OF SENSOR HEAD MOTION

TECHNICAL FIELD

This invention relates to a metal detector with one or more movement and/or position sensors.

INCORPORATION BY REFERENCE

This patent application claims priority from:

- Australian Provisional Patent Application No 2010905666 titled "Metal detector with improved functions" filed 24 December 2010.

The entire content of this application is hereby incorporated by reference.

BACKGROUND

Metal detectors are used to detect and determine the location of conductive objects that are buried in the ground. Examples of such objects include coins, relics, gold, mines and minerals. With one or more means attached to a metal detector to determine the position of the sensing head and/or the movement of the metal detector during the operation of the metal detector, the detection and the localisation performed by the metal detector may be improved.

A typical metal detector operates by transmitting electromagnetic energy into the ground where it interacts with a buried object, which in turn generates an electromagnetic energy. The sensing head of a metal detector, which includes a receiver, is used to receive and measure the electromagnetic energy from the buried object, which differs from an electromagnetic energy returned by the ground. Hence the sensing head can detect the presence of a buried object from the ground. A typical method of metal detection is to move the coil side to side over the ground to sweep an area of interest.

To provide extra functions to, or to enhance signal processing of a metal detector, one or more additional sensors can be attached to a metal detector. For example, a movement sensor or a position sensor can be attached to the sensor head of a metal detector to provide extra information regarding the received electromagnetic energy (also referred as receive field or receive signal). The extra information can be used to improve the performance of the metal detector, and to improve the interaction between the detector and the user, which improves the performance of the overall detecting process. BRIEF SUMMARY OF THE INVENTION

According to a first aspect of the present invention, there is provided a method for triggering one or more predetermined functions during a detection of a target in a soil, including: transmitting a transmit magnetic field; receiving a receive magnetic field using a magnetic field receiver to produce a receive signal; detecting a movement of the magnetic field receiver relative to the soil; analysing the movement to produce an analysis result; and performing, based on the analysis result, one or more of the one or more predetermined functions.

In one form, the one or more predetermined functions include a function to perform a compensation of an effect of the soil on the receive signal.

In one form, the analysing the movement is configured to detect a collision of the magnetic field receiver with an object external to the magnetic field receiver. Upon detection of the collision, in one form, a further processing of the receive signal, to produce an indicator output signal indicative of a presence of the target when the target is within the influence of the transmit magnetic field, is modified or abandoned.

In one form, the one or more predetermined functions include a function to store one or more characteristics of the target and of its environment when the target is within the influence of the transmit magnetic field.

In one form, the one or more predetermined functions include a function to reset or calibrate a processing of the receive signal to produce an indicator output signal indicative of a presence of the target when the target is within the influence of the transmit magnetic field.

According to a second aspect of the present invention, there is provided a method for detecting a target in a soil, including: transmitting a transmit magnetic field; receiving a receive magnetic field using a magnetic field receiver to produce a receive signal; associating the receive signal with positional references of the magnetic field receiver relative to the soil; processing one or more receive signals with one or more same or similar positional references, received over a period of time, to produce an indicator output signal indicating a target when the target is within the influence of the transmit magnetic field. In one form, the processing the one or more receive signals includes identifying different types of targets such that the indicator output signal is indicative of one or more of the different types of targets. According to a third aspect of the present invention, there is provided a method for detecting a target in a soil, including: transmitting a transmit magnetic field; receiving a receive magnetic field using a magnetic field receiver to produce a receive signal; associating the receive signal with positional references of the magnetic field receiver relative to the soil; processing one or more receive signals with one or more same or similar positional references, received over a period of time, to produce a cumulative receive signal; processing the cumulative receive signal to produce an indicator output signal indicating a target when the target is within the influence of the transmit magnetic field.

In one form, the processing the cumulative receive signal includes identifying different types of targets such that the indicator output signal is indicative of one or more of the different types of targets.

According to another aspect of the present invention, there is provided a metal detector configurable to perform the first, second and third aspects, and their various forms. To assist with the understanding of this invention, reference will now be made to the drawings:

BRIEF DESCRIPTION OF THE DRAWINGS

Figure 1 depicts a general block diagram of a metal detector with extra sensor(s);

Figure 2 depicts a general block diagram of the processing of accelerometer signals to generate acceleration, velocity and position reference signals;

Figure 3 depicts an example of detecting a movement/gesture to cause the processing unit of a metal detector to perform a ground balancing function;

Figure 4 depicts an example of detecting a collision movement/gesture that prevents an unintentional collision between the metal detector and an external object from generating a false or misleading reading;

Figure 5 depicts another example of detecting a movement gesture to cause the processing unit of a metal detector to save details relating to a detected object of interest.

Figure 6 depicts another example of a movement/gesture to cause the processing unit of a metal detector to reset or calibrate one or more internal signals in relation to the operation of the metal detector; Figures 7a and 7b depicts one embodiment of the signal processing by the metal detector to provide an improved output by accumulating the signals that have one or more the same positional references;

Figure 8 depicts a flow diagram showing an exemplary way of how the detector response signals and positional reference signals can be combined for visualisation to be presented to a user.

Figures 9a, 9b and 9c depict examples of detector response versus displacement data, generated with extra information provided by extra sensor(s);

Figures 10a, 10b and 10c depict examples of detector response versus displacement data with discrimination, generated with extra information provided by extra sensor(s); and

Figure 11 depicts one embodiment of a metal detector with extra sensor(s) capable of performing one or more of the functions described herein.

DETAILED DESCRIPTION OF INVENTION

Figure 1 is a block diagram showing main parts of one embodiment of a metal detector having a sensor.

The metal detector includes a sensing head 101, which includes a magnetic field transmitter and a magnetic field receiver (not shown), to transmit a transmit magnetic field 1 10 and to receive a receive magnetic field 111. The transmitter and the receiver can be separate coils, or can be the same coil, within the sensing head 101.

Receive signal generated by the receive magnetic field 111 received by the sensing head 101, is demodulated by demodulation module 102, filtered by filtering module 103, converted to digital form by Analogue-to-Digital Convertor (ADC) 104, prior to being further processed by signal processing unit 105. One or more of demodulation module 102, filtering module 103 and Analogue- to-Digital Convertor (ADC) 104 may be included in the signal processing unit 105. Sensor(s) 106 is connected signal processing unit 105. Sensor(s) 106 can include one or more motion and or one or more position sensors. Information provided by sensor(s) 106 can be used for enhanced detection methods to improve the operators/users ability to detect, locate, and discriminate buried objects. Sensor(s) 106 are often mounted with sensing head 101 , or being part of sensing head 101. In a case, where the transmitter and the receiver are separate entity, sensor(s) 106 can include one or more separate parts to be mounted on or integrated with both the separated receiver and the transmitter. In one embodiment, sensor(s) 106 is only mounted on or integrated with the receiver itself.

Figure 2 depicts an embodiment where sensor(s) 106 includes an accelerometer 200 as a motion sensor. The output from the accelerometer 200 can be processed to provide acceleration, velocity and position reference signals that are fed to the motion processor 290. The accelerometer 200 provides up to three raw acceleration signals along connection 201 that correspond to the x, y and z axes of the accelerometer 200. The raw signals are fed from connection 201 to the calibration module 210. The raw signals provided by accelerometer 200 may suffer from bias and scale variations in the transfer function between the input force exerted on the accelerometer and the output signals provided by the accelerometer 200. This is due to variations in the manufacturing process of the integrated circuits of the accelerometer 200. Calibration measurements are required to determine the correct calibration coefficients to apply in the calibration module 210. The calibrated signals on connection 21 are consistent between the x, y and z directions and standardised between different accelerometer devices. The signals on connection 211 are fed into the reorientation module 220. The reorientation module 220 applies a rotation so that the signals representing the x, y and z axes are aligned with a orientation or direction convention of the sensing head. For example, with the detector in front of the user on the ground, moving to the right is the positive x direction, moving forward is the positive y direction, and moving the coil vertically off the ground is the positive z direction. The reorientation module 220 ensures that the signals provided to the connection 221 are according to the orientation convention described, regardless of how the accelerometer device is mounted in the sensing head.

The module 240 includes a low pass filter (LPF) for processing the data provided by 221 to produce an indication of the current orientation of the sensing head to connection 241. By selecting a suitably low cut-off frequency for the LPF in 240, the LPF provides an approximate orientation measure. The faster the motion of the metal detector during a detection, the less accurate the orientation measure becomes. The orientation of the sensing head is provided via connection 241 to the motion processor 290. The module 230 also processes the data supplied by connection 221 with a high pass filter (HPF). The HPF removes the dc and low frequency components from the acceleration signals. Over time this tends the signals to zero. This is important because in subsequent integration module 250 any small amount of drift or dc bias will cause the integrated outputs to quickly diverge. By choosing a HPF with a low cut-off frequency allows most of the important medium to low frequency components to be passed to connection 231. The signals on connection 231, which can be considered as centred or zeroed accelerations, are supplied to the integration module 250. The integration module 250 integrates the acceleration to provide the corresponding velocity. Connection 231 also supplies the high-passed acceleration signals to the motion processor 290. The HPF in a preceding module 230 limits the problem of combining sensor drift and integration calculations. For a discrete time system, the integration in 250 can be approximated by a scaled cumulative sum of all the previous samples, the scale being the sampling period, assuming the sampling period is constant. Higher order integration implementations can be used, however, the filtering would mask any improvement in accuracy. The module 250 provides the new velocity signals to connection 251. Although the previous HPF limits the problems of integrating, there will most likely be some dc component introduced in 250. Another HPF 260, fed by connection 251, is used to centre the velocity. The filter design in 260 may be different to that in 230. The centred velocity is placed on connection 261, which is fed into the motion processor 290 and another integration module 270. The integration module 270 is identical in characteristics to integration module 250, and by suppling velocity signals the integration process produces the corresponding position. This position provided to the connection 271 will similarly contain a small dc component. Therefore, another HPF module 280 can be employed to centre the position provided from connection 271. Depending on the tasks within the motion processor 290, the last HPF module 280 may not be necessary, in which case the data on connection 281 will be identical to the data on connection 271. The connection 281 supplies the motion processor 290 with the positional reference signals. The motion processor 290, which is provided with all the spatial related data derived from the accelerometer 200, is able to do coherent motion processing, like determining when certain spatial sequences are performed by the sensor head.

In another embodiment, more sophisticated processing techniques are used to estimate the orientation, velocity and position of the sensor head. For example, this could be a Kalman filter, which uses a unified system model for the estimates rather than considering each axis individually. In addition, different types of motion sensors can be integrated into the system model. The motion sensor could consist of, but is not limited to, an accelerometer and/or a gyrometer, or an acoustic velocity sensor, or a type of measuring system that relies on components external to the metal detector. During typical operation, the operator needs to interact with the detector to change settings and detection modes. It is generally inefficient to input commands via conventional controls such as keypad, menus, switches, rotary encoders, knobs, or touch-screen. This may be due to the user wearing protective clothing which limits the mobility and dexterity of the users hand and prevents easy access to the detector. Also, it may be inconvenient for the user to see or reach the controls of the detector during operation.

With sensor(s) 106 attached to sensing head 101, it is possible to utilise the position and motion of the sensing head to trigger a command function, or to send the metal detector into different operating modes. A sensor can include processing electronics to detect and confirm a range of movement of a metal detector substantially in accordance with a predetermined movement pattern to trigger a function of the metal detector. Alternatively, the processing electronics for such purpose can be included in the signal processing unit of a metal detector. In one embodiment, sensor(s) 106 detects a movement of the magnetic field receiver, and analyses the movement to produce an analysis result, to trigger one or more of the one or more predefined functions. In other words, the sensing head itself can be used as an input device for communicating commands from the user to the detector. This is of great benefit to the user as no button presses and/or looking at the display are required.

By monitoring the motion of the sensing head, a predefined sequence of command movements, or gestures, can be detected by the detector and appropriate actions taken. An audio or visual acknowledgement could be provided once a gesture has been recognized.

In other embodiments, one or more of position, velocity, acceleration of one or more parts of the metal detector, angular position (i.e. tilt), angular velocity (i.e. rotation rate), angular acceleration of the sensor head may be considered for gesture recognition.

As an example, when ground conditions change, it is necessary to recalculate the ground balance settings by performing a ground balance. During ground balancing, the characteristics of the ground near the sensing head 101 are measured and are used in further processing to minimise the adverse effect caused by the ground upon operations of the metal detector, for example, to compensate for the interference to signals generated by electromagnetic energy returned by a buried target caused by the signals generated by the electromagnetic energy returned by the ground.

Typically, to perform a ground balancing requires some interaction of the user via key presses to tell the detector the user is going to perform a ground balance, followed by raising and lowering the sensing head 101 over the ground, allowing the detector to measure the characteristics of the ground. With the sensor(s) 106 monitoring the motion of sensing head 101, the user simply needs to start the ground balance motion of raising and lowering the sensing head 101 ; this will be recognised as a ground balance gesture by the detector and the ground balance processing automatically initiated. When the user stops the raising and lowering of the sensing head 101 and starts detecting using the normal left and right sweeping action, the ground balance procedure will cease and the processing of a receive signal due to receive magnetic field 1 11 will resume using the newly acquired ground balance settings. This process is shown in more detail in Figure 3, and described as follows. The traces 301 , 302 and 303 show the accelerometer output for the x, y and z axis respectively. In the time preceding 320 a sensing head 340 of a metal detector is being moved in a standard detecting motion, that is, swinging left and right, as shown by spatial movement depicted by 330. This leads only a little movement in the z-axis output 303, a little in the y-axis output 302, and the majority of movement in the x-axis output 301. At time 320, the detector starts the ground balancing movement, with sensing head 340 moving up and down as depicted by 331. This continues until time 321. During the ground balance process the detector is interrogating the ground to determine its characteristics. This will then be used by the detector during subsequent detecting. When the detector is being ground balanced the majority of the output is in the z-axis. By setting thresholds 310 and 31 1 for the z-axis output, and thresholds 312 and 313 for the x and y-axis outputs, the ground balance gesture can be recognised. When the thresholds 310 and 311 are exceeded and the thresholds 312 and 313 are not exceeded, the detector is considered to be in ground balancing mode. After the time 321 the detector has resumed moving with the standard detecting motion, as depicted by 332 and thus the ground balancing process stops. The process described above is only one embodiment to recognise the ground balance motion. A number of modifications can be made by using separate thresholds for each of the axes, using an envelope detector like a simple amplitude modulation (AM) filter or a Hilbert filter, estimate the frequency and amplitude and threshold the estimated parameters, or if the ground is not horizontal the low pass filtered output can be used as a reference for orientation of the sensing head 340.

Repeatedly raising and lowering the sensor head is an example of a gesture that augments the interaction between the user and the detector. The recognition of the gesture is used to toggle a particular function of the metal detector. However, a gesture is not necessarily something that is separate from the standard detecting action; there could be a particular action that occurs during the standard detecting action that would ideally be recognised and trigger some type of function.

For example, during operation of a metal detector a collision can occur when a user swings the sensing head 101, and sensing head 101 touches the ground or foreign objects such as small rocks, bushes or trees. Collisions can be common, as during typical operation, the sensing head 101 is scraped along the ground in an effort to get the sensing head as close to the ground as possible to maximise the amount of electromagnetic energy from the buried object received by the sensing head 101. This scraping and knocking of the sensing head introduces sporadic noise into the receive signal from the buried object as received by the receiver within the sensing head 101. As a consequence of this sporadic noise, the receive signal from an object of interest is more difficult to distinguish and hence the object will not be detected. Alternatively, the sporadic noise may have enough amplitude that it can falsely suggest the presence of a target. A collision may also happen during the process of ground balancing the detector, i.e. raising and lowering sensing head 101 over the ground multiple times. It is possible that the user can hit the ground with sensing head 101 which causes sporadic noise in the receive signal to be considered as part of the energy from the ground when determining the ground balance settings.

Collisions may be detected using various means. In one embodiment, sensor(s) 106 which includes a motion sensor is attached to the sensing head 101 to monitor the movement of the sensing head 101. In another embodiment, the metal detector is a single hand-held entity, and the motion sensor 106 may be attached to any part of the metal detector to monitor whether there is a collision. In another embodiment, one or more external components can be used as references or locators to assist the motion sensor.

In one embodiment, signals from a motion sensor of sensor(s) 106 are processed within the sensor(s) 106 to decide whether there is a collision. The characteristics associated with a collision can be determined empirically or theoretically and once known, a comparison or threshold or other measurement techniques can be used to determine whether there has been a collision. The required determination can be done by an associated processor or the signal processing unit 105. The collision decision signal is then made available to the signal processing unit 105. In another embodiment, signal from the motion sensor 106 is processed within the signal processing unit 105, to produce an indicator output signal 1 12 indicative of a target within the influence of the transmit magnetic field. Output signal 112 may be in audio form, video form, or the combination thereof.

To decide whether there is a collision or not, a threshold detector may be used. Such a threshold detector can be embedded within sensor(s) 106 or signal processing unit 105. A collision may be confirmed by a sudden change in velocity or in acceleration. Alternatively, collision may be confirmed when one of more conditions programmed in the threshold detector are fulfilled. For example, one or more of position, velocity, acceleration, jerk, of one or more parts of the metal detector, angular position (i.e. tilt), angular velocity (i.e. rotation rate), angular acceleration of the ^' sensor head may be considered before a collision is confirmed.

Figure 4 shows one embodiment of the processing of a receive signal 413, generated by a receive magnetic field, when a collision is detected. A collision can introduce a spike 415 into the receive signal, and after filtering by filtering module 103 (for example, band pass filtering), the spike 415 may be more pulse like (for example, receive signal 413 takes the form of 417 between time 421 and 423). For comparison purposes, without the introduction of spike 415, receive signal between time 421 and 423 may take the form of 416. The filtered signal, if processed, may produce an output signal (which can be in audio and/or video form), which may falsely suggest to a user of the metal detector that there is a target. With a motion sensor within sensor(s) 106, an acceleration output 418 would be produced and can be used in conjunction with a collision threshold 4 9 to determine the presence of a collision. Thus, the receive signal 417 can be confirmed as being due to a result of a collision. Once a collision is confirmed, the receive signal 417 within a period of time define by, for example, time 421 and 423 as shown, during which the effect of the collision is considered to be capable of producing a false target, can be identified. In an embodiment, receive signal 417 between the time 421 and 423 is abandoned from further processing which produces an output signal to the user.

In another embodiment, the receive signal 417 between the time 421 and 423 is still processed, but no output signal is presented to the user. For example, if the output signal is in audio form, the output audio is modified (for example "muted") between time 421 and 423. This embodiment may be more complex than the embodiment where received signal is not processed, but may have an extra benefit of preventing the discontinuity 429, which may be created when the receive signal 417 between time 421 and 423 is not processed. For example, if receive signal 417 between time 421 and 423 is not processed, the signal level of the receive signal 413 at time 421 is held 420 till time 423. When the processing of the receive signal resumes at time 423, the signal level of receive signal 413 at time 423 may be substantially different from the signal level of the receive signal 413 at 421 as illustrated in Figure 4. Thus, this creates discontinuity 429. Such discontinuity 429 may also create an undesirable output when filtered. For example, if the output signal is in audio form, discontinuity 429 may create unpleasant high-pitch noise, or create an audio signal which may falsely suggest a target.

In other embodiments, time 421 and 423 may be selected to suit the functionality of the metal detector. For example, the time between 421 and 423 may be increased or decreased. Alternatively, the starting time of 421 and/or 423 may be modified, independently and concurrently. Alternatively, the ending time 423 may be dependent on the size of the spike 418.

In another situation the user may wish to save the details about a target that has just been detected and the current environment with respect to the target. In this example, the user could move the sensor head in a circular motion around the target to gesture that this is a target of interest and to save some information. For example, the user may wish to save some target details like the identification and depth, and the environment as it relates to the target, like the characteristics of the ground, the location (if available), and the current detector settings.

The process is shown in Figure 5, and described as follows. The traces 501 and 502 show the accelerometer output for the x and y-axis respectively. In the time preceding 520 the sensing head 540 of a metal detector is being moved in a standard detecting motion, that is, swinging left and right, as shown by spatial movement depicted by 530. At time 520, the sensing head 540 starts the circular motion around the target, as depicted by 531. This continues until time 521. When the sensing head 540 is being moved in a circular path, the accelerometer output 501 and 502 are substantially sine and cosine functions. This means the signals 501 and 502 are in quadrature. This could be recognised by correlating the zero crossings of 501 with the peaks of 502, and conversely, the zero crossings of 502 with the peaks of 501. If the times of the peak values and the zero crossings of 501 and 502 correlate, within a specified tolerance, the detector proceeds and saves the required data. In addition, the user could be asked to confirm such an action. The data accumulated during the detecting session can be downloaded to a computer, or the internet, for subsequent analysis.

In other embodiments, the process to recognise the circular motion could be modified a number of ways, by converting the x and y-axis outputs into polar form, use velocities to improve the signal-to- noise ratio, or delaying one of the x or y-axis outputs by a quarter of a wavelength and cross- correlating them.

For certain types of metal detectors when performing certain types of processing, it may be necessary to re-zero the detector. Furthermore, it is a common requirement to re-zero when the sensing head is away from the ground to get a true zero reference. In this example, the user could simply swing the detecting head in the air which would be recognised as the re-zero gesture and the appropriate action taken in the processing.

Figure 6 shows the process in more detail. The traces 601 , 02 and 603 show the accelerometer output for the x, y and z axis respectively. Similarly to Figure 3 and Figure 5, the time preceding 620 the sensor head 640 is moving with the standard detecting action. At time 620 the sensor head 640 is lifted over the user, as depicted in 631. This continues until time 621. By setting threshold 610 for the z-axis output, and thresholds 11 and 12 for the x and y-axis outputs, the re-zero gesture can be recognised. For example, when the threshold 610 is exceeded and the thresholds 61 1 and 612 are not exceeded, the re-zero gesture can be recognised and the detector issues a re-zero command to the processing. The processing can then do any required calibrations, like reset the baseline of internal signals for example. In another embodiment the thresholds and logic are modified so the gesture is recognised when the detector head is only raised to the horizontal. It is typical for metal detectors to have a detection depth greater than the identification depth. This means there is a range of depths, near the limit of detection, where it is difficult to get an accurate identification of the target. The reason is that the received signal from the target has a low signal-to- noise ratio (SNR). A common method to increase the SNR is to take the average of multiple identical signals, and this assumes the noise on receive signal is independent between each target response. However, with a typical metal detector there is no way to correlate the receive signal to a target. For example, if the detector receives two target responses, it could be from two different targets in close proximity, or it could be from passing over the same target in each direction. Using only the received data it is impossible to determine the difference, thus it cannot be determine what signals to sum.

With the use of the motion sensor within sensor(s) 106 attached to sensing head 101 (those of Figure 1), it is possible to use the receive signal from multiple passes over the same target to provide a better detection results. For example, when a user makes a single pass over an object that is buried in the ground, and if the object is small and/or deep, the receive signal will be small. The receive signal may be large enough to be detected by the metal detector, but not large enough to extract any useful information to produce an output identifying or discriminating the object. With the use of multiple receive signals which are correlated with position, the SNR is improved by combining several receive signals. This is possible when the receive signals are correlated with position.

In one embodiment, one or more sensors associate receive signals with one or more positional references. Figure 7a shows the association with respect to time between the receive signal 700 and the positional reference 710. As the detector head passes over the target, the response in the received signal is indicated by the peaks 701, 702, 703, 704 and 705. The detection threshold 720 determines which responses will trigger an output to be indicated to the user. The response peaks shown in Figure 7a just exceed the detection threshold, except for 704 which would be undetected. There is very little information above the threshold that can be used for identification and discrimination. While the detector is passed back and fourth over the target, the motion sensor determines the positional reference 710. In Figure 7a, due to the filtering performed on the accelerometer output in calculating the positional reference 710, the target appears at position zero. Negative positions indicate the detector head is to the left of the target, and conversely, positive positions indicate the sensor head is to the right of the target.

Figure 7b shows the receive signal 700 plotted against the positional reference 710. As in Figure 7a, all of the signal peaks except 704 are just above the detection threshold 720. Now that the receive signal is correlated with position, the multiple passes can be added to produce the cumulated receive signal 750. Similarly, the mean signal could also be calculated, which would give an identical improvement in SNR. When calculating the sum or mean of the response, it may need to be interpolated or extrapolated so that it is aligned to a common position scale. In a real-time system the accumulated signal can be updated when each new sample arrives, interpolating or extrapolating as necessary, or at the end of each swing, which would be denoted by a zero crossing in the velocity output. The accumulated signal can then be processed in the same manner as the original signal.

In this embodiment, the receive signal 700 only consist of a single channel and the positional reference only considers the x-axis movements. This means that movement away from the target in the y or z direction will be ignored. For example, the target response indicated by 704 could be due to the detector head sweeping higher over the target.

In other embodiments, the receive signal can consist of a plurality of channels. In addition, the positional reference can be extended to consider movement in the y and z directions.

With the addition of a 2-dimensional visual display, correlations between the physical position of the sensor head and the processed signals can be presented to the user in a visual form that is easily understood. One axis of the display can be used to represent the position of the sensor head along the x axis while the other axis of the display can be used to represent the position of the sensor head along the y axis. The intensity of emission of the display at any particular point within the area defined by the x and y axes can be caused to indicate the strength of the signal detected by the sensor head at its corresponding position. Further, or alternatively, colour of the emission can be used to indicate strength of the received signal. Further, or alternatively, there could be a combination of colour and intensity used to display not only strength of the received signal as correlated with the position of the sensor head, but colour could simultaneously be used to indicate some aspect of the nature of the received signal as determined by the signal processing employed in the detector. Such an aspect could be the determined conductivity of the detected target, or targets. Alternatively, the aspect represented could be the degree to which the target has been determined to present a ferrous nature, or the representation could be of a function of both the conductive and ferrous natures of the target. Figure 8 shows a flow diagram of the collection of data from the sensor(s) 106 and the sensing head 101 and the high-level sequence of events that produce the representation of the signal correlated with the x-position of the sensing head 101. Pertinent data or position data 801 of the x-position of the sensing head are fed to a distance high-pass filter for high-pass filtering distance 803 with a cutoff frequency at the order of 0.1 Hz or 1 Hz. The output from the distance high-pass filter is used to determine whether a sweep has ended and a new one begun 805, through monitoring the velocity of the sensor head. If it is determined that a new sweep has begun and if the "waterfall" mode of display is selected 807, the currently shown tracks are moved up, or down, one position 809. Once that is done, display of the representation of the new sweep can commence. Without the "waterfall" display, the representation of the new sweep can be made to overwrite the track of the previous sweep as the new sweep progresses.

Before a pixel is made to represent some of the signal data 81 1, it is to be decided 813 whether the position of the sensor head can be depicted as being within the bounds of the display and, if so, the position within that display is calculated 815. Once these decisions and calculations are done, the signal data 81 1 from the signal processing of the receive signal are used to determine the colour and intensity of the screen at the x axis position 815.

That being done, another decision 817 about the positions of the pixels in the axes of the physical screen is to be made. This decision and calculations of any required interpolations 819 are performed and the selected pixels are set to their designated colour and intensity 821. The whole process is repeated for each proceeding data processing cycle.

In one embodiment of this invention, a simpler form of the visual representation of signal vs position could be employed where the position of the sensor head only along the x axis is utilised. Figure 9a is an example of the display 901 that would be shown as the sensing head 101 passes over a target. In this figure, the density of lines indicates what would be either the intensity or colour of emission of the display along its axis chosen to represent the position of the sensor head along the x axis. The representation along the y axis, in this example, has finite extent only for the purpose of making the representation easily visible to the user and has no meaning related to the position of the sensor head in the y direction, or any other direction. The intensity or colour of the display could be linearly related to the strength of the signal, but this would generally require too much dynamic range of the intensity to be of use over a large range of signal strength. Alternatively, the indication could be related to another function of signal strength, such as the logarithm of signal strength. Other means of increasing the range of signal strengths that could be meaningfully represented is to have an automatic gain control in the transformation of signal strength to luminous indication, or to have a changing baseline of signal strength that is required to produce a non-zero indication. If one of these, or some other, means of transformation is not employed, strong signals could produce a representation over the breadth of the display in which the visual indication does not vary enough to be easily interpreted by the user.

As described in the introduction to this use of the extra sensor, the colour of the emission of the display could be related to some property of the target as determined by the processing of the received signal. A problem with correlating and displaying the signal with the x-position of the sensor head is to determine, for instance, what x-positions of the sensor head are represented at the extrema of the representation of the x displacement on the visual display device. The easiest solution is to have a fixed ratio of say, one display width being equivalent to 1 metre of x-distance. The value of the equivalent distance might be selectable by the user, to best suit the sweep distance of a user, but would remain fixed throughout detection.

Another solution uses a determination of when, and where, the x-movement of the sensor head changes sign and the distance between these two points for each pair of points. This could be determined through monitoring the direction of movement and starting a new display when the velocity changes sign. The x-distance represented by the x axis of the display could be calculated from the average of the distances of the previous five sweeps, or some other number of sweeps. When a significant change in the relation between average breadth of sweep and the width of the representation occurs, any representations on the screen might need to be redrawn with the new scales.

The situation can arise that, when the user changes direction of forward movement, the centre of the sweeps of the sensor head is along a different direction than the previous sweeps. This could produce a problem with the representation, as the new data to be displayed are from a physical area that would be shown off the display screen. Placing a high-pass filter in the data stream for the x- position will allow the alignment of the central x-position of the display with the mean x-position of the sensor head. Figure 2 shows that, with the accelerometer 200 included in sensor(s) 106, the data regarding the x-position of the sensor head are passed through a high-pass filter before being used to produce the visual representation. In the absence of a high-pass filter in the position data stream, one with a relatively low cut-off frequency, say the order of 0.1 Hz, could be included to automatically re-align the mean x-position of the sensor head and the central position on the x axis of the display.

For a system in which an accelerometer is used for more than determining the relative position of the sensing head, the cut-off frequency of the high-pass filter used in the general derivation of velocity and position of the sensor head might not be suitable for the representation of signal strength vs x- position of the sensor head, to wit, if the user suddenly progresses in a direction markedly different from that followed previously, the time taken for the display to correctly correlate the centre of the swing of the sensor head with the centre of the x-displacement as shown on the display might not be suitable if the general high-pass filter is used. If so, another high-pass filter can be inserted in the calculation of the x-displacement to speed the correction.

Figure 9b shows another way of displaying success sweeps of the sensor head. Each successive sweep is represented at the bottom of a series of representations of successive sweep. As a new sweep is processed, the previous sweeps are moved up along the y axis of the display. A fixed number of successive sweeps can be caused to be represented at once. For instance, in Figure 9b, the strip 91 1 is the representation of the current pass of the sensor head. The strip 912 is of the immediately previous pass, the strip 913 is of the pass immediately before that and strip 914 is of the pass immediately before that of the strip 913.

Figure 9c shows a representation 921 of the signal detected and correlated with 2-dimensional movement of the sensor head, in both the sweep and longitudinal movements of the sensor head.

Figures 10a, 10b and 10c show more informative embodiments of the representation of signals as they are correlated with position of the sensor head. In these embodiments, the strengths of the received signals are represented by the intensity of the representation and the nature of the detected object, as determined by the signal processing of the signals, can be represented by the colour of the representation. The Figure 10a is a representation 1001 in only the x axis, similar to the method shown by Figure 9a, except that both the strength of received signal and the nature of the target are represented simultaneously, the strength of signal being represented by the intensity of the representation while the nature of the signal is represented by the colour of the representation (in this case dotted lines indicate a different colour from solid lines). Figure 10b shows the simultaneous representation 101 1 , 1012, 1013, 1014 ofthe strength of the received signal, by intensity of the representation, and the nature of the detection, by colour of the representation (in this case dotted lines indicate a different colour from solid lines), in a manner similar to that shown in Figure 9b.

Figure 10c shows the same simultaneous representation 1021 in two-dimensional position of the sensor head, in a way similar, in the correlation with position, as shown in Figure 9c.

An advantage of these embodiments of representation of the received signal correlated with position of the sensor head is that they can obviate the necessity of having a threshold of signal strength, or a minimum of signal strength, required before the detector indicates to the user that a desirable target has been detected. The manner in which many detectors with discrimination and target identification is that, in order to reduce the rate of falsely indicating the presence of a target, signals have to exceed a preset amplitude. Only when the minimum level, which can be set by the user, is exceeded by the received signal will the detector indicate the presence of a target. With this invention, with its system of continuous representation of the return signal, the user can interpret the representation of the signals and decide, through the intensity of the representation or the combination of the intensity and the colour of the representation, whether the represented signal is likely to be indicating a detected target, or a desirable target. By passing repeatedly over a particular sweep, the user can determine the whether there is a correlation of a likely signal in the one position.

Further to the advantage of not requiring that signals exceed a preset threshold is the improved ability to detect multiple targets that are close to each other. Again, many target discriminating and identifying detectors are designed to indicate the presence of only one of a collection of targets. Which of the targets is indicated can be set to be that target that produces the strongest received signal, or it can be set to be that target that is most desirable among them, perhaps determined by being that which produces that return signal with attributes that indicate that it is the most conductive. This choice is often necessitated by the fact that, were multiple targets to all be indicated with their inferred identities by the detector, the information for one target would be displayed for only a very short time before it became necessary to display the identity of the next, nearby target. Such situations are generally confusing for the user.

With the invention as described and illustrated in the Figures 9a to 10c, the presence and identities of a plurality of targets can be represented simultaneously; the user is presented with more information, of what is under the ground, in a manner that facilitates interpretation of the extra information. Further to this, numerical indications of the natures of multiple targets could be shown, each over or near its respective signal maximum as represented on the display.

Figure 1 1 shows an embodiment of a metal detector which can perform one or more of the functions described in this specification with extra information provided by one or more sensors. Sensing head 1 101, ADC 1 102, signal processing unit 1 103, motion and/or position sensor(s) 1 104 are similar to sensing head 101, ADC 104, signal processing unit 105, motion and/or position sensor(s) 106 described with reference to Figure 1 respectively. The information provided by sensor(s) 1 104 is fed to control signal generation module 1 105 and gesture recognition module 1 106. Based on the input from control signal generation module 1 105 and gesture recognition module 1 106, the signal processing unit 1 103 performs intended function(s) and produces an indicator output signal indicative of a target in audio form 1 1 10 and/or in video form 1 1 11. The video form 1 1 1 1 may be a screen showing images in the form depicted in Figures 9a to Figure 10c. The images can also be formed based on an output 1 1 13 from a displacement versus response generation module 1 107 which requires input from the sensor(s) 1 104 and signal processing unit 1 103.

A detailed description of one or more preferred embodiments of the invention is provided above along with accompanying figures that illustrate by way of example the principles of the invention. While the invention is described in connection with such embodiments, it should be understood that the invention is not limited to any embodiment. On the contrary, the scope of the invention is limited only by the appended claims and the invention encompasses numerous alternatives, modifications, and equivalents. For the purpose of example, numerous specific details are set forth in the description above in order to provide a thorough understanding of the present invention. The present invention may be practised according to the claims without some or all of these specific details. For the purpose of clarity, technical material that is known in the technical fields related to the invention has not been described in detail so that the present invention is not unnecessarily obscured.

Throughout this specification and the claims that follow unless the context requires otherwise, the words 'comprise' and 'include' and variations such as 'comprising' and 'including' will be understood to imply the inclusion of a stated integer or group of integers but not the exclusion of any other integer or group of integers.

The reference to any prior art in this specification is not, and should not be taken as, an

acknowledgment or any form of suggestion that such prior art forms part of the common general knowledge of the technical field.