CN102499638A - Living body detection system based on vision, hearing, smell and touch - Google Patents

Living body detection system based on vision, hearing, smell and touch Download PDF

Info

Publication number
CN102499638A
CN102499638A CN2011103013781A CN201110301378A CN102499638A CN 102499638 A CN102499638 A CN 102499638A CN 2011103013781 A CN2011103013781 A CN 2011103013781A CN 201110301378 A CN201110301378 A CN 201110301378A CN 102499638 A CN102499638 A CN 102499638A
Authority
CN
China
Prior art keywords
module
acoustical signal
target
filtering
motor
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN2011103013781A
Other languages
Chinese (zh)
Other versions
CN102499638B (en
Inventor
郑红
刘振强
隋强强
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beihang University
Original Assignee
Beihang University
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Beihang University filed Critical Beihang University
Priority to CN 201110301378 priority Critical patent/CN102499638B/en
Publication of CN102499638A publication Critical patent/CN102499638A/en
Application granted granted Critical
Publication of CN102499638B publication Critical patent/CN102499638B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Abstract

The invention discloses a living body detection system based on vision, hearing, smell and touch. The system comprises a vision module, a hearing module, a smell module, a touch module, a moving carrier, a motor drive module and a target information fusion module, wherein the target information fusion module receives the sensing information output by the vision module, the hearing module, the smell module and the touch module. In the invention, the target is searched by processing sound signals and vision signals; the sound signals are used for judging fuzzy directions; the vision signals are used for accurately positioning the target distance and the appearance; after the target position and appearance are determined, smell can be adopted to sense whether the target exhales CO2 to discriminate whether the target has the vital signs similar to breath to determine whether the target survives; and then the tail ends of mechanical arms are utilized to press some parts of the living body and such life states of the target are further confirmed according to the touch sensing signals as heart rate, skin elasticity and the like, thus further confirming the living body state.

Description

A kind of live body detection system of touching of smelling based on audiovisual
Technical field
The present invention relates to the general life live body of a kind of basis and have characteristics such as shape, sounding, abnormal smells from the patient, body skin sense of touch, smell by audiovisual and touch a kind of live body detection system of touching of smelling based on audiovisual that multiple sensors module and two dsp controllers are formed.
Background technology
Live body is defined as lived animal and plant body.Live body is surveyed and is meant discovery live body target and confirms whether it is the process of live body, and this process relates to target various features information collecting and judgement.
Live body Detection Techniques based on the artificial intelligence can be applied to the live body target search under danger or the complex environment, in disaster search and rescue and complex scene search procedure, have important use and are worth.The live body Detection Techniques are dabbled wider, relate generally to the acquisition technique of various information, the mode identification technology of different types of information and technological to the mechanism design that adapts to varying environment.
To the demand of above live body Detection Techniques, the present invention propose a kind of structure flexibly, convenient, the modular live body detection system of control based on many information fusion, can realize the detection quick and precisely of live body target.
Summary of the invention
The purpose of this invention is to provide a kind of live body detection system of touching of smelling based on audiovisual; This detection system is the control treatment core with two DSP; Realize information gathering, fusion, the processing of a plurality of sensor measuring modules (vision, audition, sense of touch, olfactory sensation), to come identification target direction, attribute, classification through target appearance, sound, abnormal smells from the patient, pliability.System is the control treatment core with two DSP; Control realizes the information gathering of four sensor measuring modules (audition, vision, olfactory sensation, sense of touch); And a tactic movement actuator, thereby through the fusion treatment of the sound of target, outward appearance, abnormal smells from the patient, tactile impressions information is come the identification target direction and is judged whether to be live body.
Of the present inventionly a kind ofly smell the live body detection system of touching based on audiovisual, this live body detection system includes vision module (1), audition module (2), olfactory sensation module (3), sense of touch module (4), motion carrier (5), motor drive module (6) and target information Fusion Module (7);
Vision module (1) is used to gather the image information D1{DL1 under the acquisition environment, and DR1}, DL1 represent the image information that left CMOS photographic head (11) is gathered, and DR1 representes the image information that right CMOS photographic head (12) is gathered;
Audition module (2) is used to gather the acoustic information D2{FDL2 under the acquisition environment; FDR2}; FDL2 representes that the acoustical signal DL2 that first sound transducer (2A) is gathered amplifies through high-accuracy low noise amplifier circuit (2C); Acoustical signal after the amplification is through the left-hand filtering acoustical signal of butterworth filter circuit (2D) filtering output; FDR2 representes that the acoustical signal DR2 that rising tone sound sensor (2B) is gathered amplifies through high-accuracy low noise amplifier circuit (2C), and the acoustical signal after the amplification is through the dextrad filtering acoustical signal of butterworth filter circuit (2D) filtering output;
Olfactory sensation module (3) is used to gather airborne CO under the acquisition environment 2Concentration information D3;
Sense of touch module (4) is used to gather the soft object information under the acquisition environment, and this softness object information will convert voltage signal D4 output to;
Motion carrier (5) is used to carry vision module (1), audition module (2), olfactory sensation module (3) and sense of touch module (4);
Motor drive module (6) is used for driven visual module (1), audition module (2), olfactory sensation module (3) and sense of touch module (4) motion;
Target information Fusion Module (7) issues motor control signal on the one hand and instructs the vision module (1), sense of touch module (4), the motion carrier (5) that are driven by motor drive module (6) to realize motion; Image information D1{DL1 to receiving on the other hand, DR1}, acoustic information D2{FDL2, FDR2}, pressure information D4 and CO 2Concentration information D3 handles, and obtains searching for the live body target.
The present invention is based on audiovisual smells the advantage of the live body detection system of touching and is:
1. binocular is gathered visual information in real time, but binocular is searched the different directions target respectively, and establishing target three-dimensional information outward appearance and range determination.
2. the ears microphone obtains acoustic information, and the realization Sounnd source direction is distinguished, and the strong and weak perception of sound source, as breathes power etc., can further confirm the vital sign of live body.
3. olfactory sensation module is through carbon dioxide (CO 2) concentration sensor can distinguish CO in the environment 2Concentration, detecting CO 2Concentration change, the breath state of perception live body is with its vital sign of further affirmation.
4. the PVDF touch sensor is realized the pressure measxurement to beformable body, and tactile impressions information is provided, and deeply differentiates for physical property and the live body existing state of differentiating target.
5. the live body detection that the information of touching is melted algorithm realization high efficiency, high-accuracy is smelt in audiovisual, has improved the live body detectivity.
6. vision, audition, tactile motion assist to realize accurate direction distinguishing and sense of touch excitation, and the assurance system surveys correctness.
7. system of the present invention can be used as the search of live body target and finds that its size is little, is easy to transplant.
Description of drawings
Fig. 1 the present invention is based on the structured flowchart that the live body detection system of touching is smelt in audiovisual.
Figure 1A be motor drive module of the present invention with by the structured flowchart of driven object.
Fig. 2 the present invention is based on audiovisual to smell the sketch map that the transducing signal of the live body detection system of touching flows to.
Fig. 3 the present invention is based on a kind of enforcement structure chart that the live body detection system of touching is smelt in audiovisual.
Fig. 4 the present invention is based on the process chart that target information Fusion Module in the live body detection system of touching is smelt in audiovisual.
Fig. 5 A is the circuit theory diagrams of high-accuracy low noise amplifier circuit.
Fig. 5 B and Fig. 5 C are the circuit theory diagrams of the fertile filter circuit of Bart.
Fig. 5 D is the circuit theory diagrams of pressure signal conditioning circuit.
Fig. 5 E is the circuit theory diagrams of webcam driver circuit.
Fig. 5 F is the circuit theory diagrams of motor-drive circuit.
1. vision module 11. left CMOS photographic head 12. right CMOS photographic head
13. the first webcam driver circuit, 14. second webcam driver circuit
2. the audition module 2A. first sound transducer 2B. rising tone sound sensor
2C. high-accuracy low noise amplifier circuit 2D. butterworth filter circuit
3. olfactory sensation module 4. sense of touch module 4A. feeler arm 4B.PVDF piezoelectric transducers
4C. charge amplifying circuit 4D. voltage amplifier circuit 4E. phase modulation circuit 4F. filter circuit
5. axial elongated slot 52. lower platforms of motion carrier 51. upper mounting plate 51A.X 53. left-hand rotation platforms
53A. the right spheroid of left spheroid 53B. left side U-shaped support 53C. left-hand rotation body 54. right-hand rotation platform 54A.
54B. right U-shaped support 54C. right-hand rotation body 55. mobile platforms 6. motor drive modules
6A. the right spheroid motor of left spheroid motor 6B. 6C. left-hand rotation platform motor
The into motor 6D. right-hand rotation platform motor 6E. mobile platform motor 6F. feeler arm moves
6G. motor-drive circuit 7. target information Fusion Modules
The specific embodiment
To combine accompanying drawing that the present invention is done further detailed description below.
Referring to Fig. 1, Figure 1A, shown in Figure 2; The present invention a kind ofly smells the live body detection system of touching based on audiovisual, and this live body detection system includes vision module 1, audition module 2, olfactory sensation module 3, sense of touch module 4, motion carrier 5, motor drive module 6 and target information Fusion Module 7.
Vision module 1, audition module 2, olfactory sensation module 3 and sense of touch module 4 are installed in motion carrier 5 outsides; Motor drive module 6 is formed with in the cavity with lower platform 52 with the upper mounting plate 51 that target information Fusion Module 7 is installed in motion carrier 5.
(1) motion carrier 5
Referring to Fig. 1, Figure 1A, Fig. 2, shown in Figure 3, motion carrier 5 includes upper mounting plate 51, lower platform 52, left-hand rotation platform 53, right-hand rotation platform 54, mobile platform 55.Upper mounting plate 51 is installed in the top of lower platform 52, and left-hand rotation platform 53 is installed in the X axis elongated slot 51A of upper mounting plate 51 with right-hand rotation platform 54.Olfactory sensation module 3 is installed in upper mounting plate 51 tops.The bottom installation and moving platform 55 of lower platform 52, mobile platform 55 is through a plurality of wheel walkings of bottom.
Left-hand rotation platform 53 includes left spheroid 53A, left U-shaped support 53B, left-hand rotation body 53C and left base; Left-hand rotation body 53C is installed in the top of left side base, and left U-shaped support 53B is installed in the top of left-hand rotation body 53C, and left spheroid 53A is installed at the middle part of left U-shaped support 53B; Left side spheroid 53A goes up left CMOS photographic head 11 is installed, and this left side CMOS photographic head 11 is used to gather the image information under the acquisition environment.On the outer panel of left side U-shaped support 53B the first sound transducer 2A is installed, this first sound transducer 2A is used to gather the sound under the acquisition environment.
Right-hand rotation platform 54 includes right spheroid 54A, right U-shaped support 54B, right-hand rotation body 54C and right base; Right-hand rotation body 54C is installed in the top of right base, and right U-shaped support 54B is installed in the top of right-hand rotation body 54C, and right spheroid 54A is installed at the middle part of right U-shaped support 54B; Right spheroid 54A goes up right CMOS photographic head 12 is installed, and this right side CMOS photographic head 12 is used to gather the image information under the acquisition environment.On the outer panel of right U-shaped support 54B rising tone sound sensor 2B is installed, this rising tone sound sensor 2B is used to gather the sound under the acquisition environment.
Upper mounting plate 51 is a disc structure, and the middle part of upper mounting plate 51 is provided with X axis elongated slot 51A, and the left base of left-hand rotation platform 53, the right base of right-hand rotation platform 54 are installed in this X axis elongated slot 51A.Being designed with of this X axis elongated slot 51A is beneficial to left-hand rotation platform 53, right-hand rotation platform 54 slides in its elongated slot, also is the qualification to left-hand rotation platform 53, the installation site of right-hand rotation platform 54 on upper mounting plate 51 simultaneously.X axis elongated slot 51A can guarantee platform 53, the motion of right-hand rotation platform 54 on the X axle.
Lower platform 52 is a disc structure.
Upper mounting plate 51 will be formed with a cavity after assembling with lower platform 52, and this cavity is used to be equipped with motor drive module 6 and target information Fusion Module 7.
In the present invention; The Design of Mechanical Structure of motion carrier 5 through optimizing; Can vision module 1, audition module 2, olfactory sensation module 3, sense of touch module 4, motor drive module 6 and target information Fusion Module 7 be combined, this helps the search to life entity.
In the present invention, shown in Figure 1A, the driven object of motion carrier 5 under motor drive module 6 is:
(A) motion of the left spheroid 53B in the left-hand rotation platform 53 is driven by left spheroid motor 6A; The IMAQ that the left CMOS photographic head 11 that the motion of left side spheroid 53B and then make is installed on the left spheroid 53B is accomplished on X-direction; Left spheroid 53B is positioned at the top of left-hand rotation body 53C simultaneously; Left-hand rotation body 53C moves around Y direction, thus the IMAQ that left CMOS photographic head 11 is accomplished on Y direction.
(B) motion of the right spheroid 54B in the right-hand rotation platform 54 is driven by right spheroid motor 6B; The IMAQ that the right CMOS photographic head 12 that the motion of right spheroid 54B and then make is installed on the right spheroid 54B is accomplished on X-direction; Right spheroid 54B is positioned at the top of right-hand rotation body 54C simultaneously; Right-hand rotation body 54C moves around Y direction, thus the IMAQ that right CMOS photographic head 12 is accomplished on Y direction.
(C) motion of the left-hand rotation body 53C in the left-hand rotation platform 53 is driven by left-hand rotation platform motor 6C, the motion of left-hand rotation body 53C and then make the first sound transducer 2A that is installed in left U-shaped support 53B side accomplish the sound collection on Y direction.
(D) motion of the right-hand rotation body 54C in the right-hand rotation platform 54 is driven by right-hand rotation platform motor 6D, the motion of right-hand rotation body 54C and then make the rising tone sound sensor 2B that is installed in right U-shaped support 54B side accomplish the sound collection on Y direction.
(E) motion of mobile platform 55 is driven by mobile platform motor 6E.
(2) vision module 1
Vision module 1 is used to gather the image information D1{DL1 under the acquisition environment, DR1}.
Referring to Fig. 1, shown in Figure 2, vision module 1 includes left CMOS photographic head 11, right CMOS photographic head 12, the first webcam driver circuit 13, the second webcam driver circuit 14;
The image information that left side CMOS photographic head 11 is gathered is designated as left-hand image information DL1; This left-hand image information DL1 transfers in the first processor of target information Fusion Module 7 through the first webcam driver circuit 13.
The image information that right CMOS photographic head 12 is gathered is designated as dextrad image information DR1; This dextrad image information DR1 transfers in the first processor of target information Fusion Module 7 through the second webcam driver circuit 14.Referring to shown in Figure 2, the first webcam driver circuit 13 and the second webcam driver circuit 14 are connected on the VPORT interface of first processor (selecting the DM642 chip for use).
The first webcam driver circuit 13 provides power supply and clock signal for left CMOS photographic head 11.
The second webcam driver circuit 14 provides power supply and clock signal for right CMOS photographic head 12.
Left side CMOS photographic head 11 is installed on the left spheroid 53A, and right CMOS photographic head 12 is installed on the right spheroid 54A.
The first webcam driver circuit 13,14 designs of the second webcam driver circuit are on circuit board, and shown in Fig. 5 E, the webcam driver circuit adopts the SN74CBTD3384 chip, and this circuit board is installed in the cavity that upper mounting plate 51 and lower platform 52 assembling backs form.
(3) audition module 2
Audition module 2 is used to gather the acoustic information D2{FDL2 under the acquisition environment, FDR2}.
Referring to Fig. 1, shown in Figure 2, audition module 2 includes the first sound transducer 2A, rising tone sound sensor 2B, high-accuracy low noise amplifier circuit 2C and butterworth filter circuit 2D;
On the outer panel of left side U-shaped support 53B the first sound transducer 2A is installed, the acoustical signal that the first sound transducer 2A gathers is designated as left-hand acoustical signal DL2.
On the outer panel of right U-shaped support 54B rising tone sound sensor 2B is installed, the acoustical signal that rising tone sound sensor 2B gathers is designated as dextrad acoustical signal DR2.
Left-hand acoustical signal DL2 amplifies through high-accuracy low noise amplifier circuit 2C, the acoustical signal after the amplification through butterworth filter circuit 2D filtering output filtering after acoustical signal, i.e. left-hand filtering acoustical signal FDL2;
Dextrad acoustical signal DR2 amplifies through high-accuracy low noise amplifier circuit 2C, the acoustical signal after the amplification through butterworth filter circuit 2D filtering output filtering after acoustical signal, i.e. dextrad filtering acoustical signal FDR2;
Left-hand filtering acoustical signal FDL2 and dextrad filtering acoustical signal FDR2 transfer in second processor of target information Fusion Module 7.Referring to shown in Figure 2, butterworth filter circuit 2D outfan is connected with the A/D interface of second processor (selecting the F2812 chip for use) of target information Fusion Module 7.
In the present invention, the first sound transducer 2A is identical with rising tone sound sensor 2B structure, selects electret microphone for use.
Shown in Fig. 5 A, Fig. 5 B, Fig. 5 C, in the present invention, high-accuracy low noise amplifier circuit 2C and butterworth filter circuit 2D constitute the acoustical signal modulate circuit.The design of acoustical signal modulate circuit is on circuit board, and this circuit board is installed in upper mounting plate 51 and assembles in the cavity of back formation with lower platform 52.
(4) olfactory sensation module 3
Referring to Fig. 1, Fig. 2, shown in Figure 3, olfactory sensation module 3 is used to gather airborne CO under the acquisition environment 2Concentration information D3.
Olfactory sensation module 3 is by a CO 2Concentration sensor constitutes.This CO 2Concentration sensor is according to the CO of variable concentrations 2Absorbability to the infrared light of specific wavelength is different, the CO in the Measurement of Air 2Concentration, and with CO 2Concentration signal is transported in the target information Fusion Module 7 with the form of PWM ripple.
Referring to shown in Figure 2, CO 2Concentration sensor is connected on the PWM interface of second processor (selecting the F2812 chip for use) of target information Fusion Module 7.
(5) sense of touch module 4
Sense of touch module 4 is used to gather the soft object information under the acquisition environment, and this softness object information will convert voltage signal D4 output to.
Referring to Fig. 1, Fig. 2, shown in Figure 3, sense of touch module 4 includes feeler arm 4A, PVDF piezoelectric transducer 4B, charge amplifying circuit 4C, voltage amplifier circuit 4D, phase modulation circuit 4E and filter circuit 4F.
Charge amplifying circuit 4C, voltage amplifier circuit 4D, phase modulation circuit 4E and filter circuit 4F constitute the piezoelectric signal modulate circuit, shown in Fig. 5 D.
PVDF piezoelectric transducer 4B is installed in the end of feeler arm 4A.
Feeler arm 4A is the multi freedom degree mechanical arm, can push the object under the acquisition environment, thereby make PVDF piezoelectric transducer 4B obtain the beformable body contact information.Shown in Figure 1A, the motion of feeler arm 4A is driven by feeler arm motor 6F.
PVDF piezoelectric transducer 4B utilizes the piezoelectric property of PVDF (polyvinylidene fluoride) thin film and pliability to realize the conversion of the piezoelectricity in the beformable body contact process, thereby obtains pressure information FD4; This pressure information FD4 carries out one-level through charge amplifying circuit 4C and amplifies; Carrying out secondary through voltage amplifier circuit 4D again amplifies; Then through phase modulation circuit 4E with signal inversion and increase DC component; After filter circuit 4F filtering output, pressure signal D4 transfers in second processor (selecting the F2812 chip for use) of target information Fusion Module 7 after the filtering of output.The outfan of filter circuit 4F is connected on the AD interface of second processor (selecting the F2812 chip for use).
(6) motor drive module 6
Referring to Figure 1A, shown in Figure 2, motor drive module 6 includes 6 motors (left spheroid motor 6A, right spheroid motor 6B, left-hand rotation platform motor 6C, right-hand rotation platform motor 6D, mobile platform motor 6E, feeler arm move into motor 6F) and a motor-drive circuit 6G.Motor-drive circuit 6G receives the motor control signal that second processor issues and respectively 6 motors is controlled, and issues the motion that the different motor control signal can be accomplished vision module 1, audition module 2, sense of touch module 4.Motor-drive circuit 6G adopts the TPS767D318 chip, and its former line of reasoning is shown in Fig. 5 F.
The F2812 chip provides motor control signal to motor-drive circuit through the PWM interface, and motor-drive circuit produces each road motor of electricity driving mutually according to control signal.
(7) the target information Fusion Module 7
Referring to shown in Figure 2, target information Fusion Module 7 includes the first processor and second processor; First processor adopts the DM642 chip to realize; The second processor adopting F2812 chip is realized.The first processor and second processor are serial communication mode, and promptly the SCI interface of second processor is connected with the McBSP interface of first processor.
On the F2812 chip, adopt the method for compilation and C language hybrid programming, and develop based on the CCS IDE.
The picture signal of vision module 1 output is input in the first processor.
The CO of audition module 2 output sound signals, 3 outputs of olfactory sensation module 2The pressure signal of concentration signal and 4 outputs of sense of touch module is input in second processor.
What the present invention designed smells the live body detection system of touching based on audiovisual, uses in four kinds of transducing signal target approach information fusion modules 7 and handles, and many information fusion of operation strategy is to find and the affirmation target in target information Fusion Module 7.This many information fusion strategy adopts the method for compilation and C language hybrid programming, develops based on the CCS IDE.The multiple heat transfer agent integrated treatment of many information fusion strategy of the present invention through vision module 1, audition module 2, olfactory sensation module 3 and sense of touch module 4 are produced realizes the detection of live body target step by step, and its flow process is as shown in Figure 4.Its concrete steps are following:
Step 1: judge what whether left-hand filtering acoustical signal FDL2 produced for the live body target; If then get into step 2; If not, then get into step 3;
Judge what whether dextrad filtering acoustical signal FDR2 produced for the live body target; If then get into step 2; If not, then get into step 3;
Step 2: relatively the acoustical signal of left-hand filtering acoustical signal FDL2 and dextrad filtering acoustical signal FDR2 is strong and weak;
If left-hand filtering acoustical signal FDL2 is different with the acoustical signal power of dextrad filtering acoustical signal FDR2; Target information Fusion Module 7 sends movement instruction to motor drive module 6; Instruct the side of motion carrier 5 a little less than acoustical signal to move by motor drive module 6, use audition module 2 collected sound signals then again; And return step 1;
If left-hand filtering acoustical signal FDL2 is identical with the acoustical signal power of dextrad filtering acoustical signal FDR2, motion carrier 5 stops to move, and gets into step 3;
Step 3: search judges among the left-hand image information DL1 detection of a target whether occurs; If target occurs, then get into step 4; If target do not occur, return step 1;
Search judges among the dextrad image information DR1 detection of a target whether occurs; If target occurs, then get into step 4; If target do not occur, return step 1;
Step 4: judge CO in the air 2Whether concentration information D3 is higher than normal value (1000ppm); If be higher than normal CO 2Concentration value, then live body confirms that probability increases, otherwise constant; Get into step 5;
Step 5: the degree of flexibility of confirming ferret out under the acquisition environment according to the variation of pressure signal D4 after the filtering in the real-time gatherer process; If soft variation difference is big, then live body confirms that probability increases; Otherwise reduce, get into step 6;
Step 6: four judged results of comprehensive step 1 to step 5, confirm whether ferret out is live body; If finish live body and survey; If not, return step 1, go deep into the live body target search again.
What the present invention designed smells the live body detection system of touching based on audiovisual; This system utilizes multiple sensors to gather the information under the acquisition environment; The information that pick off obtains can be carried out the efficient search and the multistage affirmation of live body target through the processing of target information Fusion Module 7, has realized the detection of live body target rapidly and accurately.Through the processing ferret out to acoustical signal and visual signal, acoustical signal is used to judge fuzzy orientation in the present invention; Visual signal is used for accurate localizing objects distance and exterior appearance.Behind definite target location and exterior appearance, can be through the detected by scent target CO that whether breathes out 2, differentiate the vital sign whether target has similar breathing, whether survive to confirm target; Then, utilize mechanical arm terminal (end is equipped with pick off) to push some position of live body (like heart, skin), further confirm to comprise the life state of target heart beating number of times, skin elasticity etc., thereby further confirm condition of living organism according to the tactilely-perceptible signal.

Claims (10)

1. smell the live body detection system of touching based on audiovisual for one kind, it is characterized in that: this live body detection system includes vision module (1), audition module (2), olfactory sensation module (3), sense of touch module (4), motion carrier (5), motor drive module (6) and target information Fusion Module (7);
Vision module (1) is used to gather the image information D1{DL1 under the acquisition environment, and DR1}, DL1 represent the image information that left CMOS photographic head (11) is gathered, and DR1 representes the image information that right CMOS photographic head (12) is gathered;
Audition module (2) is used to gather the acoustic information D2{FDL2 under the acquisition environment; FDR2}; FDL2 representes that the acoustical signal DL2 that first sound transducer (2A) is gathered amplifies through high-accuracy low noise amplifier circuit (2C); Acoustical signal after the amplification is through the left-hand filtering acoustical signal of butterworth filter circuit (2D) filtering output; FDR2 representes that the acoustical signal DR2 that rising tone sound sensor (2B) is gathered amplifies through high-accuracy low noise amplifier circuit (2C), and the acoustical signal after the amplification is through the dextrad filtering acoustical signal of butterworth filter circuit (2D) filtering output;
Olfactory sensation module (3) is used to gather airborne CO under the acquisition environment 2Concentration information D3;
Sense of touch module (4) is used to gather the soft object information under the acquisition environment, and this softness object information will convert voltage signal D4 output to;
Motion carrier (5) is used to carry vision module (1), audition module (2), olfactory sensation module (3) and sense of touch module (4);
Motor drive module (6) is used for driven visual module (1), audition module (2), olfactory sensation module (3) and sense of touch module (4) motion;
Target information Fusion Module (7) issues motor control signal on the one hand and instructs the vision module (1), sense of touch module (4), the motion carrier (5) that are driven by motor drive module (6) to realize motion; Image information D1{DL1 to receiving on the other hand, DR1}, acoustic information D2{FDL2, FDR2}, pressure information D4 and CO 2Concentration information D3 handles, and obtains ferret out.
2. according to claim 1ly smell the live body detection system of touching based on audiovisual, it is characterized in that: motion carrier (5) includes upper mounting plate (51), lower platform (52), left-hand rotation platform (53), right-hand rotation platform (54), mobile platform (55); Upper mounting plate (51) is installed in the top of lower platform (52), and left-hand rotation platform (53) is installed in the X axis elongated slot (51A) of upper mounting plate (51) with right-hand rotation platform (54); Olfactory sensation module (3) is installed in upper mounting plate (51) top; The bottom installation and moving platform (55) of lower platform (52);
Left-hand rotation platform (53) includes left spheroid (53A), left U-shaped support (53B), left-hand rotation body (53C) and left base; Left-hand rotation body (53C) is installed in the top of left side base, and left U-shaped support (53B) is installed in the top of left-hand rotation body (53C), and left spheroid (53A) is installed at the middle part of left U-shaped support (53B); Left side spheroid (53A) is gone up left CMOS photographic head (11) is installed, and first sound transducer (2A) is installed on the outer panel of left U-shaped support (53B);
Right-hand rotation platform (54) includes right spheroid (54A), right U-shaped support (54B), right-hand rotation body (54C) and right base; Right-hand rotation body (54C) is installed in the top of right base, and right U-shaped support (54B) is installed in the top of right-hand rotation body (54C), and right spheroid (54A) is installed at the middle part of right U-shaped support (54B); Right spheroid (54A) is gone up right CMOS photographic head (12) is installed, and rising tone sound sensor (2B) is installed on the outer panel of right U-shaped support (54B).
3. according to claim 1ly smell the live body detection system of touching based on audiovisual, it is characterized in that: vision module (1) includes left CMOS photographic head (11), right CMOS photographic head (12), the first webcam driver circuit (13), the second webcam driver circuit (14);
The image information that left side CMOS photographic head (11) is gathered is designated as left-hand image information DL1; This left-hand image information DL1 transfers in the first processor of target information Fusion Module (7) through the first webcam driver circuit (13);
The image information that right CMOS photographic head (12) is gathered is designated as dextrad image information DR1; This dextrad image information DR1 transfers in the first processor of target information Fusion Module (7) through the second webcam driver circuit (14);
The first webcam driver circuit (13) provides power supply and clock signal for left CMOS photographic head (11);
The second webcam driver circuit (14) provides power supply and clock signal for right CMOS photographic head (12).
4. according to claim 1ly smell the live body detection system of touching based on audiovisual, it is characterized in that: audition module (2) includes first sound transducer (2A), rising tone sound sensor (2B), high-accuracy low noise amplifier circuit (2C) and butterworth filter circuit (2D);
On the outer panel of left side U-shaped support (53B) first sound transducer (2A) is installed, the acoustical signal that first sound transducer (2A) is gathered is designated as left-hand acoustical signal DL2;
On the outer panel of right U-shaped support (54B) rising tone sound sensor (2B) is installed, the acoustical signal that rising tone sound sensor (2B) is gathered is designated as dextrad acoustical signal DR2;
Left-hand acoustical signal DL2 amplifies through high-accuracy low noise amplifier circuit (2C), the acoustical signal after the amplification through butterworth filter circuit (2D) filtering output filtering after acoustical signal, i.e. left-hand filtering acoustical signal FDL2;
Dextrad acoustical signal DR2 amplifies through high-accuracy low noise amplifier circuit (2C), the acoustical signal after the amplification through butterworth filter circuit (2D) filtering output filtering after acoustical signal, i.e. dextrad filtering acoustical signal FDR2;
Left-hand filtering acoustical signal FDL2 and dextrad filtering acoustical signal FDR2 transfer in second processor of target information Fusion Module (7);
First sound transducer (2A) is identical with rising tone sound sensor (2B) structure, selects electret microphone for use.
5. according to claim 1ly smell the live body detection system of touching based on audiovisual, it is characterized in that: olfactory sensation module (3) is a CO 2Concentration sensor.
6. according to claim 1ly smell the live body detection system of touching based on audiovisual, it is characterized in that: sense of touch module (4) includes feeler arm (4A), PVDF piezoelectric transducer (4B), charge amplifying circuit (4C), voltage amplifier circuit (4D), phase modulation circuit (4E) and filter circuit (4F);
PVDF piezoelectric transducer (4B) is installed in the end of feeler arm (4A);
The motion of feeler arm (4A) is driven by feeler arm motor (6F);
PVDF piezoelectric transducer (4B) utilizes the piezoelectric property of PVDF (polyvinylidene fluoride) thin film and pliability to realize the conversion of the piezoelectricity in the beformable body contact process, thereby obtains pressure information FD4; This pressure information FD4 carries out one-level through charge amplifying circuit (4C) and amplifies; Carrying out secondary through voltage amplifier circuit (4D) again amplifies; Then through phase modulation circuit (4E) with signal inversion and increase DC component; After filter circuit (4F) filtering output, pressure signal D4 transfers in second processor of target information Fusion Module (7) after the filtering of output.
7. according to claim 6ly smell the live body detection system of touching based on audiovisual, it is characterized in that: feeler arm (4A) is the multi freedom degree mechanical arm.
8. according to claim 1ly smell the live body detection system of touching based on audiovisual, it is characterized in that: motor drive module (6) includes left spheroid motor (6A), right spheroid motor (6B), left-hand rotation platform motor (6C), right-hand rotation platform motor (6D), mobile platform motor (6E), feeler arm move into motor (6F), motor-drive circuit (6G);
The motor control signal that second processor of motor-drive circuit (6G) receiving target information fusion module (7) issues moves control to left spheroid motor (6A), right spheroid motor (6B), left-hand rotation platform motor (6C), right-hand rotation platform motor (6D), mobile platform motor (6E), the feeler arm motor (6F) that moves into respectively.
9. according to claim 1ly smell the live body detection system of touching based on audiovisual, it is characterized in that: target information Fusion Module (7) includes the first processor and second processor; First processor adopts the DM642 chip to realize; The second processor adopting F2812 chip is realized; The first processor and second processor are serial communication mode, and promptly the SCI interface of second processor is connected with the McBSP interface of first processor.
10. according to claim 1ly smell the live body detection system of touching, it is characterized in that: the image information D1{DL1 of target information Fusion Module (7) to receiving, DR1}, acoustic information D2{FDL2, FDR2}, pressure information D4 and CO based on audiovisual 2Concentration information D3 treatment step is:
Step 1: judge what whether left-hand filtering acoustical signal FDL2 produced for the live body target; If then get into step 2; If not, then get into step 3;
Judge what whether dextrad filtering acoustical signal FDR2 produced for the live body target; If then get into step 2; If not, then get into step 3;
Step 2: relatively the acoustical signal of left-hand filtering acoustical signal FDL2 and dextrad filtering acoustical signal FDR2 is strong and weak;
If left-hand filtering acoustical signal FDL2 is different with the acoustical signal power of dextrad filtering acoustical signal FDR2; Target information Fusion Module 7 sends movement instruction to motor drive module 6; Instruct the side of motion carrier 5 a little less than acoustical signal to move by motor drive module 6, use audition module 2 collected sound signals then again; And return step 1;
If left-hand filtering acoustical signal FDL2 is identical with the acoustical signal power of dextrad filtering acoustical signal FDR2, motion carrier 5 stops to move, and gets into step 3;
Step 3: search judges among the left-hand image information DL1 detection of a target whether occurs; If target occurs, then get into step 4; If target do not occur, return step 1;
Search judges among the dextrad image information DR1 detection of a target whether occurs; If target occurs, then get into step 4; If target do not occur, return step 1;
Step 4: judge CO in the air 2Whether concentration information D3 is higher than normal value (1000ppm); If be higher than normal CO 2Concentration value, then live body confirms that probability increases, otherwise constant; Get into step 5;
Step 5: the degree of flexibility of confirming ferret out under the acquisition environment according to the variation of pressure signal D4 after the filtering in the real-time gatherer process; If soft variation difference is big, then live body confirms that probability increases; Otherwise reduce, get into step 6;
Step 6: four judged results of comprehensive step 1 to step 5, confirm whether ferret out is live body; If finish live body and survey; If not, return step 1.
CN 201110301378 2011-09-28 2011-09-28 Living body detection system based on vision, hearing, smell and touch Active CN102499638B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN 201110301378 CN102499638B (en) 2011-09-28 2011-09-28 Living body detection system based on vision, hearing, smell and touch

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN 201110301378 CN102499638B (en) 2011-09-28 2011-09-28 Living body detection system based on vision, hearing, smell and touch

Publications (2)

Publication Number Publication Date
CN102499638A true CN102499638A (en) 2012-06-20
CN102499638B CN102499638B (en) 2013-07-03

Family

ID=46211820

Family Applications (1)

Application Number Title Priority Date Filing Date
CN 201110301378 Active CN102499638B (en) 2011-09-28 2011-09-28 Living body detection system based on vision, hearing, smell and touch

Country Status (1)

Country Link
CN (1) CN102499638B (en)

Cited By (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105813011A (en) * 2016-05-12 2016-07-27 陈冠桥 Manipulator device on-site sensing signal detection method based on mobile phone client
WO2016179902A1 (en) * 2015-05-09 2016-11-17 深圳市贝沃德克生物技术研究院有限公司 Patient-end remote auxiliary diagnosis and treatment device and remote auxiliary diagnosis and treatment method
WO2016197298A1 (en) * 2015-06-08 2016-12-15 北京旷视科技有限公司 Living body detection method, living body detection system and computer program product
CN107832803A (en) * 2017-11-27 2018-03-23 胡明建 The design method that a kind of computer vision and mechanical smell are mutually mapped with the time
CN108108437A (en) * 2017-12-20 2018-06-01 胡明建 The design method that a kind of computer vision sense of hearing smell is mutually mapped with the time
CN108778097A (en) * 2016-02-25 2018-11-09 三星电子株式会社 Device and method for assessing heart failure
CN110870761A (en) * 2018-08-30 2020-03-10 中国科学院沈阳自动化研究所 Skin detection system based on mixed perception of visual sense and tactile sense
CN113607782A (en) * 2021-07-28 2021-11-05 浙江工业大学 Visual perception type water quality early warning system and method enhanced by olfactory signal
CN114512153A (en) * 2022-01-17 2022-05-17 浙江大学 Audio-visual touch data acquisition terminal for garbage

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4998282A (en) * 1987-12-09 1991-03-05 Olympus Optical Co., Ltd. Lifesaving borescope system
WO1995020171A1 (en) * 1994-01-20 1995-07-27 Selectronic Vertriebsgesellschaft Für Sicherheitstechnik Und Sonderelektronik Mbh Process and device for detecting vital functions in living bodies
JP2007180942A (en) * 2005-12-28 2007-07-12 National Institute Of Advanced Industrial & Technology In-vehicle sound source searching system
CN101807246A (en) * 2010-03-17 2010-08-18 上海大学 Multi-character vivo identification method and device

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US4998282A (en) * 1987-12-09 1991-03-05 Olympus Optical Co., Ltd. Lifesaving borescope system
WO1995020171A1 (en) * 1994-01-20 1995-07-27 Selectronic Vertriebsgesellschaft Für Sicherheitstechnik Und Sonderelektronik Mbh Process and device for detecting vital functions in living bodies
JP2007180942A (en) * 2005-12-28 2007-07-12 National Institute Of Advanced Industrial & Technology In-vehicle sound source searching system
CN101807246A (en) * 2010-03-17 2010-08-18 上海大学 Multi-character vivo identification method and device

Cited By (11)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2016179902A1 (en) * 2015-05-09 2016-11-17 深圳市贝沃德克生物技术研究院有限公司 Patient-end remote auxiliary diagnosis and treatment device and remote auxiliary diagnosis and treatment method
WO2016197298A1 (en) * 2015-06-08 2016-12-15 北京旷视科技有限公司 Living body detection method, living body detection system and computer program product
CN108778097A (en) * 2016-02-25 2018-11-09 三星电子株式会社 Device and method for assessing heart failure
CN108778097B (en) * 2016-02-25 2021-08-06 三星电子株式会社 Devices and methods for assessing heart failure
CN105813011A (en) * 2016-05-12 2016-07-27 陈冠桥 Manipulator device on-site sensing signal detection method based on mobile phone client
CN107832803A (en) * 2017-11-27 2018-03-23 胡明建 The design method that a kind of computer vision and mechanical smell are mutually mapped with the time
CN108108437A (en) * 2017-12-20 2018-06-01 胡明建 The design method that a kind of computer vision sense of hearing smell is mutually mapped with the time
CN110870761A (en) * 2018-08-30 2020-03-10 中国科学院沈阳自动化研究所 Skin detection system based on mixed perception of visual sense and tactile sense
CN110870761B (en) * 2018-08-30 2021-07-27 中国科学院沈阳自动化研究所 Skin detection system based on mixed perception of visual sense and tactile sense
CN113607782A (en) * 2021-07-28 2021-11-05 浙江工业大学 Visual perception type water quality early warning system and method enhanced by olfactory signal
CN114512153A (en) * 2022-01-17 2022-05-17 浙江大学 Audio-visual touch data acquisition terminal for garbage

Also Published As

Publication number Publication date
CN102499638B (en) 2013-07-03

Similar Documents

Publication Publication Date Title
CN102499638B (en) Living body detection system based on vision, hearing, smell and touch
CN101549498B (en) Automatic tracking and navigation system of intelligent aid type walking robots
US11898848B2 (en) Visual navigation for mobile devices operable in differing environmental lighting conditions
CN102895092A (en) Multi-sensor integration based three-dimensional environment identifying system for walker aid robot
US20190192361A1 (en) Mobility and mobility system
CN106388700B (en) Active noise reduction device for automatic cleaning equipment and automatic cleaning equipment
CN105058389A (en) Robot system, robot control method, and robot
CN206643934U (en) Multi-information acquisition perceives search and rescue robot
CN102178540B (en) Three-wheeled omnidirectional mobile control device and auxiliary system for autistic children custody
CN208543477U (en) A kind of guest-meeting robot
CN206780416U (en) A kind of intelligent medical assistant robot
CN109366504A (en) A kind of intelligence exhibition and fair service robot system
CN109259948B (en) Wheelchair for assisting driving
CN107322600A (en) A kind of programmable education robot system
US11654554B2 (en) Artificial intelligence cleaning robot and method thereof
CN107065892A (en) Life detection and environmental data collecting search and rescue car after the calamity of controlled in wireless
CN108733059A (en) A kind of guide method and robot
US20230050825A1 (en) Hands-Free Crowd Sourced Indoor Navigation System and Method for Guiding Blind and Visually Impaired Persons
CN207087877U (en) A kind of programmable education robot system
KR102366327B1 (en) Moving Robot and controlling method
WO2019127437A1 (en) Map labeling method and apparatus, and cloud server, terminal and application program
CN114260919B (en) Intelligent robot
Mazo et al. Integral system for assisted mobility
Jakubiak et al. Control and perception system for ReMeDi robot mobile platform
CN113059574A (en) Intelligent accompanying method of accompanying robot and intelligent accompanying robot

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
C14 Grant of patent or utility model
GR01 Patent grant