Summary of the invention
To solve the above problems, the purpose of the present invention is to provide a kind of automated external defibrillator intelligent assistance system, packet
It includes: cloud background server, for storing user information and sending rescue instruction to auxiliary robot according to positioning;Calling for help is set
Standby, for sending distress signals to server, distress signals include user current location information;Auxiliary robot, due to carrying
Medical Devices execute corresponding rescue task to target user according to backstage instruction, and wherein Medical Devices include automated external defibrillator
Device.
On the one hand technical solution used by the present invention solves the problems, such as it is: a kind of automated external defibrillator intelligence auxiliary system
System characterized by comprising cloud background server, for storing user information and being sent according to positioning to auxiliary robot
Rescue instruction;Equipment is called for help, for sending distress signals to server, distress signals include user current location information;Auxiliary
Robot executes corresponding rescue task to target user according to backstage instruction due to carrying Medical Devices, wherein Medical Devices
Including automated external defibrillator.
Further, the calling for help equipment includes but is not limited to mobile terminal and fixed medical first aid station.
Further, the mobile terminal device includes: locating module, for determining the current geographical location of user;Disappear
Sending module is ceased, for sending distress signals to cloud background server, wherein distress signals include that user information and user work as
Preceding geographical location.
Further, the auxiliary robot includes: drive module, for driving body mobile according to path;Navigate mould
Block, according to the location information of rescue instruction, planning action route;Obstacle avoidance module, for passing through infrared ray or acoustic detection, control
Auxiliary robot executes obstacle avoidance.
Further, the auxiliary robot includes: acquisition module, for acquiring target user's physical condition information, is wrapped
It includes but is not limited to heart rate;Display and voice module, for playing corresponding guidance content to operator according to operation equipment.
Further, the auxiliary robot includes: defibrillator, for executing defibrillation procedure to target user;Pattern is adopted
Collect module, for acquiring the face image data to defibrillation object;Characteristic extracting module, for extracting in face image data
The skin pattern feature of face;Model classifiers, the model classifiers including corresponding all age group, are made with face image data
For input, corresponding energy setting parameter is exported;Energy setup module, for the defeated of state modulator defibrillator to be arranged according to energy
Energy value out.
On the other hand technical solution used by the present invention solves the problems, such as it is: a kind of automated external defibrillator intelligence auxiliary
Method, which comprises the following steps: S100, when needing medical rescue, user is by calling for help equipment to cloud backstage
Server sends distress signals, and wherein distress signals include user current location information;S200, cloud background server are by basis
Distress signals generate emergency task and matched assignment instructions are sent to the auxiliary nearest and standby apart from user
Robot;
S300, auxiliary robot carry Medical Devices according to backstage instruction close to target user, execute corresponding rescue and appoint
Business, wherein rescue task includes implementing external defibrillation.
Further, the S300 further include: acquisition target user's physical condition information, including but not limited to heart rate;Root
Entry evaluation is done to the physical condition of target user according to the information of acquisition, and shows assessment result.
Further, the S300 further include: S310, face image data of the acquisition to defibrillation object, estimate and
Assessment in detail;S320, the skin pattern feature for extracting face in face image data carry out entry evaluation to the range of age,
Obtain a specified age bracket;S330, the method based on support vector machines establish the category of model for corresponding to different age group
Device, wherein the method for establishing category of model type is based on histogram of gradients and local binarization Model Establishment training sample database;
S340, the age bracket determined according to step S302, choose the model classifiers of corresponding age bracket, with acquisition to defibrillation object
Face image data is assessed as input terminal, exports age estimated result;S350, preset age and corresponding points are based on
Energy rule is hit, corresponding click energy is selected according to age estimated result.
Further, described 310 further include face's pre-treatment step: A. face extracts, and is carried out based on face detection device CAS
Face extracts;B. five characteristic points of face are detected using Coarse-to-Fine Auto-Encoder Network (CFAN),
It is right and left eyes center, nose, mouth left and right corner respectively;C. feature rear face image data will be extracted to be standardized, and will
Image progress picture size after standardization, which is reseted, sets, wherein standardization includes internal standardization and outer standardization two ways,
Standardized mode not only contains the intrinsic information of face at home and abroad, but also also contains whole contextual information, internal standard
Change contains only facial information.
The beneficial effects of the present invention are: executing external defibrillation to patient automatically, sue and labour for what other needs manually performed
Movement plays corresponding guidance content to operator, and the maximized success rate and treatment rate for guaranteeing defibrillation is reduced because of operation
It is lack of standardization or delay emergency time and cause first aid failure crisis patient vitals probability.
Specific embodiment
It is carried out below with reference to technical effect of the embodiment and attached drawing to design of the invention, specific structure and generation clear
Chu, complete description, to be completely understood by the purpose of the present invention, scheme and effect.
It should be noted that unless otherwise specified, when a certain feature referred to as " fixation ", " connection " are in another feature,
It can directly fix, be connected to another feature, and can also fix, be connected to another feature indirectly.In addition, this
The descriptions such as the upper and lower, left and right used in open are only the mutual alignment pass relative to each component part of the disclosure in attached drawing
For system.The "an" of used singular, " described " and "the" are also intended to including most forms in the disclosure, are removed
Non- context clearly expresses other meaning.In addition, unless otherwise defined, all technical and scientific terms used herein
It is identical as the normally understood meaning of those skilled in the art.Term used in the description is intended merely to describe herein
Specific embodiment is not intended to be limiting of the invention.Term as used herein "and/or" includes one or more relevant
The arbitrary combination of listed item.
It will be appreciated that though various elements, but this may be described using term first, second, third, etc. in the disclosure
A little elements should not necessarily be limited by these terms.These terms are only used to for same type of element being distinguished from each other out.For example, not departing from
In the case where disclosure range, first element can also be referred to as second element, and similarly, second element can also be referred to as
One element.The use of provided in this article any and all example or exemplary language (" such as ", " such as ") is intended merely to more
Illustrate the embodiment of the present invention well, and unless the context requires otherwise, otherwise the scope of the present invention will not be applied and be limited.
LBP refers to local binary patterns, full name in English: Local Binary Patterns.Original function is assistant images office
Portion's contrast is not a complete Feature Descriptor.
Histograms of oriented gradients (Histogram of Oriented Gradient, HOG) is characterized in one kind in computer
It is used to carry out the Feature Descriptor of object detection in vision and image procossing.HOG feature is by calculating and statistical picture partial zones
The gradient orientation histogram in domain carrys out constitutive characteristic.
Principal Component Analysis (PCA): Principal Component Analysis is most common linear dimensionality reduction side
Method, its target are by certain linear projection, and the data of higher-dimension, which are mapped in the space of low-dimensional, to be indicated, i.e., original n
A less m feature of feature number replaces, and new feature is the linear combination of old feature.And it is expected in the dimension projected
The variance of data is maximum, makes m new feature irrelevant as far as possible.Consolidating in the mapping capture data from old feature to new feature
There is variability.Less data dimension is used with this, while retaining the characteristic of more former data point.
It referring to Fig.1, is system schematic according to the preferred embodiment of the invention,
Include: cloud background server, refers to for storing user information and sending rescue to auxiliary robot according to positioning
It enables;Equipment is called for help, for sending distress signals to server, distress signals include user current location information;Auxiliary robot,
Corresponding rescue task is executed to target user according to backstage instruction due to carrying Medical Devices, wherein Medical Devices include automatic
External defibrillator.
Further, the calling for help equipment includes but is not limited to mobile terminal and fixed medical first aid station.
Further, the mobile terminal device includes: locating module, for determining the current geographical location of user;Disappear
Sending module is ceased, for sending distress signals to cloud background server, wherein distress signals include that user information and user work as
Preceding geographical location.
Further, the auxiliary robot includes: drive module, for driving body mobile according to path;Navigate mould
Block, according to the location information of rescue instruction, planning action route;Obstacle avoidance module, for passing through infrared ray or acoustic detection, control
Auxiliary robot executes obstacle avoidance.
Further, the auxiliary robot includes: acquisition module, for acquiring target user's physical condition information, is wrapped
It includes but is not limited to heart rate;Display and voice module, for playing corresponding guidance content to operator according to operation equipment.
Further, the auxiliary robot includes: defibrillator, for executing defibrillation procedure to target user;Pattern is adopted
Collect module, for acquiring the face image data to defibrillation object;Characteristic extracting module, for extracting in face image data
The skin pattern feature of face;Model classifiers, the model classifiers including corresponding all age group, are made with face image data
For input, corresponding energy setting parameter is exported;Energy setup module, for the defeated of state modulator defibrillator to be arranged according to energy
Energy value out.
It is method flow schematic diagram according to the preferred embodiment of the invention referring to Fig. 2,
The following steps are included: S100, when needing medical rescue, user by call for help equipment to cloud background server send out
Distress signals are sent, wherein distress signals include user current location information;S200, cloud background server will be according to distress signals
It generates emergency task and matched assignment instructions is sent to the auxiliary robot nearest and standby apart from user;
S300, auxiliary robot carry Medical Devices according to backstage instruction close to target user, execute corresponding rescue and appoint
Business, wherein rescue task includes implementing external defibrillation.
Further, the S300 further include: acquisition target user's physical condition information, including but not limited to heart rate;Root
Entry evaluation is done to the physical condition of target user according to the information of acquisition, and shows assessment result.
Further, the S300 further include: S310, face image data of the acquisition to defibrillation object, estimate and
Assessment in detail;S320, the skin pattern feature for extracting face in face image data carry out entry evaluation to the range of age,
Obtain a specified age bracket;S330, the method based on support vector machines establish the category of model for corresponding to different age group
Device, wherein the method for establishing category of model type is based on histogram of gradients and local binarization Model Establishment training sample database;
S340, the age bracket determined according to step S302, choose the model classifiers of corresponding age bracket, with acquisition to defibrillation object
Face image data is assessed as input terminal, exports age estimated result;S350, preset age and corresponding points are based on
Energy rule is hit, corresponding click energy is selected according to age estimated result.
Further, described 310 further include face's pre-treatment step: A. face extracts, and is carried out based on face detection device CAS
Face extracts;B. five characteristic points of face are detected using Coarse-to-Fine Auto-Encoder Network (CFAN),
It is right and left eyes center, nose, mouth left and right corner respectively;C. feature rear face image data will be extracted to be standardized, and will
Image progress picture size after standardization, which is reseted, sets, wherein standardization includes internal standardization and outer standardization two ways,
Standardized mode not only contains the intrinsic information of face at home and abroad, but also also contains whole contextual information, internal standard
Change contains only facial information.
Present artificial intelligence has obtained further extensive attention in computer field.And in robot, economic politics is determined
Plan, control system are used widely in analogue system.AED first aid auxiliary intelligent robot applies the artificial of pinpoint accuracy
Intelligence system is used to handle the data that rescue patient is and calculates, and sets in the AED defibrillator of cooperation product carrying and other rescues
It is standby, substantially meet the needs of product.Finally, the design of product mainly includes both sides content: artificial intelligence design and relief
The design of equipment, in the research research range of this paper, the design of the artificial intelligence mainly various relief by comparison on the market
The technological parameter of equipment, the final specific data of typing, determines the final design of product.
Auxiliary robot mainly needs to have following factor:
(1) safety in intelligent first aid procedures, because there are the AED defibrillators of high current in salvage device, once go out
Now electric leakage or error shock pole will will cause serious consequence, thus safety analysis become in product salvage device it is primary because
Element.(2) the operation conveniency of salvage device is bound to be applied to the salvage device in product, salvage device during relief
For medical professional equipment, the ordinary people without learning training is difficult directly to apply such equipment, therefore products application high precision
Artificial intelligence system instructs ordinary people to carry out rescue service by accurately calculating by voice and display screen.It is difficult to reduce operation
Degree.(3) relief efficiency is improved, in order to improve the success rate of first aid, salvage device is equipped with AED defibrillator and first aid medicine, object
Product first-aid kit completes first aid within the shortest time by the calculating of artificial intelligence, catches the prime time of first aid.
In the modern times of development in science and technology, pressure is also increasing, and also along with numerous sudden illness, sudden cardiac death is roared
Asthma, heart disease come one after another, and sudden cardiac death is most representative, and sudden cardiac death occurs in 1 hour after being broken out with acute symptom
Characterized by consciousness is lost suddenly, sudden cardiac arrest patient early stage 85%~90% is ventricular fibrillation, treats ventricular fibrillation most efficient method
It is early AED defibrillation.Defibrillation is every to postpone 1 minute, and survival rate reduces by 7%~10%.The early stage of CPR and AED is effectively used cooperatively,
It is the most effective rescue means for rescuing cardiopulmonary arrest sudden death patient.AED defibrillator is small in size etc. excellent because having high efficiency
Point is applied to this product.The AED defibrillator of application is designed herein, as long as structure is as shown, the system includes following composition
Component: power supply, defibrillation electrode sheet, electrode slice are placed indicator light, electrode slice connector interface, diagnosis panel, loudspeaker, electric shock and are pressed
Button, adult-children's switching push button and accessory part.
The working principle of AED: arrhythmia cordis is eliminated by heart with stronger pulse current, is allowed to restore the normal heart
Rule, defibrillation is to treat arrhythmia cordis using exogenous electric current, is the method for modern age treatment arrhythmia cordis.Defibrillation
Principle is: cardioactive when defibrillation conversion is primary instantaneous high energy pulse, and general persistence is 4~10ms, and electric energy exists
In 40~400J (joule).It can complete Electrical Cardioversion, i.e. defibrillation.When serious tachy-arrhythmia occurs for patient, such as atrium
It flutters, atrial fibrillation, supraventricular or Ventricular Tachycardia etc., often results in different degrees of hemodynamics obstacle.Especially work as trouble
When ventricular fibrillation occurs in person, since ventricle is terminated without overall shrinkage ability, cardiac ejection and blood circulation, such as rescue not in time,
Patient is often resulted in because brain hypoxic exposure is too long and dead.Defibrillator is such as used, the electric current for controlling certain energy passes through heart, energy
Certain arrhythmias are eliminated, the rhythm of the heart can be made to restore normal, so that heart disease patients be made to be rescued and be treated.Artificial intelligence system
System first can intelligent identifying system can be judged automatically according to the heart rate of patient, given in the case where permission electric shock removal
It quivers, front of the car has Intelligent heart rate display instrument, more accurately can provide processing information and application method, system meeting for rescuer
The size of current of defibrillation and the time of electric shock are controlled, thoroughly solves the problems, such as that ordinary people is not available professional medical equipment.After and
Vehicle body is equipped with other first-aid equipments, is mainly used for needed for other first aid occasions in addition to AED first aid, as various types are detoxified
Medicament, hemostasis medical instrument etc., after actuation when vehicle first aid, rear engine cover rotation bounces and can show intelligent tutoring system, according to applying
The information analysis first aid grade of the person's of rescuing input, corresponding pop-up drug, and the display screen of rear engine cover attachment will pop up doctor
It treats instrument study course guidance auxiliary rescuer and rescues patient.AED first aid auxiliary intelligent robot first aid rushes to the scene under form, front truck
Body braking, rear back-drive, and vehicle body is recessed, and the site of the accident is hurried in the form of the driving of automobile, wakes up artificial intelligence in the process
Energy system, every first-aid apparatus is in running order, opens simultaneously alarm, and infrared detector, sounding an alarm avoids pedestrian,
Infra-red detection pedestrian makes robot avoid the pedestrian that passes by one's way in time.It finally arrives at the site of the accident to start operation, succours patient.
Age level estimation, which is roughly divided into, estimates and assesses in detail two stages.It estimates the stage: extracting face in photo
Skin pattern feature does a rough assessment to the range of age, obtains a specific age bracket;
Detailed evaluation stage:
By the method for support vector machines, multiple model classifiers corresponding to multiple age brackets are established, and select to close
Suitable model is matched.Among these, we merge the face age algorithm for estimating of LBP and HOG feature.
Using the face age algorithm for estimating of LBP and HOG feature process as shown in figure 3,
The face age algorithm for estimating for merging LBP and HOG feature extracts and the part of the face of change of age close relation
Statistical nature.LBP (local binarization mode) feature and HOG (histogram of gradients) feature, and with CCA's (canonical correlation analysis)
Method fusion, is trained and tests to face database finally by the method for SVR (Support vector regression).
Carrying out age estimation using face includes two committed steps:
(1) age characteristics is expressed: all there is general visual problem in this, and the method used before is all using artificial
The feature extractor of setting carries out feature extraction, but this mode is largely determined by the characteristics of feature extractor
It is fixed, it tends in the case where emphasis is looked after in a certain respect, performance is not fine in other aspects.This kind of feature extractor
Mainly there is (BIF, LBP, HOG etc.).And can enable the network to widely learn various features using the method for deep learning,
The feature that cost of labor can not only be reduced, but also extracted also compares with versatility.
(2) learning process of age estimator: the estimation at age is divided into recurrence, classification, classification+recurrence the problems such as progress
Study, but the previous method extracted using manual features extractor, feature extraction is to separate with age estimation procedure,
Neural network or SVR etc. namely are fed for later first with the progress feature extraction of some feature extractors to be estimated, are done so
Bad is some the selection of final accuracy dependence characteristics, and the study toward device of backward classifying/return can not feed back to spy
Extractor is levied, so that feature extraction cannot be adjusted according to the accuracy of estimation.This point be similar to target detection in RCNN and
The training method of Fast RCNN, it is not only big in feature extraction phases operand, but also it is easy to produce redundancy, and subsequent place
Reason cannot be fed back to front, so that effect is general.And process end to end then may be implemented by using deep neural network, from
And promote the accuracy of estimation.
The conventional method of human face training and test is described below:
The training method of network:
(1) age returns device training:
The method of recurrence is employed herein, the last layer of primitive network is revised as a neurode first, and
Using sigmoid activation primitive, result can be limited between [0,1] in this way.It is so also required to normalize on age label
So that their scales having the same, not so loss function will be unable to calculate between [0,1].The normalized method of label be by
Institute's has age is all divided by 100.
(2) character classification by age device training:
Scale label are described using the age, cross entropy is employed herein and is trained as loss function.
The step of network training:
(1) since the data set of face is all small, first large-scale facial recognition data collection CASIA-WebFace into
Pre-training is gone.It is primarily due to recognition of face and face characteristic has all been used in the estimation of face age, therefore in recognition of face number
The feature of face can be preferably extracted according to pre-training is carried out on collection, this initialization mode certainly will be more preferable than random initializtion,
Serious over-fitting is less likely to occur.
(2) it is finely adjusted using the data set comprising real age.Although real age estimation and surface age estimation are not
Equally, but still there are many similitudes in the two.It is finely tuned on CACD, Morph-II, WebFaceAge.It mentions above
To classifier and return device be the study carried out herein.
(3) it is finely adjusted using competition data.
Face's pretreatment:
(1) face extracts: having carried out facial extraction using the face detection device CAS that the laboratory VIPL is researched and developed.
(2) it face feature point location: is had detected using Coarse-to-Fine Auto-Encoder Network (CFAN)
Five characteristic points of face are right and left eyes center, nose, mouth left and right corner respectively.
(3) face standardization: being used herein internal standardization and outer standardization two ways, wherein outer standardized mode
The intrinsic information of face is not only contained, but also also contains whole contextual information;Internal standardization contains only facial letter
Breath.Then the image after standardization the resize of 256*256 has been subjected to.
It should be appreciated that the embodiment of the present invention can be by computer hardware, the combination of hardware and software or by depositing
The computer instruction in non-transitory computer-readable memory is stored up to be effected or carried out.Standard volume can be used in the method
Journey technology-includes that the non-transitory computer-readable storage media configured with computer program is realized in computer program,
In configured in this way storage medium computer is operated in a manner of specific and is predefined --- according in a particular embodiment
The method and attached drawing of description.Each program can with the programming language of level process or object-oriented come realize with department of computer science
System communication.However, if desired, the program can be realized with compilation or machine language.Under any circumstance, which can be volume
The language translated or explained.In addition, the program can be run on the specific integrated circuit of programming for this purpose.
In addition, the operation of process described herein can be performed in any suitable order, unless herein in addition instruction or
Otherwise significantly with contradicted by context.Process described herein (or modification and/or combination thereof) can be held being configured with
It executes, and is can be used as jointly on the one or more processors under the control of one or more computer systems of row instruction
The code (for example, executable instruction, one or more computer program or one or more application) of execution, by hardware or its group
It closes to realize.The computer program includes the multiple instruction that can be performed by one or more processors.
Further, the method can be realized in being operably coupled to suitable any kind of computing platform, wrap
Include but be not limited to PC, mini-computer, main frame, work station, network or distributed computing environment, individual or integrated
Computer platform or communicated with charged particle tool or other imaging devices etc..Each aspect of the present invention can be to deposit
The machine readable code on non-transitory storage medium or equipment is stored up to realize no matter be moveable or be integrated to calculating
Platform, such as hard disk, optical reading and/or write-in storage medium, RAM, ROM, so that it can be read by programmable calculator, when
Storage medium or equipment can be used for configuration and operation computer to execute process described herein when being read by computer.This
Outside, machine readable code, or part thereof can be transmitted by wired or wireless network.When such media include combining microprocessor
Or other data processors realize steps described above instruction or program when, invention as described herein including these and other not
The non-transitory computer-readable storage media of same type.When methods and techniques according to the present invention programming, the present invention
It further include computer itself.
Computer program can be applied to input data to execute function as described herein, to convert input data with life
At storing to the output data of nonvolatile memory.Output information can also be applied to one or more output equipments as shown
Device.In the preferred embodiment of the invention, the data of conversion indicate physics and tangible object, including the object generated on display
Reason and the particular visual of physical objects are described.
The above, only presently preferred embodiments of the present invention, the invention is not limited to above embodiment, as long as
It reaches technical effect of the invention with identical means, all within the spirits and principles of the present invention, any modification for being made,
Equivalent replacement, improvement etc., should be included within the scope of the present invention.Its technical solution within the scope of the present invention
And/or embodiment can have a variety of different modifications and variations.