-magnetic metal sample block is classified accordingly; the specific determination rules are as follows:
[0018] When the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block to the first calibration characteristic curve is less than or equal to 0.0015, the material of the non-magnetic metal sample block is determined to be titanium;
[0019] When the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block to the second calibration characteristic curve is less than or equal to 0.0015, the material of the non-magnetic metal sample block is determined to be lead;
[0020] When the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block to the third calibration characteristic curve is less than or equal to 0.0015, the material of the non-magnetic metal sample block is determined to be tin;
[0021] When the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block to the fourth calibration characteristic curve is less than or equal to 0.0015, the material of the non-magnetic metal sample block is determined to be zinc;
[0022] When the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block to the fifth calibration characteristic curve is less than or equal to 0.0015, the material of the non-magnetic metal sample block is determined to be aluminum;
[0023] When the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block to the sixth calibration characteristic curve is less than or equal to 0.0015, the material of the non-magnetic metal sample block is determined to be copper.
[0024] Furthermore, the excitation frequency of the single-frequency excitation signal is 60kHz, and the excitation voltage is 1V.
[0025] Furthermore, the first calibration characteristic curve is obtained by the following steps: selecting a plurality of titanium sample blocks with flat upper surfaces, the same length and width but different heights, and measuring the height and characteristic value of each titanium sample block, and then fitting the first calibration characteristic curve using the height as the independent variable and the characteristic value as the dependent variable.
[0026] Furthermore, the second calibration characteristic curve is obtained by the following steps: selecting multiple lead sample blocks with flat upper surfaces, the same length and width but different heights, and measuring the height and characteristic value of each lead sample block, and then fitting the second calibration characteristic curve using the height as the independent variable and the characteristic value as the dependent variable.
[0027] Furthermore, the third calibration characteristic curve is obtained by the following steps: selecting a plurality of tin sample blocks with flat upper surfaces, the same length and width but different heights, and measuring the height and characteristic value of each tin sample block, and then fitting the third calibration characteristic curve with the height as the independent variable and the characteristic value as the dependent variable.
[0028] Furthermore, the fourth calibration characteristic curve is obtained by the following steps: selecting a plurality of zinc sample blocks with flat upper surfaces, the same length and width but different heights, and measuring the height and characteristic value of each zinc sample block, and then fitting the fourth calibration characteristic curve with the height as the independent variable and the characteristic value as the dependent variable.
[0029] Furthermore, the fifth calibration characteristic curve is obtained by the following steps: selecting multiple aluminum sample blocks with flat upper surfaces, the same length and width but different heights, and measuring the height and characteristic value of each aluminum sample block, and then fitting the fifth calibration characteristic curve with the height as the independent variable and the characteristic value as the dependent variable.
[0030] Furthermore, the sixth calibration characteristic curve is obtained by the following steps: selecting multiple copper sample blocks with flat upper surfaces, the same length and width but different heights, and measuring the height and characteristic value of each copper sample block, and then fitting the sixth calibration characteristic curve with the height as the independent variable and the characteristic value as the dependent variable.
[0031] Compared with the existing non-magnetic metal classification technology, the present invention uses image recognition technology to convert the uneven upper surface of the non-magnetic metal sample block into a flat upper surface, and uses the phase of the mutual inductance signal as the classification basis, thereby effectively reducing the influence of the flatness of the upper surface of the non-magnetic metal sample block on the classification result, thereby greatly improving the classification accuracy. Experiments show that when the upper surface of the non-magnetic metal sample block is uneven, the classification accuracy of the present invention is as high as 96.67%.
[0032] The invention effectively solves the problem that when the upper surface of the non-magnetic metal sample block is uneven, the classification accuracy of the existing non-magnetic metal classification technology is low, and is suitable for the classification of non-magnetic metals. BRIEF DESCRIPTION OF THE DRAWINGS
[0033] Figure 1 It is a partial structural schematic diagram of the device described in the present invention.
[0034] Figure 2 yes Figure 1 Top view of the .
[0035] Figure 3 yes Figure 1 Left view of .
[0036] Figure 4 It is another partial structural schematic diagram of the device of the present invention.
[0037] Figure 5 It is a schematic diagram of step four in the method of the present invention.
[0038] Figure 6 It is a schematic diagram of the first to sixth calibration characteristic curves in the method of the present invention.
[0039] In the figure: 1-belt conveyor, 2-support rod, 3-mounting plate, 401-excitation coil, 402-receiving coil, 501-transmitter of the opposite-beam photoelectric switch, 502-receiver of the opposite-beam photoelectric switch, 6-camera, 7-signal generator, 8-signal conditioner, 9-host computer, 10-non-magnetic metal sample block; S represents the actual area of the side of the non-magnetic metal sample block; L represents the bottom side length of the side of the non-magnetic metal sample block; H represents the equivalent height of the side of the non-magnetic metal sample block. DETAILED DESCRIPTION
[0040] A non-magnetic metal classification device for reducing the influence of upper surface flatness, comprising a belt conveyor 1, two support rods 2, a mounting plate 3, an eddy current sensor, a beam-type photoelectric switch, a camera 6, a signal generator 7, a signal conditioner 8, and a host computer 9;
[0041] The belt conveyor 1 is fixed horizontally on the ground; each support rod 2 is L-shaped, and each support rod 2 includes a horizontal section and a vertical section arranged downward; the vertical sections of the two support rods 2 are respectively fixed on the ground on both sides of the belt conveyor 1; the horizontal sections of the two support rods 2 are both located above the belt conveyor 1; the mounting plate 3 is fixed horizontally between the horizontal ends of the two support rods 2, and the mounting plate 3 is located above the belt conveyor 1;
[0042] The eddy current sensor includes a coaxially arranged excitation coil 401 and a receiving coil 402; the excitation coil 401 is horizontally fixed to the lower surface of the mounting plate 3; the receiving coil 402 is horizontally fixed to the upper surface of the mounting plate 3; the transmitter 501 and the receiver 502 of the opposing photoelectric switch are respectively fixed to the vertical section sides of the two support rods 2; the camera 6 is fixed to the vertical section side of one of the support rods 2; the host computer 9 is electrically connected to the excitation coil 401 through the signal generator 7; the receiving coil 402 is electrically connected to the host computer 9 through the signal conditioner 8; the receiver 502 and the camera 6 of the opposing photoelectric switch are both electrically connected to the host computer 9.
[0043] The vertical sections of the two support rods 2 are telescopic structures with adjustable lengths; the mounting plate 3 is an acrylic plate with a thickness of 2 mm, and its height can be adjusted with the lengths of the vertical sections of the two support rods 2; the excitation coil 401 and the receiving coil 402 are cylindrical hollow coils of the same size, with an inner diameter of 20 mm, an outer diameter of 20.2 mm, a height of 3.5 mm, and 31 turns.
[0044] A non-magnetic metal classification method for reducing the influence of upper surface flatness is implemented based on a non-magnetic metal classification device for reducing the influence of upper surface flatness according to the present invention. The method is implemented by the following steps:
[0045] Step 1: Start the beam-type photoelectric switch, and the transmitter 501 of the beam-type photoelectric switch transmits a light beam to the receiver 502;
[0046] Step 2: placing the non-magnetic metal sample block 10 with an uneven upper surface to be classified on the upper surface of the conveyor belt of the belt conveyor 1, and the non-magnetic metal sample block 10 moves along the conveyor belt of the belt conveyor 1;
[0047] Step 3: When the non-magnetic metal sample block 10 moves between the vertical sections of the two support rods 2, the light beam is blocked by the non-magnetic metal sample block 10, and the receiver 502 of the beam-type photoelectric switch outputs a switch signal. The host computer 9 controls the camera 6 and the signal generator 7 to start according to the switch signal; then, the camera 6 collects the image of the side of the non-magnetic metal sample block 10, and transmits the collected image to the host computer 9; at the same time, the signal generator 7 outputs a single-frequency excitation signal; the single-frequency excitation signal is transmitted to the excitation coil 401, so that an eddy current is induced in the non-magnetic metal sample block 10, thereby inducing a mutual inductance signal in the receiving coil 402; the mutual inductance signal is conditioned by the signal conditioner 8 and transmitted to the host computer 9; the host computer 9 demodulates the mutual inductance signal, thereby obtaining the real part M of the mutual inductance signal. R and the imaginary part M I ;
[0048] Step 4: Perform Canny edge detection on the image of the side of the non-magnetic metal sample block 10 to obtain the pixel area of the side of the non-magnetic metal sample block 10, and convert the pixel area into the actual area; then, divide the actual area of the side of the non-magnetic metal sample block 10 by the length of the bottom side to obtain the equivalent height of the side of the non-magnetic metal sample block 10, thereby converting the uneven upper surface of the non-magnetic metal sample block 10 into a flat upper surface;
[0049] Step 5: According to the real part M of the mutual inductance signal R and the imaginary part M I , calculate the phase p of the mutual inductance signal and use it as the characteristic value of the non-magnetic metal sample block 10; the specific calculation formula is as follows:
[0050]
[0051] Step six: using the equivalent height of the side of the non-magnetic metal sample block 10 as the horizontal coordinate and the characteristic value of the non-magnetic metal sample block 10 as the vertical coordinate, determine the characteristic coordinate point of the non-magnetic metal sample block 10; then, respectively calculate the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block 10 to the first calibration characteristic curve, the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block 10 to the second calibration characteristic curve, the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block 10 to the third calibration characteristic curve, the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block 10 to the fourth calibration characteristic curve, the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block 10 to the fifth calibration characteristic curve, and the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block 10 to the sixth calibration characteristic curve;
[0052] Step 7: According to the calculation result of step 6, the material of the non-magnetic metal sample block 10 is determined, and the non-magnetic metal sample block 10 is classified accordingly; the specific determination rules are as follows:
[0053] When the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block 10 to the first calibration characteristic curve is less than or equal to 0.0015, the material of the non-magnetic metal sample block 10 is determined to be titanium;
[0054] When the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block 10 to the second calibration characteristic curve is less than or equal to 0.0015, the material of the non-magnetic metal sample block 10 is determined to be lead;
[0055] When the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block 10 to the third calibration characteristic curve is less than or equal to 0.0015, the material of the non-magnetic metal sample block 10 is determined to be tin;
[0056] When the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block 10 to the fourth calibration characteristic curve is less than or equal to 0.0015, the material of the non-magnetic metal sample block 10 is determined to be zinc;
[0057] When the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block 10 to the fifth calibration characteristic curve is less than or equal to 0.0015, the material of the non-magnetic metal sample block 10 is determined to be aluminum;
[0058] When the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block 10 to the sixth calibration characteristic curve is less than or equal to 0.0015, the material of the non-magnetic metal sample block 10 is determined to be copper.
[0059] The excitation frequency of the single-frequency excitation signal is 60kHz and the excitation voltage is 1V.
[0060] The first calibration characteristic curve is obtained by the following steps: selecting a plurality of titanium sample blocks with flat upper surfaces, the same length and width but different heights, and measuring the height and characteristic value of each titanium sample block, and then fitting the first calibration characteristic curve using the height as the independent variable and the characteristic value as the dependent variable.
[0061] The second calibration characteristic curve is obtained by the following steps: selecting multiple lead sample blocks with flat upper surfaces, the same length and width but different heights, and measuring the height and characteristic value of each lead sample block, and then fitting the second calibration characteristic curve using the height as the independent variable and the characteristic value as the dependent variable.
[0062] The third calibration characteristic curve is obtained by the following steps: selecting a plurality of tin sample blocks with flat upper surfaces, the same length and width but different heights, and measuring the height and characteristic value of each tin sample block, and then fitting the third calibration characteristic curve with the height as the independent variable and the characteristic value as the dependent variable.
[0063] The fourth calibration characteristic curve is obtained by the following steps: selecting a plurality of zinc sample blocks with flat upper surfaces, the same length and width but different heights, and measuring the height and characteristic value of each zinc sample block, and then fitting the fourth calibration characteristic curve using the height as the independent variable and the characteristic value as the dependent variable.
[0064] The fifth calibration characteristic curve is obtained by the following steps: selecting multiple aluminum sample blocks with flat upper surfaces, the same length and width but different heights, and measuring the height and characteristic value of each aluminum sample block, and then fitting the fifth calibration characteristic curve with the height as the independent variable and the characteristic value as the dependent variable.
[0065] The sixth calibration characteristic curve is obtained by the following steps: selecting a plurality of copper sample blocks with flat upper surfaces, the same length and width but different heights, and measuring the height and characteristic value of each copper sample block, and then fitting the sixth calibration characteristic curve using the height as the independent variable and the characteristic value as the dependent variable.
[0066] Although the specific embodiments of the present invention are described above, it should be understood by those skilled in the art that these are only examples, and the protection scope of the present invention is defined by the appended claims. Those skilled in the art may make various changes or modifications to these embodiments without departing from the principles and essence of the present invention, but these changes and modifications all fall within the protection scope of the present invention.
Claims
1. A non-magnetic metal sorting device for reducing the influence of upper surface flatness, characterized in that: It comprises a belt conveyor (1), two support rods (2), a mounting plate (3), an eddy current sensor, a beam-type photoelectric switch, a camera (6), a signal generator (7), a signal conditioner (8), and a host computer (9); The belt conveyor (1) is fixed horizontally on the ground; each support rod (2) is L-shaped, and each support rod (2) comprises a horizontal section and a vertical section arranged downward; the vertical sections of the two support rods (2) are respectively fixed on the ground on both sides of the belt conveyor (1); the horizontal sections of the two support rods (2) are both located above the belt conveyor (1); the mounting plate (3) is fixed horizontally between the horizontal ends of the two support rods (2), and the mounting plate (3) is located above the belt conveyor (1); The eddy current sensor comprises an excitation coil (401) and a receiving coil (402) which are coaxially arranged; the excitation coil (401) is horizontally fixed to the lower surface of a mounting plate (3); the receiving coil (402) is horizontally fixed to the upper surface of the mounting plate (3); the transmitter (501) and the receiver (502) of the opposing photoelectric switch are respectively fixed to the vertical section side surfaces of two support rods (2); the camera (6) is fixed to the vertical section side surface of one of the support rods (2); the host computer (9) is electrically connected to the excitation coil (401) through a signal generator (7); the receiving coil (402) is electrically connected to the host computer (9) through a signal conditioner (8); and the receiver (502) and the camera (6) of the opposing photoelectric switch are both electrically connected to the host computer (9).
2. A non-magnetic metal classification device for reducing the influence of upper surface flatness according to claim 1, characterized in that: The vertical sections of the two support rods (2) are telescopic structures with adjustable lengths; the mounting plate (3) is an acrylic plate with a thickness of 2 mm, and its height can be adjusted according to the lengths of the vertical sections of the two support rods (2); the excitation coil (401) and the receiving coil (402) are cylindrical hollow coils of the same size, with an inner diameter of 20 mm, an outer diameter of 20.2 mm, a height of 3.5 mm, and 31 turns.
3. A non-magnetic metal classification method for reducing the influence of upper surface flatness, the method is implemented based on the non-magnetic metal classification device for reducing the influence of upper surface flatness as claimed in claim 1, characterized in that: This method is implemented by the following steps: Step 1: activating the beam-type photoelectric switch, wherein the transmitter (501) of the beam-type photoelectric switch transmits a light beam to the receiver (502); Step 2: placing a non-magnetic metal sample block (10) with an uneven upper surface to be classified on the upper surface of a conveyor belt of a belt conveyor (1), and the non-magnetic metal sample block (10) moves along the conveyor belt of the belt conveyor (1); Step 3: When the non-magnetic metal sample block (10) moves between the vertical sections of the two support rods (2), the light beam is blocked by the non-magnetic metal sample block (10), and the receiver (502) of the beam-to-beam photoelectric switch outputs a switch signal. The host computer (9) controls the camera (6) and the signal generator (7) to start according to the switch signal; then, the camera (6) collects an image of the side of the non-magnetic metal sample block (10), and transmits the collected image to the host computer (9); at the same time, the signal generator (7) outputs a single-frequency excitation signal; the single-frequency excitation signal is transmitted to the excitation coil (401), so that an eddy current is induced in the non-magnetic metal sample block (10), thereby inducing a mutual induction signal in the receiving coil (402); the mutual induction signal is conditioned by the signal conditioner (8) and transmitted to the host computer (9); the host computer (9) demodulates the mutual induction signal, thereby obtaining the real part M of the mutual induction signal. R and the imaginary part M I ; Step 4: Performing Canny edge detection on the image of the side of the non-magnetic metal sample block (10) to obtain the pixel area of the side of the non-magnetic metal sample block (10), and converting the pixel area into an actual area; then, dividing the actual area of the side of the non-magnetic metal sample block (10) by the length of the bottom side to obtain the equivalent height of the side of the non-magnetic metal sample block (10), thereby converting the uneven upper surface of the non-magnetic metal sample block (10) into an equivalent flat upper surface; Step 5: According to the real part M of the mutual inductance signal R and the imaginary part M I , calculate the phase p of the mutual inductance signal and use it as the characteristic value of the non-magnetic metal sample block (10); The specific calculation formula is as follows: Step six: using the equivalent height of the side of the non-magnetic metal sample block (10) as the horizontal coordinate and the characteristic value of the non-magnetic metal sample block (10) as the vertical coordinate, determining the characteristic coordinate point of the non-magnetic metal sample block (10); then, respectively calculating the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block (10) to the first calibration characteristic curve, the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block (10) to the second calibration characteristic curve, the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block (10) to the third calibration characteristic curve, the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block (10) to the fourth calibration characteristic curve, the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block (10) to the fifth calibration characteristic curve, and the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block (10) to the sixth calibration characteristic curve; Step 7: According to the calculation result of step 6, the material of the non-magnetic metal sample block (10) is determined, thereby classifying the non-magnetic metal sample block (10); the specific determination rules are as follows: When the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block (10) to the first calibration characteristic curve is less than or equal to 0.0015, the material of the non-magnetic metal sample block (10) is determined to be titanium; When the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block (10) to the second calibration characteristic curve is less than or equal to 0.0015, the material of the non-magnetic metal sample block (10) is determined to be lead; When the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block (10) to the third calibration characteristic curve is less than or equal to 0.0015, the material of the non-magnetic metal sample block (10) is determined to be tin; When the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block (10) to the fourth calibration characteristic curve is less than or equal to 0.0015, the material of the non-magnetic metal sample block (10) is determined to be zinc; When the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block (10) to the fifth calibration characteristic curve is less than or equal to 0.0015, the material of the non-magnetic metal sample block (10) is determined to be aluminum; When the longitudinal distance from the characteristic coordinate point of the non-magnetic metal sample block (10) to the sixth calibration characteristic curve is less than or equal to 0.0015, the material of the non-magnetic metal sample block (10) is determined to be copper.
4. A non-magnetic metal classification method for reducing the influence of upper surface flatness according to claim 3, characterized in that: The excitation frequency of the single-frequency excitation signal is 60kHz and the excitation voltage is 1V.
5. A non-magnetic metal classification method for reducing the influence of upper surface flatness according to claim 3, characterized in that: The first calibration characteristic curve is obtained by the following steps: selecting a plurality of titanium sample blocks with flat upper surfaces, the same length and width but different heights, and measuring the height and characteristic value of each titanium sample block, and then fitting the first calibration characteristic curve using the height as the independent variable and the characteristic value as the dependent variable.
6. A non-magnetic metal classification method for reducing the influence of upper surface flatness according to claim 3, characterized in that: The second calibration characteristic curve is obtained by the following steps: selecting multiple lead sample blocks with flat upper surfaces, the same length and width but different heights, and measuring the height and characteristic value of each lead sample block, and then fitting the second calibration characteristic curve using the height as the independent variable and the characteristic value as the dependent variable.
7. A non-magnetic metal classification method for reducing the influence of upper surface flatness according to claim 3, characterized in that: The third calibration characteristic curve is obtained by the following steps: selecting a plurality of tin sample blocks with flat upper surfaces, the same length and width but different heights, and measuring the height and characteristic value of each tin sample block, and then fitting the third calibration characteristic curve with the height as the independent variable and the characteristic value as the dependent variable.
8. A non-magnetic metal classification method for reducing the influence of upper surface flatness according to claim 3, characterized in that: The fourth calibration characteristic curve is obtained by the following steps: selecting a plurality of zinc sample blocks with flat upper surfaces, the same length and width but different heights, and measuring the height and characteristic value of each zinc sample block, and then fitting the fourth calibration characteristic curve using the height as the independent variable and the characteristic value as the dependent variable.
9. A non-magnetic metal classification method for reducing the influence of upper surface flatness according to claim 3, characterized in that: The fifth calibration characteristic curve is obtained by the following steps: selecting multiple aluminum sample blocks with flat upper surfaces, the same length and width but different heights, and measuring the height and characteristic value of each aluminum sample block, and then fitting the fifth calibration characteristic curve with the height as the independent variable and the characteristic value as the dependent variable.
10. A non-magnetic metal classification method for reducing the influence of upper surface flatness according to claim 3, characterized in that: The sixth calibration characteristic curve is obtained by the following steps: selecting a plurality of copper sample blocks with flat upper surfaces, the same length and width but different heights, and measuring the height and characteristic value of each copper sample block, and then fitting the sixth calibration characteristic curve using the height as the independent variable and the characteristic value as the dependent variable.