JP7124692B2 - Foot sole shape estimation method and pressure sensor device - Google Patents

Description

The present invention relates to a sole shape estimating method and a pressure sensor device, and more particularly to a sole shape estimating method for estimating the shape of the sole of a person walking on the sensor panel using a plurality of detection cells for detecting pressure. , and pressure sensor devices.

2. Description of the Related Art In recent years, pressure sensor devices in which a plurality of detection cells for detecting pressure are arranged in a matrix have been used in various fields. For example, by installing the pressure-sensitive sensor device on the floor, it is possible to specify the shape of the foot of a person who passes over the pressure-sensitive sensor device, and apply it to personal authentication and the like.

Patent Literature 1 discloses a personal authentication system that can measure the distribution of the load on the soles of people's feet and use the distribution of the load to perform personal authentication. In the personal authentication system according to Patent Document 1, by using a pressure-sensitive sensor device in which piezoelectric elements are densely arranged, the shape and load distribution of a person's sole can be obtained with high accuracy.

JP 2008-250996 A

When measuring the shape of a person's sole and applying it to personal authentication, etc., it is necessary to obtain the shape with high resolution. In general, in order to obtain the shape of the sole with high resolution, as described in Patent Document 1, a method of arranging detection cells for detecting pressure in a pressure-sensitive sensor device at a high density is conceivable. . However, when the detection cells are densely arranged in the pressure-sensitive sensor device, there is a problem that the manufacturing cost and the control load of the pressure-sensitive sensor device increase.

SUMMARY OF THE INVENTION The present invention has been made to solve such a problem. An object of the present invention is to provide a shape estimation method and a pressure sensor device.

A foot sole shape estimation method according to an aspect of the present invention uses a sensor panel in which a plurality of detection cells for detecting pressure are arranged in a matrix in row and column directions. A sole shape estimation method for estimating the shape of a sole, comprising: a first step of acquiring temporal walking data of a person walking on the sensor panel using the plurality of detection cells; a second step of respectively extracting detection regions corresponding to the shape of the sole of the person using the obtained walking data; a third step of obtaining a center position of each of the plurality of extracted detection regions; a fourth step of obtaining a grounding direction of the person's sole in each of the extracted detection regions using temporal information of the detection cells that detected the pressure; and a fifth step of estimating the shape of the person's sole by superimposing them so that the center position and the ground contact direction are the same.

A pressure-sensitive sensor device according to an aspect of the present invention comprises a sensor panel in which a plurality of detection cells for detecting pressure are arranged in a matrix in row and column directions; and a processing circuit for estimating the shape of the sole of a person who has walked on the sensor panel, wherein the processing circuit stores the temporal walking data of the person who has walked on the sensor panel in the plurality of detection cells. using the acquired walking data, respectively extracting detection areas corresponding to the shape of the sole of the person's foot, acquiring the center position of each of the plurality of extracted detection areas, and obtaining the pressure Using the temporal information of the detection cells that detected the contact direction of the sole of the person in each of the extracted detection regions, The shape of the person's sole is estimated by superimposing them so that they match each other.

ADVANTAGE OF THE INVENTION According to the present invention, it is possible to provide a sole shape estimating method and a pressure sensor device capable of estimating the shape of a person's sole with high accuracy while suppressing the arrangement density of detection cells that detect pressure. .

It is a figure for demonstrating the pressure sensor apparatus concerning embodiment. It is a cross-sectional view showing a detection cell provided in the pressure-sensitive sensor device according to the embodiment. 3 is a cross-sectional view showing a state in which pressure is applied to the detection cell shown in FIG. 2; FIG. It is a figure for demonstrating the sole shape estimation method concerning embodiment. It is a figure for demonstrating the sole shape estimation method concerning embodiment. 4 is a flowchart for explaining a sole shape estimation method according to an embodiment; It is a figure for demonstrating the sole shape estimation method concerning embodiment. It is a figure for demonstrating the sole shape estimation method concerning embodiment. It is a flow chart for explaining a detailed flow of step S50. It is a figure for demonstrating the sole shape estimation method concerning embodiment. It is a figure for demonstrating the sole shape estimation method concerning embodiment. It is a figure for demonstrating the sole shape estimation method concerning embodiment. It is a figure for demonstrating the preliminary processing in the sole shape estimation method concerning embodiment. It is a figure for demonstrating the sole shape estimation method concerning a modification. It is a figure for demonstrating the sole shape estimation method concerning a modification. It is a figure for demonstrating the sole shape estimation method concerning a modification.

Specific embodiments are described in detail below with reference to the drawings. In each drawing, the same reference numerals are given to the same or corresponding elements, and redundant description will be omitted as necessary for clarity of description.

<Configuration of pressure sensor device>
FIG. 1 is a block diagram for explaining the pressure sensor device according to this embodiment. As shown in FIG. 1, the pressure sensor device 1 according to this embodiment includes a sensor panel 5, a processing circuit 8, and a recording device 9. As shown in FIG. The sensor panel 5 has a plurality of detection cells 10 for detecting pressure arranged in a matrix in row and column directions. FIG. 1 shows a configuration example including detection cells 10 of 10 rows×18 columns, but in the pressure-sensitive sensor device 1 according to the present embodiment, the number of detection cells 10 provided in the row direction and the column direction is arbitrary. can be determined to

The processing circuit 8 is a circuit that estimates the shape of a person's sole based on the output of the sensor panel 5 (that is, the pressure detected by the sensor panel 5). For example, the processing circuit 8 includes a processor, a memory, and the like, and the processor executes a program to implement the sole shape estimation method according to the present embodiment. The memory stores a program and predetermined data for carrying out the sole shape estimation method.

The recording device 9 is a device for recording information output by the processing circuit 8 . For example, the recording device 9 includes a nonvolatile recording medium such as an SSD (Solid State Drive). Note that the recording device 9 may be detachable, or may be a device provided outside the pressure sensor device 1 .

FIG. 2 is a cross-sectional view showing the detection cell 10 included in the pressure-sensitive sensor device according to this embodiment. As shown in FIG. 2, sensing cell 10 is formed on substrate 11 . The substrate 11 can be configured using a rigid substrate such as a printed wiring board. Lower electrodes 12 and 13 are arranged on the upper surface of the substrate 11 so as to be adjacent to each other. As materials for the lower electrodes 12 and 13, for example, metal materials such as copper and aluminum, and carbon-based materials such as carbon black and graphite can be used. The bottom electrodes 12, 13 may be formed on the substrate 11 using, for example, a printing process. Alternatively, for example, a film on which the lower electrodes 12 and 13 are printed may be attached to the substrate 11 to form it. A driving voltage is supplied to the lower electrode 12 .

A film 15 is arranged on the substrate 11 . A spacer 17 is provided between the substrate 11 and the film 15 . The spacers 17 are arranged on both sides of the detection cell 10 in the row direction (in other words, on both sides of the lower electrodes 12 and 13). By providing the spacer 17, the substrate 11 and the film 15 can be spaced apart from each other. An upper electrode 14 is formed on the bottom surface of the film 15 . The upper electrode 14 is arranged so as to face the lower electrodes 12 and 13 .

FIG. 3 is a cross-sectional view showing a state in which pressure is applied to the detection cell 10 shown in FIG. As shown in FIG. 3, in the pressure-sensitive sensor device 1 according to the present embodiment, when stress F1 is applied to the upper surface of each detection cell 10, the upper electrode 14 contacts the lower electrodes 12 and 13. is configured to detect pressure at That is, a drive voltage is supplied to the lower electrode 12 of each detection cell 10, and the film 15 is displaced in a direction approaching the substrate 11, and the upper electrode 14 contacts the lower electrodes 12 and 13, thereby 13 becomes conductive, and by detecting that a driving voltage is applied to the lower electrode 13, the pressure is detected.

The configuration of each detection cell 10 shown in FIGS. 2 and 3 is an example, and in the present embodiment, the detection cell 10 may be configured using, for example, a piezoelectric element or a pressure-sensitive element. In other words, in the present embodiment, it is sufficient that a plurality of detection cells 10 for detecting pressure are arranged in a matrix, and the method by which each detection cell detects pressure is not particularly limited.

Next, a method for estimating the shape of a person's sole using the sensor panel 5 will be described.

<Algorithm for Determining Human Feet>
FIG. 4 is a diagram for explaining the sole shape estimation method according to the present embodiment, and is a diagram for explaining the processing (algorithm) for determining whether or not the detection area corresponds to a human foot. be. As shown in the upper diagram of FIG. 4, when the sensor panel 5 is stepped on by a person's foot, the detection cell 10 detects pressure. In the drawings of the present application, the detection cells 10 that detect pressure are indicated by hatching. For example, when the sensor panel 5 is stepped on by a person's foot and the detection area 21 (hatched area; the same applies hereinafter) is detected, the processing circuit 8 determines whether the detection area 21 corresponds to the person's foot. judge. Specifically, the processing circuit 8 compares the detection area 21 with a reference value for determining whether or not it corresponds to a human foot, and if the detection area 21 is within the range of the reference value, the detection area 21 corresponds to a human foot. In the lower diagram of FIG. 4, an area 24 indicates an area in which the detection area 21 is determined to correspond to a human foot.

Here, the reference value for determining whether or not it is a human foot is, for example, the size of a human foot (the length of a person's foot, including the length of a shoe). If the length (longitudinal length) of the longest part of 21 is within the range of the reference value for the length of the human leg, it is determined that the detection region 21 corresponds to the human leg. Specifically, the reference value of the leg length has a minimum value and a maximum value, and the length of the detection area 21 is equal to or greater than the minimum reference value of the leg length and equal to or less than the maximum reference value of the leg length. , it is determined that the detection area 21 corresponds to a human foot. For example, the reference value of the leg length may be determined in consideration of the adult's leg length, or may be determined in consideration of the child's leg length in addition to the adult's leg length.

Further, in the present embodiment, the width of a person's foot (including the width of a shoe) may be used as a reference value for determining whether it is a human foot. By using the width of a person's foot in addition to the length of the person's foot, it is possible to more accurately determine whether or not the object is a person's foot.

Further, in the present embodiment, the area of a human foot (including the area of a shoe) may be used as a reference value for determining whether or not it is a human foot. By using the area of a person's foot in addition to the length and width of the person's foot, it is possible to more accurately determine whether or not it is a person's foot. In addition, in the present embodiment, it may be determined whether or not the object is a human foot by using the length of the human foot and the area of the human foot.

The reference value for determining whether or not the object is a human foot is stored in advance in a memory included in the processing circuit 8, for example.

Although the detection area 21 shown in FIG. 4 is a continuous area, the detection area 22 may be divided into a plurality of areas, such as the detection area 22 shown in FIG. For example, when a user steps on the sensor panel 5 while wearing high heels, the detection area 22 is divided between the heel portion and the toe portion. In such a case, as shown in the lower diagram of FIG. 4, the distance between the center point 26 of one of the detection areas 22 and the center point 27 of the other area is obtained, and this distance is a predetermined reference value. Whether or not it corresponds to a human foot can be determined based on whether or not it is within the range. Here, of the detection areas 22, the area with the smaller area (area including the center point 26) corresponds to the heel portion, and the area with the larger area (area including the center point 27) corresponds to the toe area. there is That is, when the detection area 22 is divided into a plurality of areas like the detection area 22 in FIG. It is possible to determine whether or not In the lower diagram of FIG. 4, a region 28 indicates a region determined to correspond to a human foot. In the case of the detection area 22, the reference value for determining whether or not it is a human foot is the distance between the center point of the heel portion and the center point of the toe portion. Such reference values are stored in advance in a memory provided in the processing circuit 8 .

<Algorithm for determining foot contact direction>
Next, an algorithm for determining the contact direction of a person's foot will be described with reference to FIG. In general, when a person walks, the heel strikes the ground, then the whole foot touches the ground, and then the heel rises and the toes leave the ground. In the present embodiment, the direction of contact of the foot is determined by tracking the series of motions of the foot of the person. The algorithm for determining the ground contact direction of the foot will be specifically described below.

As shown in FIG. 5(a), when the heel of the foot lands on the sensor panel 5, the detection cell 10 detects pressure. The center point of the detection cell 10 detecting the pressure at this time is the center point 31 (corresponding to the position of the heel of the foot). After that, as shown in FIG. 5(b), when the foot further touches the sensor panel 5, the area where the detection cell 10 detects the pressure becomes wider. The center point of the detection cell 10 detecting the pressure at this time is the center point 32 . After that, as shown in FIG. 5(c), when the entire foot touches the sensor panel 5, the area in which the detection cell 10 detects pressure becomes wider. The center point of the detection cell 10 detecting the pressure at this time is the center point 32 . After that, as shown in FIG. 5(d), when the heel of the foot is lifted, only the toe touches the sensor panel 5, so the detection cell 10 detects the pressure of the toe. The center point of the detection cell 10 detecting the pressure at this time is the center point 32 (corresponding to the position of the toe).

In this embodiment, the foot contact direction is determined based on the temporal movement direction of the center point 32 of the detection cell 10 that detects the pressure in each of FIGS. 5(a) to 5(d). 5(b) to 5(d) with respect to the center point 31 (corresponding to the position of the heel of the foot) in FIG. 5(a). , which determines the contact direction of the person's foot.

<Algorithm for estimating the shape of the sole of a person>
Next, an algorithm for estimating the shape of a person's sole will be described. FIG. 6 is a flowchart for explaining an algorithm for estimating the shape of a person's sole. 7 and 8 are diagrams for explaining the flowchart of FIG. FIG. 7 shows an example of walking data 100 of a person walking on the sensor panel 5 from left to right. FIG. 8 shows the result of analyzing the walking data 100 according to FIG.

As shown in FIG. 6, the step of acquiring and recording information about the detection area mainly includes steps S10 to S50.

First, in step S<b>10 , the processing circuit 8 acquires temporal walking data 100 of a person walking on the sensor panel 5 using a plurality of detection cells 10 . For example, the processing circuit 8 first acquires the coordinate position and detection time of the detection cell 10 that detected the pressure. Then, the processing circuit 8 records the acquired combination of the coordinate position and the detection time as the walking data 100 in the recording device 9 .

As a specific example, a case where there is a trace of walking as shown in FIG. 7A will be described below. FIG. 7( a ) shows the sensor panel 5 , the footprints of a person who walked across the sensor panel 5 from left to right, and other people's footprints different from the person. Here, it is assumed that several detection cells 10 detect pressure as shown in FIG. 7(b) along with the history of walking as shown in FIG. 7(a). That is, it is assumed that pressure is detected in detection areas 40 and 50 corresponding to the pedestrian's left foot, detection areas 45 and 55 corresponding to the pedestrian's right foot, and detection area 110 corresponding to another pedestrian's foot. . Assume that the pressure is erroneously detected in the detection area 120 regardless of walking. At this time, the processing circuit 8 records the combination of the coordinate position and detection time of each detection cell 10 in the detection areas 40 , 45 , 50 , 55 , 110 , 120 in the recording device 9 as walking data 100 .

After step S10, the process proceeds to step S20, and the processing circuit 8 uses the walking data 100 acquired in step S10 to extract detection regions corresponding to the shape of the sole of the person (see FIG. 6). For example, the processing circuit 8 extracts detection regions corresponding to the shape of the sole of a person's foot, based on the above-described algorithm (algorithm for determining a person's foot).

FIG. 8A is a diagram showing an example of extraction of a detection region corresponding to the shape of a person's sole from walking data 100. FIG. In the example of FIG. 8A, detection areas 40, 45, 50, 55, and 110 are extracted as corresponding to human feet 43, 48, 53, 58, and 113, respectively. On the other hand, the detection area 120 does not correspond to the shape of the human foot because its size is less than the size of the human foot. Therefore, detection region 120 is not extracted in step S20.

Next, proceeding to step S30, the processing circuit 8 obtains the center position of each of the plurality of detection areas extracted in step S20 (see FIG. 6). In the example of this embodiment, the processing circuit 8 determines the center point of the detection cells 10 forming each detection area as the center position of the detection area. The processing circuit 8 may determine the centroid of the plurality of detection cells 10 forming each detection area as the central position, or may determine a position determined based on an arbitrary formula specified by the user as the central position. You may

Next, proceeding to step S40, the processing circuit 8 acquires the contact direction of the sole of the person in each of the detection regions extracted in step S20 using the temporal information of the detection cells 10 that have detected pressure ( See Figure 6). For example, processing circuitry 8 determines the orientation of the sole of the person's foot in each of the detection regions based on the algorithm described above (algorithm for determining the orientation of foot contact).
In this embodiment, an example in which step S40 is performed after step S30 has been described, but step S30 may be performed after step S40.

FIG. 8(b) shows an example in which the respective center positions and grounding directions of the detection areas 40, 45, 50, 55, and 110 are determined. For example, the processing circuitry 8 determines a foot contact direction 44 based on the direction of movement of the center points 41, 42 over time relative to the detection area 40. FIG. Here, in FIG. 8B, the center point 41 corresponds to the position of the heel, and the center point 42 corresponds to the center position of the foot. The processing circuit 8 can determine the grounding direction 44 of the foot by obtaining the movement direction over time from the position of the heel of the foot (center point 41) to the position of the center of the foot (center point 42). Further, the processing circuit 8 sets the position of the center of the foot (the center point 42 ) as the center position of the detection area 40 .
Note that FIG. 8B shows a case in which the heel position (center point 41) and the center position (center point 42) of the foot are used to determine the grounding direction 44 of the foot. 5, the center point corresponding to the toe of the foot may be used to determine the ground contact direction 44 .

Similarly, the processing circuit 8 determines a foot contact direction 49 based on the movement direction of the center points 46 and 47 over time for the detection area 45, and determines the position of the center of the foot (center point 47) as follows: Let it be the center position of the detection area 45 . In addition, the processing circuit 8 determines a grounding direction 54 of the foot based on the temporal movement direction of the center points 51 and 52 in the detection area 50, and detects the position of the center of the foot (center point 52). Let it be the center position of the area 50 . In addition, the processing circuit 8 determines a grounding direction 59 of the foot based on the movement direction of the center points 56 and 57 over time with respect to the detection area 55, and the processing circuit 8 determines the position of the center of the foot (center point 57) is determined as the center position of the detection area 55. FIG. In addition, the processing circuit 8 determines a grounding direction 114 of the foot based on the temporal movement direction of the center points 111 and 112 in the detection area 110, and detects the position of the center of the foot (center point 112). Let it be the center position of the area 110 .

After step S40, the process proceeds to step S50, and the processing circuit 8 parallelly and rotationally moves and superimposes the plurality of detection areas so that the center positions and grounding directions are aligned with each other, so that the soles of the person's feet are superimposed on each other. is estimated, and the flow ends (see FIG. 6). By superimposing the detection areas so that the center positions and grounding directions match, a shape obtained by averaging the respective detection areas is obtained. It can be said that the shape obtained in this manner reproduces the shape of the sole better than the shape of each detection area. Therefore, the sole shape can be estimated with high accuracy.

Here, an example of a detailed process of step S50 will be described. FIG. 9 is a detailed flowchart of step S50. As shown in FIG. 9, step S50 mainly includes steps S51 to S53. FIG. 10 is a diagram for explaining step S52.

First, in step S51, the processing circuit 8 extracts the detection area corresponding to the foot on the same side from among the plurality of detection areas extracted in step S20 (see FIG. 9).

For example, the processing circuit 8 extracts a set of detection areas that have a small difference in detection time, a short distance, and a small angle between grounding directions, thereby extracting detection areas corresponding to the feet of the same person. can be done. In this case, the processing circuit 8 first determines that the difference in the detection times at which the detection cells 10 detect the pressure in the plurality of detection regions extracted in step S20 is within a predetermined time range (for example, 0 to 3 seconds). Extract the set of detection regions in . Then, the processing circuit 8 extracts a set of detection areas in which the distance between the respective detection areas is within a range of a predetermined distance (for example, human stride length) from the extracted sets of detection areas. Furthermore, after that, the processing circuit 8 extracts a set of detection areas in which the angles formed by the grounding directions of the respective detection areas are within a predetermined angle range (eg, 0 degrees to 90 degrees). As described above, the processing circuit 8 can extract detection regions corresponding to the same person's feet.

For example, in FIG. 8B, it is assumed that detection regions 40, 45, 50, and 55 are detected within a predetermined time range. It is also assumed that the distance between the detection areas is within a predetermined distance range. Furthermore, the angles formed by the grounding directions 44, 49, 54, and 59 of the detection area are each within a predetermined angle range. In this case, it can be determined that the detection areas 40, 45, 50, 55 correspond to the feet of the same person. On the other hand, the grounding direction 114 of the detection area 110 has a predetermined angular range with respect to the grounding directions (grounding directions 44, 49, 54, 59) of the other detection areas (detection areas 40, 45, 50, 55). Over. In this case, it can be determined that the detection area 110 corresponds to another person's foot.

In addition, the processing circuit 8 synthesizes the grounding directions 44, 49, 54, and 59 of the detection areas 40, 45, 50, and 55 determined to correspond to the feet of the same person, thereby determining whether the person walked from left to right. can be detected. Since the detection areas 40 and 50 are on the left side when viewed from the person's walking direction, the processing circuit 8 can determine that the detection areas 40 and 50 correspond to the person's left foot. Similarly, since the detection areas 45 and 55 are on the right side when viewed from the person's walking direction, the processing circuit 8 can determine that the detection areas 45 and 55 correspond to the person's right foot. As described above, the processing circuit 8 can extract that the detection areas 40 and 50 correspond to the same person's left foot, and the detection areas 45 and 55 correspond to the same person's right foot, as shown in FIG. 8(c).

After step S51, the process proceeds to step S52, in which the detection areas corresponding to the same foot of the same person are moved parallel and rotationally and superimposed so that the center position and the contact direction match (see FIG. 9).

For example, as shown in FIG. 10(a), the processing circuit 8 translates and rotates the detection areas 40 and 50 corresponding to the left foot of the same person so that the respective center positions and ground contact directions match. and superimpose them. Specifically, first, the detection area 50 is rotated so that the grounding direction 44 and the grounding direction 54 are the same. After that, the detection regions 50 are moved in parallel so that the center points 42 and 52, which are the respective center positions, coincide with each other, and the respective detection regions are superimposed. In this manner, an integration area 61 obtained by superimposing the detection area 40 and the detection area 50 can be obtained. In FIG. 10(a), a portion of the integration area 61 where the detection area 40 and the detection area 50 overlap is indicated by dark hatching. Note that FIG. 10(a) shows an example in which the grounding directions 44 and 54 are substantially the same to simplify the explanation. An example of superimposing detection regions having different contact directions will be described later.

Similarly, as shown in FIG. 10(b), the processing circuit 8 translates and shifts the detection areas 45 and 55 corresponding to the right foot of the same person so that the respective center positions and ground contact directions match. Rotate and move to superimpose. Specifically, first, the detection area 55 is rotated so that the grounding direction 49 and the grounding direction 59 are the same. After that, the detection regions 55 are moved in parallel so that the center points 47 and 57, which are the respective center positions, coincide with each other, and the respective detection regions are superimposed. In this manner, an integration area 62 obtained by superimposing the detection area 45 and the detection area 55 can be obtained. In FIG. 10(b), a portion of the integration area 61 where the detection area 40 and the detection area 50 overlap is indicated by dark hatching. Note that FIG. 10B shows an example in which the grounding directions 49 and 59 are substantially the same to simplify the explanation. An example of superimposing detection regions having different contact directions will be described later.

In the above example, the detection areas are rotationally moved so that their grounding directions match, and then parallelly moved so that their center positions match. You can run them in order. That is, the detection areas may be moved in parallel so that their center positions are aligned, and then rotationally moved so that their grounding directions are aligned. Moreover, the rotational movement and the parallel movement may be performed at the same time.

After step S52, the process advances to step S53 to estimate the shape of each sole based on the integrated areas 61 and 62 obtained in step S52, and the flow ends (see FIG. 9). For example, the processing circuit 8 estimates the shape of a person's sole by extracting a portion where a relatively large number of detection regions are overlapped in the integration regions 61 and 62 .

Specifically, the processing circuit 8 extracts a portion of the integration area 61 where the detection areas 40 and 50 overlap, and estimates the shape of the foot (left foot) corresponding to the integration area 61 . Similarly, the processing circuit 8 extracts a portion of the integration area 62 where the detection areas 45 and 55 overlap, and estimates the shape of the foot (right foot) corresponding to the integration area 62 .

In addition, in the example shown in FIGS. 10A and 10B, the case where the shape of the sole is estimated by superimposing two detection regions has been described for the sake of simplicity. However, in the present embodiment, the larger the number of overlapping detection regions, the more accurately the shape of the sole can be estimated. The mechanism of the present invention will be described in detail below.

11 and 12 are diagrams for explaining the mechanism of the present invention. FIGS. 11(a)-(d) are detection regions 71, 74, 77, 80 respectively acquired for the same foot of the same person. The respective center positions in the detection areas 71, 74, 77, 80 are the center points 72, 75, 78, 81, and the respective ground directions are the ground directions 73, 76, 79, 82. FIG. FIG. 12 shows an integrated area 83 obtained by superimposing the detection areas in FIG. 11, and a sole shape 84 (actual sole shape) corresponding to each detection area. The integration area 83 overlaps the detection areas 71, 74, 77, 80 so that the center positions (center points 72, 75, 78, 81) and the grounding directions (grounding directions 73, 76, 79, 82) match. It was obtained by Note that, in FIG. 12, portions of the integrated area 83 where a plurality of detection areas overlap are indicated by darker hatching according to the number of overlapping detection areas. That is, the portions of the integration region 83 where the detection regions 71, 74, 77, and 80 all overlap are indicated by the darkest hatching. In addition, the portions where three of the detection areas 71, 74, 77, and 80 overlap are indicated by the next darkest hatching.

As shown in FIG. 12, when the integrated area 83 and the actual sole shape 84 are superimposed and compared, it can be seen that the integrated area 83 is darkly hatched on the inner side of the foot and lightly hatched on the outer side of the foot. . This is because the detection cells 10 on the inside of the foot have a high probability of detecting pressure, and the detection cells 10 on the outside of the foot have a low probability of detecting pressure. In other words, the shape of the sole shape 84 can be estimated with high accuracy by extracting a portion where relatively many detection regions overlap in the integration region 83 .

More specifically, the actual sole shape 84 can be obtained with high accuracy by extracting a portion where a predetermined ratio or more of the detection regions overlap in the integrated region 83 (for example, a portion where 50% or more of the detection regions overlap). can be estimated by This is because a portion in the integrated region 83 where the ratio of the overlapping detection regions does not reach the predetermined ratio is considered to correspond to the outer side of the foot, since the detection cell 10 has a low probability of detecting pressure.

As shown in FIGS. 10 and 12, the processing circuit 8 creates an image in which different colors and patterns are added to the areas where many detection areas overlap in the integration area and the areas where the detection areas do not overlap so much. may Such an image can prompt the user to intuitively understand the shape of the sole.

Further, by superimposing the detection areas with different contact directions, the shape of the sole can be obtained with higher accuracy. For example, the grounding directions 73, 76, 79, 82 of the detection areas 71, 74, 77, 80 are different. At this time, an integration area 83 obtained by rotating and overlapping the detection areas 71, 74, 77, and 80 includes not only information in the row and column directions in which the detection cells 10 are arranged, but also information in the oblique direction. It tends to have a rounded shape. Therefore, the curved contour of the sole shape 84 can be accurately represented. Specifically, it is preferable that at least two detection regions having an angle of 10 degrees or more formed by the grounding directions are superimposed on each other.

In the example of FIG. 12, the case where the integration area 83 is created by superimposing four detection areas has been described. By overlapping more detection areas, the S/N ratio of the integration area is improved, and the distinction between the inside and outside of the foot can be made clear.

<Preprocessing of detection area>
FIG. 13 is a diagram for explaining pre-processing in the moving direction identifying method according to the present embodiment. As shown in FIG. 13, the sensor panel 5 has a plurality of detection cells 10 arranged in a matrix in row and column directions. Each detection cell 10 cannot detect pressure with a resolution higher than the sensor pitch. Therefore, in FIG. 13, the detection area is expanded by pseudo-expanding the resolution of each detection cell 10 .

That is, in the preliminary processing shown in FIG. 13, the detection areas 91 and 92 detected by each detection cell 10 shown in the upper diagram of FIG. 13 are expanded as shown in the lower diagram of FIG. Specifically, the detection area detected by each detection cell 10 is expanded concentrically 93 . Then, using each of the expanded detection areas, areas 94 and 95 corresponding to human feet are specified.

By expanding the detection area in this way, the determination processing of the area corresponding to the human foot can be performed with high accuracy. Also, if the detection area is not expanded, the resolution will be according to the sensor pitch, so the detection result will be a discrete detection result. detection results can be obtained. Further, by expanding each detection area, even if the pitch of each detection cell 10 is large compared to the size of a person's foot, it is possible to determine the contact direction of the foot and the movement direction of the person. . Therefore, an area equivalent to the size of a person's foot can be used as a detection target.

As described above, the present invention has been described in accordance with the above embodiments, but the present invention is not limited only to the configurations of the above embodiments, and is applicable within the scope of the invention of the claims of the present application. Needless to say, it includes various modifications, modifications, and combinations that can be made by a trader.
For example, in the above-described embodiment, the processing circuit 8 executes each step related to the sole shape estimation method, but the user may execute part or all of each step.

Further, in step S52, if the number of detection regions corresponding to one leg (first leg) of the same person is less than a predetermined number (for example, five), the other leg (second leg) of the same person is detected. foot) may be used to estimate the shape of the first foot. In this case, the processing circuit 8 first turns over the detection area corresponding to the second foot. After that, the processing circuit 8 translates and rotates the detection area corresponding to the first foot and the detection area corresponding to the turned-over second foot so that the center position and the contact direction match. Then overlap. By doing so, the shape of the sole can be obtained with higher accuracy.

For example, in the case of FIG. 8C, there are two detection areas corresponding to the left foot (here, the first foot), detection area 40 and detection area 50, and the number of detection areas is less than the predetermined number. Therefore, the processing circuit 8 turns over the detection areas 45, 55 corresponding to the right foot (here, the second foot) of the same person and superimposes the respective detection areas. This is because the shape in which the right foot is turned inside out can be approximated as a shape representing the left foot.

FIG. 14 shows an example in which an integration area 62_2 obtained by flipping an integration area 62, which is an overlay of the detection areas 45 and 55, is overlaid on an integration area 61, which is an overlay of the detection areas 40 and 50. FIG. An integration area 63 obtained by superimposing the integration area 61 and the integration area 62_2 is substantially an area obtained by overlapping four detection areas corresponding to the left foot. Therefore, it can be said that the shape of the sole 64 is represented with higher accuracy than the integration area 61 obtained by superimposing two detection areas corresponding to the left foot. That is, by estimating the shape of the sole based on the integrated area 63 obtained in this manner, the shape of the sole can be estimated with higher accuracy.

Further, in the above embodiment, the rectangular area of the detection cell 10 in which the pressure is detected is regarded as one unit, and the set of these rectangular areas is defined as the detection area. A sensing point may be regarded as one unit, and a detection area may be extracted based on a set of this sensing point. FIGS. 15(a)-(d) are sensor panels 5 corresponding to FIGS. 11(a)-(d), respectively. 15(a) to (d), each circle drawn in each detection cell 10 indicates a pressure detection sensing point 18 of the detection cell 10. In FIG. The sensing point 18 of the sensing cell 10 that sensed pressure is indicated as pressure sensing point 19 .

As an example of determining the detection area based on the sensing points, a set of concentric circles of the sensing points (pressure detection points 19) of the detection cell 10 that detected the pressure can be set as the detection area. FIG. 15 shows an example in which detection regions 171 , 174 , 177 , and 180 are set of concentric circles of respective pressure detection points 19 . When the detection area is determined in this way, the shape of the detection area does not depend on the shape of the detection cells 10, so the detection area more strongly reflects the shape of the sole. Therefore, the shape of the sole can be estimated with higher accuracy.

Also, at this time, the shape of the person's sole may be estimated based on a plurality of sensing points included in a plurality of superimposed detection regions. FIG. 16 shows a distribution 183 of each sensing point 18 and pressure detection point 19 when the detection areas 171, 174, 177, 180 are superimposed so that the center position and ground contact direction are the same. Here, it can be said that the portion where the density of the pressure detection points 19 is high in the distribution 183 corresponds to the portion where relatively many detection areas overlap. Therefore, by extracting a portion with a high density of pressure detection points 19 in the distribution 183, the sole shape 84 can be estimated with high accuracy.

In addition, the foot sole shape estimation method according to the present invention can estimate the shape of a person's sole with high accuracy, so that the walking trace of a pedestrian can be estimated with high accuracy. Therefore, it is possible to accurately analyze walking information unique to a walker, such as stride length (the distance between both feet in the walking direction) and stride distance (the distance between both feet in the direction orthogonal to the walking direction). By using such walking information together with information on the shape of the sole, it is possible to perform personal authentication of the walker with higher accuracy.

1 pressure sensor device 5 sensor panel 8 processing circuit 9 recording device 10 detection cell 11 substrates 12, 13 lower electrode 14 upper electrode 15 film 17 spacer

Claims (10)

A sole shape estimation method for estimating the shape of the sole of a person walking on a sensor panel using a sensor panel in which a plurality of detection cells for detecting pressure are arranged in a matrix in row and column directions. hand,
a first step of acquiring temporal walking data of a person walking on the sensor panel using the plurality of detection cells;
a second step of extracting detection regions corresponding to the shape of the sole of the person using the acquired walking data;
a third step of obtaining a center position of each of the plurality of extracted detection areas;
a fourth step of obtaining a grounding direction of the person's sole in each of the extracted detection regions, using temporal information of the detection cells that detected the pressure;
a fifth step of estimating the shape of the person's foot sole by superimposing the plurality of detection areas so that the center position and the ground contact direction match, respectively;
Foot sole shape estimation method. In the fifth step, estimating the shape of the person's sole by extracting a portion where relatively many detection areas overlap,
The sole shape estimation method according to claim 1. In the fifth step, estimating the shape of the person's sole by extracting a portion where a predetermined ratio or more of the detection areas overlap,
The sole shape estimation method according to claim 2. extracting a detection region corresponding to the first foot of the same person before the fifth step;
In the fifth step, estimating the shape of the sole of the first foot by superimposing the detection regions corresponding to the first foot.
The sole shape estimation method according to any one of claims 1 to 3. Before the fifth step, extracting a detection area corresponding to a first foot of the same person; and extracting a detection area corresponding to a second foot on a side different from the first foot of the same person. and extracting
In the fifth step, after turning over the detection area corresponding to the second foot, the detection area corresponding to the first foot and the detection area corresponding to the turned over second foot are overlapped. By matching, estimating the shape of the sole of the first foot,
The sole shape estimation method according to any one of claims 1 to 4. In the second step, the plurality of sensing areas are extracted based on the sensing points of the sensing cells that detected the pressure.
The sole shape estimation method according to any one of claims 1 to 5. In the fifth step, estimating the shape of the sole of the person based on the plurality of sensing points included in the plurality of superimposed detection areas;
The sole shape estimation method according to claim 6. In the fifth step, estimating the shape of the sole of the person by extracting a portion with a high density of the plurality of sensing points included in the plurality of superimposed detection regions;
The sole shape estimation method according to claim 7. The grounding direction is obtained based on the moving direction over time of the center point of the detection cell that detects the pressure in the detection area.
The sole shape estimation method according to any one of claims 1 to 8. a sensor panel in which a plurality of detection cells for detecting pressure are arranged in a matrix in row and column directions;
a processing circuit for estimating the shape of the sole of a person walking on the sensor panel based on the pressure detected by the plurality of detection cells;
The processing circuit is
Acquiring temporal walking data of a person walking on the sensor panel using the plurality of detection cells;
Extracting detection regions corresponding to the shape of the sole of the person using the acquired walking data,
Acquiring the center position of each of the plurality of extracted detection areas,
Acquiring the contact direction of the person's sole in each of the extracted detection regions using the temporal information of the detection cells that detected the pressure;
estimating the shape of the sole of the person by superimposing the plurality of detection areas so that the center position and the ground contact direction are aligned with each other;
A pressure sensor device.

Images