Metal foreign matter detection method, metal foreign matter detection equipment and storage medium

Abstract

The invention relates to the technical field of metal foreign matter detection, in particular to a metal foreign matter detection method, metal foreign matter detection equipment and a storage medium, and the method comprises the steps: obtaining the detection statistics of a to-be-detected product; according to the detection statistical magnitude, on the basis of a preset point taking rule, calculating to obtain average noise power; according to the average noise power, a detection threshold value is obtained through calculation; and judging whether the detection statistical magnitude is greater than the detection threshold value, if so, determining that the detection result contains the metal foreign matters, and otherwise, determining that the detection result does not contain the metal foreign matters. The method has the advantages that the precision and accuracy of metal foreign matter detection are improved, and meanwhile the detection rate and the false alarm rate can be taken into account under the condition of the low signal-to-noise ratio.

Description

Technical Field

[0001] The present application relates to the technical field of metal foreign object detection, and in particular, to a method for detecting metal foreign objects, a metal foreign object detection device, and a storage medium. Background Art

[0002] At present, metal foreign object detection is widely used in the quality control links of industries such as food, medicine, and chemical industry. Detection equipment usually works based on the principle of electromagnetic induction. By detecting the change in the electromagnetic field between the transmitting and receiving coils, it is determined whether the tested product contains metal foreign objects. The level of detection accuracy is directly related to product safety and compliance.

[0003] The detection signal of a metal detector contains various noises. The change in the electromagnetic field detected by the receiving coil in the probe is very small, such as a signal in the range of nV to μV. After a series of circuit processing and amplification, the final detection signal is obtained, but this process will introduce various noises, including the thermal noise of electronic devices, power supply noise, interference noise from other circuits on the circuit board received by the demodulation circuit, etc. At the same time, the probe receiving coil, as a receiver, will also pick up some external noises, and these noises will all appear in the final detection signal.

[0004] Existing metal foreign object detection technologies mainly adopt a fixed threshold judgment method, that is, comparing the detection signal with a preset threshold. When the signal amplitude exceeds the threshold, it is determined that there is a metal foreign object. To avoid false alarms, the threshold is usually set relatively conservatively. However, when the metal foreign object is small and the signal-to-noise ratio is low, the above-mentioned noises may affect metal detection, resulting in an increase in the false alarm rate. To ensure a zero false alarm rate, existing methods often suppress the noise by increasing the threshold, but this method will cause a significant decrease in detection accuracy, and thus lead to the missed detection of small metal foreign objects. Therefore, there is room for improvement. Summary of the Invention

[0005] In order to improve the accuracy and accuracy of metal foreign object detection, and at the same time, the detection rate and false alarm rate can be balanced under low signal-to-noise ratio conditions, the present application provides a method for detecting metal foreign objects, a metal foreign object detection device, and a storage medium.

[0006] In a first aspect, the present application provides a method for detecting metal foreign objects, including: Obtaining a detection statistic of a product to be tested; Calculating an average noise power based on a preset sampling rule according to the detection statistic; Calculating a detection threshold according to the average noise power; Judging whether the detection statistic is greater than the detection threshold. If so, the detection result is that there is a metal foreign object; otherwise, the detection result is that there is no metal foreign object.

[0007] By adopting the above technical solution, by obtaining the detection statistic of the product to be tested and calculating the average noise power based on a preset sampling rule, the noise level in the current detection environment can be dynamically evaluated, thereby improving the anti-interference ability and adaptability of the detection result. By calculating the detection threshold according to the average noise power, the detection threshold can be matched with the current environmental noise level, thereby avoiding false alarms caused by noise changes, improving the detection accuracy, and then determining whether the detection statistic is greater than the detection threshold, which can accurately determine whether there is a metallic foreign object. A high detection rate and a low false alarm rate can be achieved even in a low signal-to-noise ratio environment, ensuring the accuracy of the detection result, reducing the possibility of missed detection, and thus improving the precision and accuracy of metallic foreign object detection.

[0008] Optionally, the detection statistic follows an exponential distribution.

[0009] By adopting the above technical solution, since the detection statistic follows an exponential distribution, the change law of the detection statistic can be accurately reflected according to the characteristics of the exponential distribution, thereby providing a reliable statistical basis for subsequent threshold calculation.

[0010] Optionally, the detection threshold is expressed as , where T is the detection threshold, α is the threshold adjustment coefficient, and Z is the average noise power.

[0011] By adopting the above technical solution, by introducing the threshold adjustment coefficient, the threshold can be dynamically adjusted on the premise of a given false alarm rate, avoiding the inadaptability problem caused by using a fixed threshold, thereby optimizing the decision boundary in actual detection, improving the accuracy of metallic foreign object detection, and achieving an increase in the detection rate on the premise of a given false alarm rate.

[0012] Optionally, the obtaining of the detection statistic of the product to be tested specifically includes: Collecting a plurality of time-domain detection signal sequences of the product to be tested through the receiving coil of the metal detection device; Performing coherent demodulation processing on the time-domain detection signal sequences respectively to obtain the detection statistic.

[0013] By adopting the above technical solution, by collecting a plurality of time-domain signal sequences through the receiving coil of the metal detection device, the electromagnetic reflection or scattering characteristics of the product to be tested at different frequencies can be comprehensively covered, thereby improving the detection sensitivity and ensuring the all-round monitoring of metallic foreign objects. Furthermore, by performing coherent demodulation processing on the time-domain detection signal sequences, the background noise interference can be effectively removed, the signal components related to metallic foreign objects can be retained, the risk of misjudgment caused by signal distortion can be reduced, and thus the recognition ability of weak metallic foreign object signals can be enhanced and the detection sensitivity can be improved.

[0014] Optionally, based on the detection statistic and a preset sampling rule, the average noise power is calculated as follows: In the sampling sequence corresponding to the detection statistic, a detection point is selected as the current detection point, and with the current detection point as the center, several symmetric sampling points are selected from the sampling sequence; Based on the noise power calculation formula, the average noise power corresponding to the symmetric sampling points is calculated according to the several symmetric sampling points.

[0015] By adopting the above technical solution, by selecting a detection point as the center in the sampling sequence corresponding to the detection statistic and extracting several symmetric sampling points from the sampling sequence, representative local noise data can be extracted in space or time, thereby improving the stability and representativeness of noise estimation. Furthermore, based on the noise power calculation formula, the average noise power in the symmetric sampling points is calculated, and the true noise level corresponding to this detection point can be obtained. On this basis, the detection threshold is calculated to ensure the detection accuracy and reduce the negative impact of environmental noise on the detection accuracy.

[0016] Optionally, the average noise power is expressed as , where Z is the average noise power of the symmetric sampling points, W is the number of symmetric sampling points, i is the i-th sampling point, and N i represents the noise power corresponding to the i-th sampling point.

[0017] By adopting the above technical solution, expressing the average noise power as the average of the noise values of several sampling points can smooth the interference of extreme values caused by instantaneous fluctuations, thereby improving the stability and accuracy of the average noise power estimation and reducing the impact of noise fluctuations on the detection threshold calculation in a low signal-to-noise ratio environment.

[0018] Optionally, calculating the detection threshold based on the average noise power includes the following steps: Obtain a preset target false alarm rate, and determine a threshold adjustment coefficient based on the fact that the detection statistic follows an exponential distribution and the target false alarm rate; Calculate the detection threshold corresponding to the detection statistic according to the threshold adjustment coefficient and the average noise power.

[0019] By adopting the above technical solution, by obtaining a preset target false alarm rate and combining that the detection statistic follows an exponential distribution, a threshold adjustment coefficient is determined, which can reasonably model and control the false alarm risk in a statistical sense, thereby improving the robustness and false judgment suppression ability of the detection method in a low signal-to-noise ratio environment. By multiplying the threshold adjustment coefficient by the average noise power to obtain the detection threshold, it can dynamically adapt to the change of the noise level in different detection scenarios, thereby enhancing the self-adaptability and reliability of the judgment threshold, and thus improving the detection ability for various metal foreign objects.

[0020] Optionally, when determining the threshold adjustment coefficient, there is , so there is , where α is the threshold adjustment coefficient, W is the number of symmetric sampling points, and P fa is the target false alarm rate.

[0021] By adopting the above technical solution, by deriving a formula to determine the threshold adjustment coefficient, it is possible to make a judgment on the premise that the overall of multiple samples does not exceed the threshold, thereby ensuring that the overall false alarm rate meets the preset target at the mathematical level. By clarifying the functional relationship between the threshold adjustment coefficient, the false alarm rate, and the number of samples through formula transformation, it is possible to accurately adjust the detection threshold in the case of a low signal-to-noise ratio, thereby increasing the detection rate of metal foreign objects and effectively controlling the false alarm rate at the same time.

[0022] In a second aspect, the present application provides a metal foreign object detection device, including a memory, a processor, and a program stored in the memory and executable on the processor. When the processor executes the program, the steps of the above metal foreign object detection method are implemented.

[0023] In a third aspect, the present application provides a storage medium, where the storage medium stores a program, and when the program is executed by a processor, the steps of the above metal foreign object detection method are implemented.

[0024] In summary, the present application includes at least one of the following beneficial technical effects: 1. By obtaining the detection statistic of the product to be tested and calculating the average noise power based on a preset sampling rule, the noise level in the current detection environment can be dynamically evaluated, thereby improving the anti-interference ability and adaptability of the detection result. By calculating the detection threshold according to the average noise power, the detection threshold can be matched with the current environmental noise level, thereby avoiding false alarms caused by noise changes, improving the detection accuracy. Furthermore, by judging whether the detection statistic is greater than the detection threshold, it is possible to accurately judge whether there is a metal foreign object, and a high detection rate and a low false alarm rate can be achieved even in a low signal-to-noise ratio environment, ensuring the accuracy of the detection result, reducing the possibility of missed detection, and thus the accuracy and precision of metal foreign object detection; 2. By introducing a threshold adjustment coefficient, the threshold can be dynamically adjusted on the premise of a given false alarm rate, avoiding the inadaptability problem caused by using a fixed threshold, thereby optimizing the decision boundary in actual detection, improving the accuracy of metal foreign object detection, and achieving an increase in the detection rate on the premise of a given false alarm rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 is a flowchart of the implementation of the method for obtaining a balanced signal in an embodiment of the present application; Figure 2 is a flowchart of the implementation of the method for processing a balanced signal in an embodiment of the present application; Figure 3 is a flowchart of the implementation of a method for detecting metal foreign objects in an embodiment of the present application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0026] The following embodiments will help those skilled in the art further understand the role of the present application, but do not limit the present application in any form. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can be made. These all belong to the protection scope of the present application.

[0027] In the following description, for the purpose of illustration rather than limitation, specific details such as specific system structures and technologies are presented to thoroughly understand the embodiments of the present application. However, those skilled in the art should clearly understand that the present application can also be implemented in other embodiments without these specific details. In other cases, the detailed descriptions of well-known systems, devices, circuits, and methods are omitted to avoid unnecessary details from interfering with the description of the present application.

[0028] It should be understood that when used in the specification and appended claims of the present application, the term "comprising" indicates the presence of the described features, wholes, steps, operations, elements, and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0029] The present application will be further described in detail below with reference to the accompanying drawings.

[0030] A method for detecting metal foreign objects provided in this embodiment can balance the detection rate and false alarm rate under low signal-to-noise ratio conditions, and improve the precision and accuracy of metal foreign object detection.

[0031] Generally, it can be through such as Figure 1 and 2The detection signal is obtained in the manner shown. The transmitting signal is amplified by the power amplifier circuit and then sent to the transmitting coil through the transmitting transformer, which excites an electromagnetic field of a certain intensity in the probe. The receiving coil senses the change of the electromagnetic field, and its output enters the receiving amplifier circuit through the receiving transformer and outputs a balanced signal, as shown in FIG. Figure 1 As shown, the balanced signal is processed by the demodulation circuit to obtain the detection signal.

[0032] Among them, for balanced signals, common processing methods are as follows Figure 2 As shown, an orthogonal carrier signal is outputted through a carrier signal generator, so that a balanced signal and a carrier signal and a balanced signal and another carrier signal are respectively inputted into different calculation channels for signal demodulation, signal filtering and signal amplification, so as to obtain a first detection signal DT1 and a second detection signal DT2. Then DT1 and DT2 are sampled by an AD sampling circuit and sent to a controller. The controller processes the sampled DT1 and DT2, generally based on a preset threshold value as a judgment standard. If DT1 or DT2 exceeds the threshold value, it is judged that a foreign object is detected.

[0033] Furthermore, when no product passes, the detection signal generated is a noise signal, which represents the real part and imaginary part of the detection signal respectively, and DT1 and DT2 are combined to form a complex Gaussian noise DT, DT = DT1 + jDT2. For the convenience of expression, Expressed as follows, where X, I, Q represent DT, DT1, DT2 respectively, and j is the imaginary unit.

[0034] Furthermore, for complex Gaussian noise processed by the coherent demodulation circuit, X generally obeys a complex Gaussian distribution, that is, I and Q obey Gaussian distributions respectively, I~N(0,σ 2 ), Q~N(0,σ 2 ), the power of X (I 2 +Q 2 ) obeys exponential distribution. For the complex Gaussian noise processed by the incoherent demodulation circuit, its modulus obeys Rayleigh distribution, that is, |X| obeys Rayleigh distribution, which can be converted to obey exponential distribution through the square law detector. In short, no matter which demodulation method is used, the complex Gaussian noise signal obtained by balanced signal processing, the noise power (I 2 +Q 2 ) can be considered to obey exponential distribution.

[0035] In metal detection, the presence of noise can cause false alarms and missed detections. To ensure a false alarm rate of zero, the threshold is often increased to suppress noise. However, this approach significantly reduces the detection accuracy, leading to missed detections of small metal foreign objects and failing to balance the detection rate and false alarm rate. Therefore, in this embodiment, the lowest missed detection rate is achieved within the given upper limit of the false alarm rate, that is, the highest detection probability is achieved. In statistics, from the perspective of simple binary hypothesis testing, the following assumptions can be made: H0: No metal foreign object is detected in the detection signal.

[0036] H1: A metal foreign object is detected in the detection signal.

[0037] And each time the detection result is judged, the following four situations will occur:

[0038] Among them, the false alarm rate is , and the detection probability is , x is the value of the sample to be measured, that is, the sampling data of a group of detection signals, and T is the decision threshold.

[0039] The essence of the above solution is to find a decision threshold T to achieve the goal of maximizing the detection probability under the given false alarm rate. Therefore, in this embodiment, the balanced signal is represented by a sampling sequence, and the balanced signal is denoted as BAL(n). After passing through the coherent demodulation module, namely the multiplier, low-pass filter, and amplifier, the detection statistic N is obtained, where N = X 2 , that is, the detection statistic is expressed as the above noise power. Therefore, the aforementioned assumptions can be expressed as: H0: N < T.

[0040] H1: N > T.

[0041] Referring to Figure 3 , Figure 3 is the implementation flowchart of a metal foreign object detection method in the embodiment of the present application. The method for judging the detection of metal foreign objects in this embodiment is as follows: S1. Obtain the detection statistic of the product to be measured. Among them, the detection statistic of the product to be measured is obtained as follows: Collect multiple time-domain detection signal sequences of the product to be measured through the receiving coil of the metal detection device; perform coherent demodulation processing on the time-domain detection signal sequences respectively to obtain the detection statistic.

[0042] Specifically, the product to be tested is conveyed through the detection area by a conveyor belt or via a pipeline, ensuring that there are no obstructions between it and the receiving coil. Then, the metal detection device emits electromagnetic waves through the receiving coil and records the reflected signals. Signal acquisition usually includes multiple time-domain detection signal sequences, which means that the device will repeat sampling at different time intervals to obtain multiple independent signal waveforms.

[0043] More specifically, according to the obtained time-domain detection signal sequences, perform coherent demodulation processing on the time-domain detection signal sequences. By using a multiplier, a low-pass filter, and an amplifier, eliminate the interference factors in the original signal and retain the useful signal part related to the reflection or scattering of metal foreign objects, thereby obtaining the detection statistic N.

[0044] Among them, the detection statistic follows an exponential distribution.

[0045] Specifically, from the above content, it can be known that the noise power of the complex Gaussian noise signal X follows an exponential distribution. Since N = X 2 , it can be known that N also follows an exponential distribution, which can be expressed as N ~ Exp(2σ 2 ).

[0046] S2. According to the detection statistic, calculate the average noise power based on a preset sampling rule.

[0047] Specifically, in the sampling sequence of the detection statistic N, for the current decision target point n, based on a preset sampling rule, for example, take a number of sampling points before and after the current detection point for symmetric distribution, ensuring that these sampling points are evenly distributed in time to reduce sampling errors. In the generated signal samples, calculate the mean of the amplitudes to generate the corresponding average noise power Z, which represents the noise level of the environment to be tested.

[0048] Among them, the determination of the average noise power Z is as follows. In the sampling sequence corresponding to the detection statistic, select a detection point as the current detection point, and based on the current detection point, select a number of symmetric sampling points from the sampling sequence. Then, based on the noise power calculation formula, calculate the average noise power corresponding to the symmetric sampling points according to the number of symmetric sampling points.

[0049] Specifically, take the current decision target point n as the current detection point. In the sampling sequence of the detection statistic N, with this current detection point n as the center, take a number of symmetric sampling points before and after, a total of W sampling points. For example, take 5 sampling points before and after, thus forming a set of 10 symmetric sampling points.

[0050] More specifically, then use the following formula to calculate the average noise power Z. Denote the background noise power of these W points as , where Z is the average noise power of symmetric sampling points, W is the number of symmetric sampling points, and N i represents the noise power corresponding to the i-th sampling point. Among them, since N follows an exponential distribution, that is, N i follows an exponential distribution, that is, N i ~Exp(2σ 2 ), and because N i are independent of each other, based on statistical principles, it can be deduced that the sum of N i follows a gamma distribution, that is , then the average noise power Z follows a gamma distribution, that is . In this way, the noise level of the surrounding environment can be reflected by the selected symmetric sampling points.

[0051] S3. Calculate the detection threshold according to the average noise power.

[0052] Specifically, through the calculated average noise power N and the preset adjustment coefficient α, the detection threshold T can be calculated through the formula .

[0053] Among them, the determination of the threshold adjustment coefficient α is as follows: obtain the preset target false alarm rate, determine the threshold adjustment coefficient based on the detection statistic following an exponential distribution and the target false alarm rate; calculate the detection threshold corresponding to the detection quantity according to the threshold adjustment coefficient and the average noise power.

[0054] Specifically, in the sampling sequence of the detection statistic N, symmetrically take sampling points before and after the current monitoring point n, a total of W points. Since the role of the threshold adjustment coefficient α is to make the probability that the value of each of these W points does not exceed the detection threshold T be P fa , that is, the preset target false alarm rate is P fa . Since , that is , since N~Exp(2σ 2 ), and the probability density function (PDF) of this exponential distribution is , then , so .

[0055] More specifically, since , and the moment generating function (MGF) of this gamma distribution is , let , substitute it into the moment generating function, then there is , so the threshold adjustment coefficient is determined α when there is: , thus obtaining: , where α is the threshold adjustment coefficient, W is the number of symmetric sampling points, and Pfa is the target false alarm rate.

[0056] Therefore, when the target false alarm rate P fa is a preset value and the number of symmetric sampling points W is known, the threshold adjustment coefficient α can be obtained such that the probability that the values of these W points do not exceed the detection threshold T is P fa , and the threshold adjustment coefficient α is independent of the standard deviation σ of the average noise power Z. Therefore, according to the above derivation, the value of the corresponding detection threshold T can be obtained as .

[0057] S4. Determine whether the detection statistic is greater than the detection threshold. If it is, the detection result is that there is a metallic foreign object; otherwise, the detection result is that there is no metallic foreign object.

[0058] Specifically, by determining whether the detection statistic is greater than the detection threshold, if it is, the detection result is that there is a metallic foreign object; otherwise, the detection result is that there is no metallic foreign object. Therefore, when the detection statistic N meets the conditional judgment formula, it is considered that a metallic foreign object is found in the detection result. Therefore, in actual detection, the detection threshold T obtained by solving in the above formula can be used, and since the detection probability , so when the given target false alarm rate P fa and the number of symmetric sampling points W are given, the solution of the detection threshold T can be realized, so that when the given target false alarm rate P fa and the detection threshold T are given, the detection probability P d can be obtained. Based on the Neyman - Pearson criterion, there can be a detection threshold T such that the maximum detection probability P fa is achieved under the given false alarm rate P d , where the detection probability P d can be calculated by the solution of the probability density function, which is not limited here.

[0059] More specifically, since the value of the threshold adjustment coefficient α is independent of the noise power, the method provided in this embodiment can achieve a certain detection accuracy even in the case of a low signal-to-noise ratio. Moreover, for the detection scenario where products continuously pass through, such as the sauce continuously flowing in a pipeline or the materials continuously moving on a conveyor belt, the detection effect caused by the products themselves can also be regarded as background noise. In this case, through the method of this application, the interference caused by the continuous product flow can be effectively eliminated, and the accuracy of metal foreign object detection can be improved. In addition, in the detection scenario with external environmental interference, such as electromagnetic compatibility (EMC) interference, mechanical vibration interference, etc., the method of this application can dynamically adjust the detection threshold. Even in the case of sudden changes in environmental noise, it can adapt to the changes in background noise in real time, so as to ensure that the metal foreign object detection device can operate under the lowest detection accuracy requirement, avoid the shutdown of the production line caused by continuous false alarms, and thus improve the applicability and stability of the metal foreign object detection device. An embodiment of this application provides a metal foreign object detection device, including a memory, a processor, and a program stored in the memory and executable on the processor. When the processor executes the program, it implements the metal foreign object detection method in the above embodiment.

[0060] An embodiment of this application provides a storage medium, on which a program is stored. When the program is executed by a processor, the following steps are implemented: Obtain the detection statistic of the product to be tested; Based on the detection statistic and a preset sampling rule, calculate the average noise power; Calculate the detection threshold according to the average noise power; Judge whether the detection statistic is greater than the detection threshold. If so, the detection result is that there is a metal foreign object; otherwise, the detection result is that there is no metal foreign object.

[0061] The above embodiments are only used to illustrate the technical solutions of this application, rather than to limit them; although this application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of this application, and should all be included in the protection scope of this application.

Claims

1. A method for detecting metal foreign objects, characterized in that, The metal foreign object detection method includes: Obtaining a detection statistic of a product to be detected; Calculating an average noise power based on a preset sampling rule according to the detection statistic; Calculating a detection threshold according to the average noise power; Determining whether the detection statistic is greater than the detection threshold. If so, the detection result is that there is a metal foreign object; otherwise, the detection result is that there is no metal foreign object.

2. The metal foreign object detection method according to claim 1, characterized in that The detection statistic follows an exponential distribution.

3. The metal foreign object detection method according to claim 1, characterized in that, The detection threshold is expressed as , where T is the detection threshold, α is the threshold adjustment coefficient, and Z is the average noise power.

4. The metal foreign object detection method according to claim 2, characterized in that The obtaining of the detection statistic of the product to be detected specifically includes: Collecting a plurality of time-domain detection signal sequences of the product to be detected through a receiving coil of a metal detection device; Performing coherent demodulation processing on the time-domain detection signal sequences respectively to obtain a detection statistic.

5. The metal foreign object detection method according to claim 4, characterized in that, Calculating the average noise power based on a preset sampling rule according to the detection statistic specifically includes: In the sampling sequence corresponding to the detection statistic, selecting a detection point as the current detection point, and selecting a plurality of symmetric sampling points from the sampling sequence with the current detection point as the center; Calculating the average noise power corresponding to the symmetric sampling points based on a noise power calculation formula according to the plurality of symmetric sampling points.

6. The metal foreign object detection method according to claim 3, characterized in that, The average noise power is expressed as , where Z is the average noise power of symmetric sampling points, W is the number of symmetric sampling points, i is the i-th sampling point, and N i represents the noise power corresponding to the i-th sampling point.

7. The metal foreign object detection method according to claim 5, characterized in that Calculating the detection threshold according to the average noise power specifically includes: Obtaining a preset target false alarm rate, and determining a threshold adjustment coefficient based on the fact that the detection statistic follows an exponential distribution and the target false alarm rate; Calculating the detection threshold corresponding to the detection statistic according to the threshold adjustment coefficient and the average noise power.

8. The metal foreign object detection method according to claim 7, wherein When determining the threshold adjustment coefficient, there is , so there is , where α is the threshold adjustment coefficient, W is the number of symmetric sampling points, and P fa is the target false alarm rate.

9. A metal foreign object detection device, comprising a memory, a processor, and a program stored in the memory and running on the processor, characterized in that, When the processor executes the program, the steps of the metal foreign object detection method according to any one of claims 1 to 8 are implemented.

10. A storage medium, the storage medium stores a program, characterized in that, When the program is executed by the processor, the steps of the metal foreign object detection method according to any one of claims 1 to 8 are implemented.

Images