CN107423658B

CN107423658B - Gait recognition method and device

Info

Publication number: CN107423658B
Application number: CN201610349345.7A
Authority: CN
Inventors: 不公告发明人
Original assignee: Beijing Hai Bai Sichuan Science And Technology Co ltd
Current assignee: Beijing Hai Bai Sichuan Science and Technology Co.,Ltd.
Priority date: 2016-05-24
Filing date: 2016-05-24
Publication date: 2022-03-22
Anticipated expiration: 2036-05-24
Also published as: CN107423658A

Abstract

The application discloses a gait recognition method and a device, wherein pressure states corresponding to a toe part, an arch part and a heel part of a sole under each gait phase are defined firstly. And then acquiring the pressure states of the toe part, the arch part and the heel part at the same time. The identification method divides the states of all parts into high pressure states or low pressure states, and compares the pressure states collected by the toe part, the arch part and the heel part with the defined gait phase at the same time, thereby judging the gait phase at the moment and obtaining more accurate judgment results.

Description

Gait recognition method and device

Technical Field

The application relates to the field of human exoskeleton robots, in particular to human gait recognition.

Background

Human gait is an important aspect of human motion research, relates to the fields of motion rehabilitation, robot control and the like, and is particularly important for accurately identifying human gait in mechanical exoskeleton robot control. The mechanical exoskeleton technology is a hot spot at present, and can enhance the motion function of the human body and improve the motion capability of the human body by providing additional power for the human body, so that the mechanical exoskeleton technology is used in the military and industrial fields; on the other hand, the rehabilitation training device can also assist patients with lower limb dysfunction to perform rehabilitation training in the medical field.

In human exoskeleton robot control, there are two main ways to identify lower limb gait: based on electromyographic signals or based on plantar pressure sensor signals. The electromyographic signals are high in cost, need to be in complete contact with the surface of a human body, are weak and are difficult to identify in mode. It is common today to perform gait recognition by measuring plantar pressure sensor signals. However, the accuracy of the existing gait recognition method by plantar pressure detection still needs to be improved.

Disclosure of Invention

The application provides a novel gait recognition method and a novel gait recognition device.

The gait recognition method provided by the application comprises the following steps:

a defining step: defining pressure states corresponding to a toe part, an arch part and a heel part of a sole under each gait phase respectively, wherein the pressure states are divided into a high pressure state and a low pressure state;

a signal acquisition step: acquiring pressure detection signals of a toe part, an arch part and a heel part of a sole at the same time;

a conversion step: converting the acquired pressure detection signal into a digital signal representing a high pressure state or a low pressure state;

a judging step: comparing the acquired pressure states of the toe part, the arch part and the heel part at the same time with the pressure states defined by each gait phase, and if the comparison result is consistent with the comparison result of a certain step phase, judging that the toe part, the arch part and the heel part are in the gait phase at present; if the gait phases are different from all the gait phases, judging the gait phases to be abnormal;

the execution steps are as follows: and when the current step phase is judged, sending an execution instruction to control an execution mechanism to execute the action matched with the gait.

As a further improvement of the gait recognition method, in the conversion step, after the acquired pressure detection signals are subjected to filtering processing, each pressure detection signal is divided into a high pressure state or a low pressure state by adopting a fuzzy recognition algorithm.

As a further improvement of the gait recognition method, a phase change sequence of each gait is defined, and each gait phase jumps according to the sequence.

As a further improvement of the gait recognition method, if the currently detected gait phase is not located after the executing gait phase in the defined change sequence, no jump occurs, and the state of the executing gait phase is maintained.

As a further improvement of the gait recognition method, the toe region pressure collection point includes a big toe, and the arch region pressure collection point includes an arch inner side and an arch outer side.

As a further improvement of the gait recognition method, the gait phase includes:

s1: the initial contact phase, the big toe area under the initial contact phase is correspondingly in a low-pressure state, the inner side area of the arch of foot is correspondingly in a low-pressure state, the outer side area of the arch of foot is correspondingly in a low-pressure state, and the heel area is correspondingly in a high-pressure state;

s2: loading a feedback phase, wherein a large toe area under the feedback phase corresponds to a low-pressure state, an arch inner area corresponds to a low-pressure state, an arch outer area corresponds to a high-pressure state, and a heel area corresponds to a high-pressure state;

s3: an intermediate support phase, which is the following:

the big toe area is correspondingly in a low-pressure state, the inner side area of the arch is correspondingly in a high-pressure state, the outer side area of the arch is correspondingly in a high-pressure state, and the heel area is correspondingly in a high-pressure state;

or the big toe area is in a low pressure state correspondingly, the inner side area of the arch is in a high pressure state correspondingly, the outer side area of the arch is in a low pressure state correspondingly, and the heel area is in a high pressure state correspondingly;

or the big toe area is correspondingly in a high pressure state, the inner side area of the arch is correspondingly in a high pressure state, the outer side area of the arch is correspondingly in a low pressure state, and the heel area is correspondingly in a high pressure state;

or the big toe area is correspondingly in a high-pressure state, the inner side area of the arch is correspondingly in a high-pressure state, the outer side area of the arch is correspondingly in a high-pressure state, and the heel area is correspondingly in a high-pressure state;

s4: supporting the end phase, which is the following:

the big toe area is correspondingly in a low-pressure state, the inner side area of the arch is correspondingly in a high-pressure state, the outer side area of the arch is correspondingly in a high-pressure state, and the heel area is correspondingly in a low-pressure state;

or the big toe area is in a low pressure state correspondingly, the inner side area of the arch is in a low pressure state correspondingly, the outer side area of the arch is in a high pressure state correspondingly, and the heel area is in a low pressure state correspondingly;

or the big toe area is in a low pressure state correspondingly, the inner side area of the arch is in a high pressure state correspondingly, the outer side area of the arch is in a low pressure state correspondingly, and the heel area is in a low pressure state correspondingly;

or the big toe area is correspondingly in a high pressure state, the inner side area of the arch is correspondingly in a high pressure state, the outer side area of the arch is correspondingly in a high pressure state, and the heel area is correspondingly in a low pressure state;

s5: pre-swing phase, this phase is the following:

the big toe area is in a high pressure state correspondingly, the inner side area of the arch of foot is in a low pressure state correspondingly, the outer side area of the arch of foot is in a low pressure state correspondingly, and the heel area is in a low pressure state correspondingly;

or the big toe area is in a high pressure state correspondingly, the inner side area of the arch is in a high pressure state correspondingly, the outer side area of the arch is in a low pressure state correspondingly, and the heel area is in a low pressure state correspondingly;

s6: a swing phase, wherein the area of the big toe is correspondingly in a low-pressure state, the area of the inner side of the arch is correspondingly in a low-pressure state, the area of the outer side of the arch is correspondingly in a low-pressure state, and the area of the heel is correspondingly in a low-pressure state;

the phases are sequentially cyclically jumped in the order of S1 to S2, S2 to S3, S3 to S4, S4 to S5, S5 to S6, S6 to S1.

As a further improvement of the gait recognition method, in the abnormal state:

the big toe area is in a high pressure state correspondingly, the inner side area of the arch of foot is in a low pressure state correspondingly, the outer side area of the arch of foot is in a low pressure state correspondingly, and the heel area is in a high pressure state correspondingly;

or the big toe area is in a high pressure state correspondingly, the inner side area of the arch is in a low pressure state correspondingly, the outer side area of the arch is in a high pressure state correspondingly, and the heel area is in a low pressure state correspondingly;

or the big toe area is correspondingly in a high pressure state, the inner side area of the arch is correspondingly in a low pressure state, the outer side area of the arch is correspondingly in a high pressure state, and the heel area is correspondingly in a high pressure state;

and when the abnormal state is judged, jumping does not occur, and the state of the executing gait phase is kept.

The application provides a gait recognition device, includes:

the device comprises a defining module, a judging module and a judging module, wherein the defining module is used for defining pressure states corresponding to a toe part, an arch part and a heel part of a sole under each gait phase respectively, and the pressure states are divided into a high pressure state and a low pressure state;

the signal acquisition module is used for acquiring pressure detection signals of a toe part, an arch part and a heel part of a sole at the same time;

the conversion module is used for converting the acquired pressure detection signal into a digital signal representing a high pressure state or a low pressure state;

the judging module is used for comparing the pressure states of the toe part, the arch part and the heel part at the same time with the pressure states defined under each gait phase, and judging that the gait phase is currently positioned if the comparison result is consistent with the comparison result of a certain step phase; if the gait phases are different from all the gait phases, judging the gait phases to be abnormal;

and the execution module sends out an execution instruction when judging that the current gait phase is in a certain step phase, and controls the execution mechanism to execute the action matched with the gait.

As a further improvement of the gait recognition device, the conversion module divides each pressure detection signal into a high pressure state or a low pressure state by adopting a fuzzy recognition algorithm after filtering the acquired pressure detection signals.

As a further improvement of the gait recognition device, the gait recognition device further comprises a gait phase jumping module which is used for defining the sequence of the phase change of each gait and jumping each gait phase according to the sequence.

As a further improvement of the gait recognition device, if the currently detected gait phase is not located after the executing gait phase in the defined change sequence, no jump occurs, and the state of the executing gait phase is maintained.

As a further improvement of the gait recognition device, the toe region pressure collection point comprises a big toe, and the arch region pressure collection point comprises an arch inner side and an arch outer side.

As a further improvement of the gait recognition device, the gait phase includes:

s3: an intermediate support phase, which is the following:

s4: supporting the end phase, which is the following:

s5: pre-swing phase, this phase is the following:

As a further improvement of the gait recognition device, in the abnormal state:

The beneficial effect of this application is:

in the gait recognition method and device provided by the application, the pressure states corresponding to the toe part, the arch part and the heel part of the sole under each gait phase are defined. And then acquiring the pressure states of the toe part, the arch part and the heel part at the same time. The identification method divides the states of all parts into high pressure states or low pressure states, and compares the pressure states collected by the toe part, the arch part and the heel part with the defined gait phase at the same time, thereby judging the gait phase at the moment and obtaining more accurate judgment results.

Drawings

Fig. 1 is an exploded view of an embodiment of a plantar pressure detection device according to the present application;

FIG. 2 is a schematic view of a supporting layer structure in an embodiment of the plantar pressure detection device of the present application;

FIG. 3 is a schematic view of a covering layer structure according to an embodiment of the present invention;

FIG. 4 is a block flow diagram of a gait recognition method of the present application;

FIG. 5 is a diagram illustrating a hardware configuration of a gait recognition system according to the present application;

FIG. 6 is a schematic block diagram of a conversion step in the gait recognition method of the present application;

fig. 7 is a schematic diagram of a jump sequence of each gait phase in the gait recognition method of the present application.

Detailed Description

The present invention will be described in further detail with reference to the following detailed description and accompanying drawings. The present application may be embodied in many different forms and is not limited to the embodiments described in the present embodiment. The following detailed description is provided to facilitate a more thorough understanding of the present disclosure, and the words used to indicate orientation, top, bottom, left, right, etc. are used solely to describe the illustrated structure in connection with the accompanying figures.

One skilled in the relevant art will recognize, however, that one or more of the specific details can be omitted, or other methods, components, or materials can be used. In some instances, some embodiments are not described or not described in detail.

Furthermore, the technical features, aspects or characteristics described herein may be combined in any suitable manner in one or more embodiments. It will be readily appreciated by those of skill in the art that the order of the steps or operations of the methods associated with the embodiments provided herein may be varied. Thus, any sequence in the figures and examples is for illustrative purposes only and does not imply a requirement in a certain order unless explicitly stated to require a certain order.

The first embodiment is as follows:

the first embodiment provides a sole pressure detection mechanism, which can collect pressure signals of all parts of the sole of a human body. The control system can utilize the pressure signal collected by the mechanism to carry out subsequent processing, such as judging the motion of the feet of the human body, thereby controlling the execution mechanism to execute corresponding motion.

Referring to fig. 1, the sole pressure detecting mechanism includes:

a support layer 100;

a pressure sensor (which is conventional and not shown) mounted on the support layer 100;

and a cover layer 200 for being disposed on the sole of a human body, the cover layer 200 including a toe region a for diffusing pressure applied to the toe portion of the human body, an arch region B for diffusing pressure applied to the arch portion of the human body, and a heel region C for diffusing pressure applied to the heel portion of the human body. The cover layer 200 is attached to the sensing part of the pressure sensor with a gap between the cover layer 200 and the support layer 100. At least three pressure sensors are arranged on the support layer 100, and at least one pressure sensor is respectively corresponding to each of the toe area a, the arch area B and the heel area C, so as to realize pressure detection of the three areas.

The cover layer 200 is used to contact the bare foot or the sole of the subject to transmit the pressure generated from the bare foot or the sole to the pressure sensor of the support layer 100. When the cover layer 200 is provided to be fitted to the sole, there is a benefit in that the user does not need to barefoot, but only needs to wear the shoe in contact with the cover layer 200.

The gap is formed between the cover layer 200 and the support layer 100, which enables the pressure of the sole of the foot to be intensively transmitted to the pressure sensor of the support layer 100, thereby improving the accuracy of detection. Meanwhile, after the force of the sole is transmitted to the covering layer 200, regional change can be caused, the sensor can collect signals more easily, and the applicability is stronger.

For the covering layer 200, it may be a unitary structure, such as a unitary sheet-like structure; or it may be formed by combining three separate components, such as a separate component for each of the three regions, and finally joining the three regions together to form the entire cover 200. It is also possible that two of the three regions are integrated and the other region is separately provided and finally the entire cover layer 200 is split.

The toe region a, the arch region B, and the heel region C are divided in different areas from the sole contact area of the human body. The sole of a human body is divided into a toe part, an arch part and a heel part, wherein the toe part comprises all toes including a thumb, the heel part is a rear heel part, and an area between the toe part and the heel part is the arch part. The area of the present cover layer 200 that is used for indirect or direct contact with the toe portion is the toe area a, the area that is in indirect or direct contact with the arch portion is the arch area B, and the area that is in indirect or direct contact with the heel portion is the heel area C.

The direct contact means that the cover layer 200 is directly laid on the sole of the foot, and the indirect contact means that the cover layer 200 is laid on the sole of the shoe or the like.

Referring to fig. 1, the mechanism is provided with at least one pressure sensor in a toe area a, an arch area B and a heel area C respectively to acquire pressure signals in the areas. When the detected person moves, the movement of the detected person is judged by the pressure change of the three areas.

As an example, the present application collects pressure signals of the big toe in the toe portion, and thus the toe area a of the cover layer 200 includes a big toe area corresponding to the big toe of the human body, and at least one pressure sensor for detecting pressure in the area is provided corresponding to the big toe area.

As an example, the arch portion of the present application collects pressure signals on the outside of the arch and on the inside of the arch. Specifically, the areas of the left and right human arch parts close to each other are the inner sides, and the areas away from each other are the outer sides. The arch region B of the cover layer 200 includes an outer arch region corresponding to the outer arch of the human body and an inner arch region corresponding to the inner arch of the human body, and the outer arch region and the inner arch region are respectively provided with at least one pressure sensor for detecting pressure in the regions.

Specifically, referring to fig. 1 and 2, in the present embodiment, four pressure sensors are provided, and correspond to the big toe region, the outer region of the arch of foot, the inner region of the arch of foot, and the heel region one by one, respectively. The four sensors can collect pressure signals at four different places, so that the further analysis of the sole pressure of the human body is facilitated. The change rule of the four sensors can effectively reflect the change condition of the gait of the human body.

In particular, the support layer 100 is provided with recessed

cavities

111, 112, 113, 114 for mounting pressure sensors in correspondence of the big toe area, the lateral arch area, the medial arch area and the heel area. The hollow cavity 111 corresponds to the big toe area, the hollow cavity 112 corresponds to the inner side of the arch, the hollow cavity 113 corresponds to the outer side of the arch, and the hollow cavity 114 corresponds to the heel area. The pressure sensor may employ a pressure sensitive sensor element of a thin film type. Meanwhile, the support layer 100 is further provided with a concave wire groove 120 for routing, and the wires are hidden in the wire groove 120 and led out from the side of the arch region, so that the protrusion of the wires is prevented from interfering with the force transmission of the covering layer 200.

In addition, a plurality of fixing holes 130 are further formed in the supporting layer 100, the covering layer 200 and the sole can be connected through the fixing holes 130 and a wearable structure, the whole supporting layer 100 and the covering layer 200 are connected together, wearing by a detected person is facilitated, and contact with the sole during walking is guaranteed.

The covering layer 200 can spread the pressure of the human body from the stress point to a certain range, for example, taking the big toe as an example, in the existing detection structure, the pressure sensor is directly aligned with the big toe for detection, but the size of the sole of each person is different, and the general position of the used sensor is fixed, so the size difference is easy to cause dislocation, and the applicability is not good. In the embodiment, the pressure of the big toe acting on the covering layer 200 can cause regional change and spread to a certain range, and the pressure sensor can detect a corresponding pressure signal when contacting with the range, so that the influence caused by dislocation is avoided, and the adaptability of the mechanism is improved.

The cover layer 200 may alternatively be made of plastic, for example a laminate of 1-3mm (e.g. 2mm) PVC (polyvinyl chloride) material.

Further, the cover layer 200 is spaced apart from the two planes of the support layer 100, so that the force can be transmitted to the four pressure sensors of the support layer 100.

The spacing may be selected to be between 0.3 and 0.7mm, preferably 0.5 mm.

Further, referring to fig. 3, a surface of the cover layer 200 facing the pressure sensor has a raised contact pad 210, and the contact pad 210 is attached to the detection portion of the pressure sensor. When there are four pressure sensors, the number of the contact pads 210 is also four, and the position distribution of the contact pads corresponds to the pressure sensors, and corresponds to the support layer 100 one by one, as shown in fig. 3.

Furthermore, it is also possible to arrange the detection portion of the pressure sensor protruding from the support layer 100, thereby ensuring that the cover layer 200 only contacts the pressure sensor and not the support layer 100.

Example two:

the second embodiment provides a gait recognition method, which is used for recognizing human gait and further outputting a corresponding instruction to an execution mechanism so that the execution mechanism cooperates with the human gait motion.

Referring to fig. 4, the gait recognition method includes:

a signal acquisition step: acquiring pressure detection signals of a toe part, an arch part and a heel part at the same time;

In the defining step, the pressure states corresponding to the toe part, the arch part and the heel part of the sole under each gait phase are defined according to the gait rule of the human body, for example, the pressure states of the toe part, the arch part and the heel part under the actions of standing still, walking and the like of the human body are defined.

The pressure collection points of the toe part, the arch part and the heel part can be selected according to the requirement. In order to ensure accuracy, the toe pressure collection points can be big toes, and the arch pressure collection points can be the inner side and the outer side of the arch.

The identification method may be based on a hardware system as shown in fig. 5. The sensor array is used for acquiring pressure signals of the soles of the human bodies, and the specific acquisition mode can be performed as the structure shown in the first embodiment, or can be realized by adopting any structure in the prior art.

The amplification circuit in fig. 5 is used to amplify the acquired signal. The AD circuit converts an analog signal into a digital signal. The central processor is used for judging the gait phase and sending a corresponding instruction to the executing mechanism.

The sensors respectively collect pressure signals of the soles of the human bodies, the analog signals are converted into digital quantity through signal amplification, and data are transmitted to the central processing unit through the USB high-speed interface.

Of course, the fig. 5 shows only one hardware structure for implementing the identification method, and the identification method is not limited to be implemented by the structure shown in fig. 5.

Further, in the converting step, after the acquired pressure detection signals are subjected to filtering processing, each pressure detection signal is divided into a high pressure state or a low pressure state by using a fuzzy recognition algorithm.

Specifically, as shown in fig. 6, the filter is designed to filter the pressure signal. The pressure signal collected by the sensor is mixed with a lot of noise and interference, and a low-pass filter is designed by using Butterworth according to the sampling frequency and the gait walking period to remove the interference signal.

And then, fusing data of the filtered signals according to the human gait rule. The method adopts the existing fuzzy recognition algorithm, selects a membership function with high matching degree, divides each sensor signal into two states of high pressure and low pressure, and combines the change rule of gait to recognize the gait phase.

Further, the pressure signals are not completely consistent in the walking process, and large signal changes or interference of filtering can occur, so that sudden change of gait phases can be caused, and the sent instructions are inconsistent with the current gait of the human body.

To solve the above problem, the sequence of each gait phase can be defined, and each gait phase jumps in sequence. Particularly, the change sequence of each phase is defined through a state machine, and the stability of gait recognition is improved. When the gait phase identified by the sensor is suddenly changed and is not in the appointed jump phase, the identification error is explained.

Specifically, if the currently detected gait phase is not located after the executing gait phase in the defined change sequence, no jump occurs, and the state of the executing gait phase is maintained. Thereby preventing sudden changes in gait phase.

Referring to the following table, the following table shows the pressure status of the corresponding parts under each gait:

the method comprises the following steps:

s3: an intermediate support phase, which is the following:

s4: supporting the end phase, which is the following:

s5: pre-swing phase, this phase is the following:

referring to fig. 7, the above phases are cyclically jumped in sequence in the order of S1 to S2, S2 to S3, S3 to S4, S4 to S5, S5 to S6, and S6 to S1.

According to the characteristics of a person when walking, each gait phase can be divided into an initial contact phase, a loading feedback phase, a swing phase and a support phase when the person walks, wherein the swing phase comprises a pre-swing phase and the swing phase, and the support phase comprises a middle support phase and a support end phase.

The above phases are detected for a single foot, wherein:

the initial contact phase (S1) is in a swing state and just turns into a support state, and the heel just touches the ground;

the loading feedback phase (S2) is a state that the landing leg starts to be mainly stressed;

the middle supporting phase (S3) mainly transfers the gravity center to the landing leg to become a supporting leg, and the other leg swings forwards;

supporting the end phase (S4) to force the arch of the supporting leg forward;

the pre-swing phase (S5) is that the other leg is supported by landing, and the supporting leg only bears the force of the tiptoe;

the swing phase (S6) is the swing state that the original supporting leg is in the forward leg lifting state.

For each phase, the control system sends out a corresponding command, the output command is mainly determined according to the specific control quantity of the system, generally the rotating position or speed of each joint, and the action of the execution structure is to drive the joint to move to the expected position or speed.

Further, the definition of the abnormal state is as follows:

When a certain motion state is determined by the identification method, the central processing unit sends out an instruction to control the execution mechanism to carry out corresponding motion.

Example three:

a third embodiment provides a gait recognition device corresponding to the second embodiment, including:

Furthermore, after the obtained pressure detection signals are filtered by the conversion module, each pressure detection signal is divided into a high pressure state or a low pressure state by adopting a fuzzy recognition algorithm.

Furthermore, the walking phase jumping module is used for defining the sequence of each walking phase change, and each walking phase jumps according to the sequence.

If the currently detected gait phase is not located after the executing gait phase in the defined change sequence, no jump occurs and the state of the executing gait phase is maintained.

Further, the toe position pressure collection point comprises the big toe, and the arch position pressure collection point comprises the inner arch side and the outer arch side.

Further, the gait phase comprises:

s3: an intermediate support phase, which is the following:

s4: supporting the end phase, which is the following:

s5: pre-swing phase, this phase is the following:

According to the characteristics of people when walking, the walking state of people can be divided into a swinging phase and a supporting phase.

The above phases are detected for a single foot, wherein:

supporting the end phase (S4) to force the arch of the supporting leg forward;

Further, the abnormal state is defined as follows:

Example four:

the fourth embodiment provides a mechanical exoskeleton device.

The mechanical exoskeleton device comprises a control system, an execution mechanism and a sole pressure detection mechanism as shown in the first embodiment, wherein the control system judges the action of a human body according to a sole pressure signal of the human body acquired by the sole pressure detection mechanism and controls the execution mechanism to execute corresponding action.

The control system judges the action of the human body based on the pressure signals acquired by the plantar pressure detection mechanism, and when the pressure signals acquired by the plantar pressure detection mechanism are more accurate, the judgment of the control system is more accurate, so that the action of the actuating mechanism can be more consistent with the action of the human body.

In addition, the control system of the mechanical exoskeleton device can adopt the identification method or device as shown in the second and third embodiments to complete the identification of the human body action. Alternatively, other gait recognition methods may be used to complete the operation in cooperation with the plantar pressure detection mechanism shown in the first embodiment.

The foregoing is a more detailed description of the present invention that is presented in conjunction with specific embodiments, and the practice of the invention is not to be considered limited to those descriptions. It will be apparent to those skilled in the art that a number of simple derivations or substitutions can be made without departing from the inventive concept.

Claims

1. A gait recognition method, characterized by comprising:

a defining step: defining pressure states corresponding to a toe part, an inner side of an arch, an outer side of the arch and a heel part of the sole of a foot under each gait phase respectively, wherein the pressure states are divided into a high pressure state and a low pressure state;

a signal acquisition step: acquiring pressure detection signals of a toe part, an inner side of an arch, an outer side of the arch and a heel part of a sole at the same time; pressure detection signals of the toe part, the inner side of the arch of foot, the outer side of the arch of foot and the heel part are acquired through a plantar pressure detection mechanism, and the plantar pressure detection mechanism comprises a supporting layer, a pressure sensor and a covering layer; a plurality of pressure sensors are arranged on the supporting layer; the covering layer is attached to the detection part of the pressure sensor, and a gap is formed between the covering layer and the supporting layer;

a judging step: comparing the acquired pressure states represented by digital signals at the toe part, the inner side of the arch, the outer side of the arch and the heel part at the same time with the pressure states represented by digital signals defined by all gait phases, wherein each gait phase comprises an initial contact phase, a loading feedback phase, a middle support phase, a support final phase, a pre-swing phase and a swing phase, and if the comparison result is consistent with that of a certain step phase, judging that the current gait phase is in; if the gait phases are different from all the gait phases, judging the gait phases to be abnormal; the definition of the pressure states corresponding to the toe part, the arch part and the heel part of the sole under each gait phase in the definition step is carried out according to the gait rule of the human body, and comprises the definition according to the pressure states of the toe part, the arch part and the heel part under the standing state of the human body or the walking of the human body; each gait phase under the static standing of the human body comprises an initial contact phase and a loading feedback phase, or each gait phase under the walking of the human body comprises an initial contact phase, a loading feedback phase, a middle support phase, a support end phase, a pre-swing phase and a swing phase;

2. A gait recognition method according to claim 1, characterized in that in the conversion step, after the acquired pressure detection signals are subjected to filtering processing, each pressure detection signal is classified into a high pressure state or a low pressure state using a fuzzy recognition algorithm.

3. A gait recognition method according to claim 1, characterized in that a phase change sequence is defined for each gait phase, each gait phase jumping in sequence.

4. A gait recognition method according to claim 3, characterized in that if the currently detected gait phase is not located after the executing gait phase in the defined change sequence, no jump takes place, keeping the state of the executing gait phase.

5. A gait recognition method according to claim 3, characterized in that said toe region pressure collection point comprises the big toe.

6. A gait recognition method according to claim 5, characterized in that the gait phases include:

s3: an intermediate support phase, which is the following:

s4: supporting the end phase, which is the following:

s5: pre-swing phase, this phase is the following:

7. A gait recognition method according to claim 6, characterized in that in said abnormal state:

8. A gait recognition apparatus, characterized by comprising:

the device comprises a defining module, a judging module and a judging module, wherein the defining module is used for defining pressure states corresponding to a toe part, an inner side of an arch, an outer side of the arch and a heel part of a sole under each gait phase respectively, and the pressure states are divided into a high pressure state and a low pressure state;

the signal acquisition module is used for acquiring pressure detection signals of a toe part, an inner side of an arch, an outer side of the arch and a heel part of a sole at the same time; the signal acquisition module comprises a plantar pressure detection mechanism, and the plantar pressure detection mechanism comprises a supporting layer, a pressure sensor and a covering layer; a plurality of pressure sensors are arranged on the supporting layer; the covering layer is attached to the detection part of the pressure sensor, and a gap is formed between the covering layer and the supporting layer;

the judging module is used for comparing the pressure states which are represented by digital signals at the toe part, the inner side of the arch, the outer side of the arch and the heel part at the same time with the pressure states which are defined by all gait phases and are represented by the digital signals, wherein all the gait phases comprise an initial contact phase, a loading feedback phase, a middle support phase, a support last phase, a pre-swing phase and a swing phase, and if the comparison result of the initial contact phase, the loading feedback phase, the middle support phase, the support last phase, the pre-swing phase and the swing phase is consistent, the current gait phase is judged; if the gait phases are different from all the gait phases, judging the gait phases to be abnormal; the definition of the pressure states respectively corresponding to the toe part, the arch part and the heel part of the sole under each gait phase in the definition step is carried out according to the gait rule of the human body, and comprises the definition according to the pressure states of the toe part, the arch part and the heel part under the static standing state and the walking state of the human body; each gait phase under the static standing of the human body comprises an initial contact phase and a loading feedback phase, and each gait phase under the walking of the human body comprises an initial contact phase, a loading feedback phase, a middle support phase, a support end phase, a pre-swing phase and a swing phase;

9. A gait recognition device according to claim 8, characterized in that the conversion module divides each pressure detection signal into a high pressure state or a low pressure state by using a fuzzy recognition algorithm after filtering the acquired pressure detection signals.

10. A gait recognition apparatus according to claim 8, further comprising a gait phase jump module for defining a sequence of phase changes for each gait, each gait phase jumping in sequence.

11. A gait recognition apparatus according to claim 10, characterized in that if the currently detected gait phase is not located after the executing gait phase in the defined change sequence, no jump takes place, keeping the state of the executing gait phase.

12. The gait recognition apparatus of claim 10, wherein the toe region pressure collection point comprises the big toe.

13. A gait recognition device according to claim 12, characterized in that the gait phase includes:

s3: an intermediate support phase, which is the following:

s4: supporting the end phase, which is the following:

s5: pre-swing phase, this phase is the following:

14. A gait recognition apparatus according to claim 13, characterized in that in the abnormal state: