CN109447176A - Bursting work personnel detection method, device, system, medium and server - Google Patents
Bursting work personnel detection method, device, system, medium and server Download PDFInfo
- Publication number
- CN109447176A CN109447176A CN201811332421.9A CN201811332421A CN109447176A CN 109447176 A CN109447176 A CN 109447176A CN 201811332421 A CN201811332421 A CN 201811332421A CN 109447176 A CN109447176 A CN 109447176A
- Authority
- CN
- China
- Prior art keywords
- mask
- cnn
- bursting work
- personnel
- training
- 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.)
- Pending
Links
- 230000009172 bursting Effects 0.000 title claims abstract description 137
- 238000001514 detection method Methods 0.000 title claims abstract description 58
- 238000012549 training Methods 0.000 claims abstract description 60
- 238000005422 blasting Methods 0.000 claims abstract description 23
- 238000007689 inspection Methods 0.000 claims abstract description 19
- 239000002360 explosive Substances 0.000 claims description 29
- 238000004880 explosion Methods 0.000 claims description 28
- 239000000463 material Substances 0.000 claims description 28
- 238000003860 storage Methods 0.000 claims description 22
- 239000000843 powder Substances 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 10
- 239000004744 fabric Substances 0.000 claims description 3
- 238000000034 method Methods 0.000 abstract description 33
- 238000010276 construction Methods 0.000 abstract description 3
- 238000012790 confirmation Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000000926 separation method Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- 238000004140 cleaning Methods 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005474 detonation Methods 0.000 description 1
- 239000003814 drug Substances 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000000737 periodic effect Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
- G06Q50/265—Personal security, identity or safety
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Tourism & Hospitality (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Educational Administration (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Economics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Computer Security & Cryptography (AREA)
- General Business, Economics & Management (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of bursting work personnel detection method, device, system, medium and a kind of server based on Mask R-CNN, method include: to obtain the default collected realtime graphic of pickup area;The realtime graphic is detected using Mask R-CNN model after training, obtains the essential information of the default pickup area borehole blasting operating personnel;Wherein, the essential information is for judging whether the configuration of the bursting work personnel meets default bursting work specification;Mask R-CNN model is the model being trained using the history picture sample comprising bursting work personnel's essential information to the blank model constructed based on Mask R-CNN algorithm after the training.From the above, it can be seen that, the present invention detects the collected realtime graphic of default pickup area using Mask R-CNN model after training, it avoids in manual inspection method since wrong leakage detection caused by administrative staff's work carelessness interrogates topic, the safety of bursting work construction working is ensured, and human resources are further saved, improve detection efficiency.
Description
Technical field
The present invention relates to technical field of image processing, in particular to a kind of bursting work personnel inspection based on Mask R-CNN
Survey method, apparatus, system, medium and server.
Background technique
It is well known that explosion project need to strictly observe the relevant regulations of public security bureau's explosion project security management, and in public security
Put on record at office, explosion project security management provides that each bursting work project management department should be according to bursting work project scene and setting
Quantity be equipped with enough bursting work personnel, wherein security official's quantity that each bursting work project management department is equipped with should meet " every
The standard of 1 security official of a bursting work project outfit ", is responsible for that project management responsible person, person in charge of the technology aspect is assisted to carry out relatively
The field management of some fixed bursting work project works;Administrator's quantity that each bursting work project management department is equipped with should expire
Relatively-stationary some bursting work scene is responsible in the standard of foot " each bursting work project is equipped with 2 administrators ", sole duty
The work such as keeping, nurse, transmitting-receiving, registration, the cleaning cancelling stocks of explosive;Dynamiter's number that each bursting work project management department is equipped with
Amount should meet the standard of " each bursting work project is equipped with 2 dynamiters ", and sole duty is responsible for some relatively-stationary explosion and is made
Cloth hole in industry, verify, powder charge, on line, protection, detonation, warning, it is quick-fried after the operation such as check.
Require to be responsible for operating personnel's color separation dressing in the confirmation of blast working specification according to related request progress dressing.Wherein bear
The blasting engineering technical staff of duty field operation unifies purple clothes, Dai Hongse safety cap, other bursting works personnel must not
Wear red safety cap;Dynamiter, security official distinguish yellow, red clothes;Period warehouseman on duty should unify blue
Clothes.Non- color separation dressing or the operating personnel for not wearing " bursting work personnel job qualification certificate " must not enter civil explosive material loading place
Domain, the interim storage area of civil explosive material and powder charge warning region.Before bursting work starts, work should be verified at the scene by public security bureau
Industry number can start to work when events in operation scene is equipped with enough bursting work personnel.
Instantly it is applied in the blast working specification confirmation of industrial circle for operating personnel's number and work post dressing really
Recognizing inspection is usually related management personnel manual inspection.Manual inspection method inevitably exists because administrative staff's work carelessness due to causes
Under-enumeration mistake the problem of looking into, the safety that may cause blast working work cannot get basic guarantee.In addition, multiple quick-fried when having
When broken operation field needs to supervise, public security bureau sends the quantity of supervisor also to increase accordingly, and causes police strength to waste, low efficiency
Under.
Summary of the invention
In view of this, the purpose of the present invention is to provide a kind of bursting work personnel detection side based on Mask R-CNN
Method replaces manual inspection method using COMPUTER DETECTION, improves detection efficiency, and can reduce manual inspection bring work
Make error.Its concrete scheme is as follows:
In a first aspect, the invention discloses a kind of bursting work personnel's detection method based on Mask R-CNN, comprising:
Obtain the default collected realtime graphic of pickup area;
The realtime graphic is detected using Mask R-CNN model after training, is obtained in the default pickup area
The essential information of bursting work personnel;Wherein, the essential information is for judging whether the configuration of the bursting work personnel accords with
Close default bursting work specification;Mask R-CNN model is to utilize going through comprising bursting work personnel's essential information after the training
The model that history picture sample is trained the blank model constructed based on Mask R-CNN algorithm.
Optionally, the default pickup area includes that explosion waiting area, civil explosive material loading and unloading area, civil explosive material are interim
Storage area and powder charge warning region.
Optionally, described that the realtime graphic is detected using Mask R-CNN model after training, it obtains described pre-
If the essential information of pickup area borehole blasting operating personnel, comprising:
It is detected, is obtained described using realtime graphic of the Mask R-CNN model after training to the explosion waiting area
The quantity and corresponding work post information of explosion waiting area borehole blasting operating personnel.
Optionally, described that the realtime graphic is detected using Mask R-CNN model after training, it obtains described pre-
If after the essential information of pickup area borehole blasting operating personnel, further includes:
The quantity and corresponding work post information are sent to supervision side, so that supervision side judges current bursting work personnel
Configuration whether meet default bursting work specification.
Optionally, described that the realtime graphic is detected using Mask R-CNN model after training, it obtains described pre-
If the essential information of pickup area borehole blasting operating personnel, comprising:
Using Mask R-CNN model after training to the civil explosive material loading and unloading area, the interim storage area of the civil explosive material
The realtime graphic of domain and the powder charge warning region is detected, and the work post information of corresponding bursting work personnel is obtained.
Optionally, described that the realtime graphic is detected using Mask R-CNN model after training, it obtains described pre-
If after the essential information of pickup area borehole blasting operating personnel, further includes:
The work post information of the bursting work personnel is sent to supervision side, so that supervision side judges the bursting work people
Whether the work post information of member meets corresponding working region.
Optionally, further includes:
The judging result of the supervision side is sent to the display screen of the corresponding default pickup area.
Optionally, further includes:
The result that Mask R-CNN model inspection obtains after the training is saved into distributed data base, to utilize institute
Distributed data base is stated to be updated Mask R-CNN model after the training.
Second aspect, bursting work personnel's detection device based on Mask R-CNN that the invention discloses a kind of, comprising:
Image collection module, for obtaining the collected realtime graphic of default pickup area;
Image detection module is obtained for being detected using Mask R-CNN model after training to the realtime graphic
The essential information of the default pickup area borehole blasting operating personnel;Wherein, the essential information is for judging that the explosion is made
Whether the configuration of industry personnel meets default bursting work specification;Mask R-CNN model is to make using comprising explosion after the training
The history picture sample of industry personnel's essential information is trained the blank model constructed based on Mask R-CNN algorithm
Model.
The third aspect, the invention discloses a kind of servers, comprising:
Memory, for storing computer program;
Processor, the aforementioned disclosed bursting work personnel based on Mask R-CNN when for executing the computer program
The step of detection method.
Fourth aspect, bursting work personnel's detection system based on Mask R-CNN that the invention discloses a kind of, including it is preceding
State disclosed server.
5th aspect, the invention discloses a kind of computer readable storage medium, the computer readable storage medium is used
In storage computer program, the computer program is realized aforementioned disclosed based on the quick-fried of Mask R-CNN when being executed by processor
The step of broken operating personnel's detection method.
As it can be seen that the present invention obtains the default collected realtime graphic of pickup area first;Utilize Mask R-CNN after training
Model detects the realtime graphic, obtains the essential information of the default pickup area borehole blasting operating personnel;Wherein,
The essential information is for judging whether the configuration of the bursting work personnel meets default bursting work specification;After the training
Mask R-CNN model is to be calculated using the history picture sample comprising bursting work personnel's essential information based on Mask R-CNN
The model that the blank model of method building is trained.From the foregoing, it will be observed that the present invention utilizes Mask R-CNN model pair after training
The default collected realtime graphic of pickup area is detected, and is avoided in manual inspection method and is neglected since administrative staff work
Caused by wrong leakage detection interrogate topic, ensured the safety of bursting work construction working, and further save human resources, mentioned
High detection efficiency.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
The embodiment of invention for those of ordinary skill in the art without creative efforts, can also basis
The attached drawing of offer obtains other attached drawings.
Fig. 1 is a kind of bursting work personnel's detection method flow chart based on Mask R-CNN disclosed in the present application;
Fig. 2 is a kind of specifically bursting work personnel's detection method process based on Mask R-CNN disclosed in the present application
Figure;
Fig. 3 is a kind of structural block diagram of bursting work personnel's detection device based on Mask R-CNN disclosed in the present application;
Fig. 4 is a kind of structural schematic diagram of server disclosed in the present application.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
In the prior art, in the blast working specification confirmation of industrial circle for operating personnel's number and work post dressing really
Recognizing inspection is usually related management personnel manual inspection.Manual inspection method inevitably exists because administrative staff's work carelessness due to causes
Under-enumeration mistake the problem of looking into;When there is multiple bursting work scenes to need to supervise, public security bureau sends the quantity of supervisor also phase
It should increase, police strength is caused to waste, inefficiency.
Bursting work personnel's detection method based on Mask R-CNN that the embodiment of the invention discloses a kind of, referring to Fig. 1 institute
Show, this method comprises:
Step S101: the default collected realtime graphic of pickup area is obtained;
In the present embodiment, the default collected realtime graphic of pickup area is obtained first.Specifically, the present embodiment is default
Pickup area installs camera in advance, collects the realtime graphic in current region using camera, and send it to cloud
Server.
Step S102: detecting the realtime graphic using Mask R-CNN model after training, obtains described default
The essential information of pickup area borehole blasting operating personnel;Wherein, the essential information is for judging the bursting work personnel's
Whether configuration meets default bursting work specification;Mask R-CNN model is to utilize to include bursting work personnel's base after the training
The model that the history picture sample of this information is trained the blank model constructed based on Mask R-CNN algorithm.
Specifically, after the present embodiment gets realtime graphic, the Mask R-CNN model obtained using preparatory training is to institute
It states realtime graphic to be detected, identifies the bursting work personnel in the realtime graphic in the default pickup area, further
Obtain the essential information of the bursting work personnel.
It should be noted that the Mask R-CNN model is to utilize the history figure comprising bursting work personnel's essential information
The model that piece sample is trained the blank model constructed based on Mask R-CNN algorithm.Wherein, the bursting work
Personnel's essential information include bursting work personnel dressing color, whether safe wearing cap and corresponding work post information.
It is understood that after getting the essential information of the default pickup area borehole blasting operating personnel, this reality
Applying example can judge whether the configuration of current bursting work personnel meets default bursting work specification using essential information, realize quick-fried
Broken operation starts the preceding detection to bursting work personnel.
As it can be seen that the present invention obtains the default collected realtime graphic of pickup area first;Utilize Mask R-CNN after training
Model detects the realtime graphic, obtains the essential information of the default pickup area borehole blasting operating personnel;Wherein,
The essential information is for judging whether the configuration of the bursting work personnel meets default bursting work specification;After the training
Mask R-CNN model is to be calculated using the history picture sample comprising bursting work personnel's essential information based on Mask R-CNN
The model that the blank model of method building is trained.From the foregoing, it will be observed that the present invention utilizes Mask R-CNN model pair after training
The default collected realtime graphic of pickup area is detected, and is avoided in manual inspection method and is neglected since administrative staff work
Caused by wrong leakage detection interrogate topic, ensured the safety of bursting work construction working, and further save human resources, mentioned
High detection efficiency.
It is shown in Figure 2, the embodiment of the invention discloses a kind of bursting work personnel detection side based on Mask R-CNN
Method, comprising:
Step S201: the default collected realtime graphic of pickup area is obtained;Wherein, the default pickup area includes quick-fried
Broken waiting area, civil explosive material loading and unloading area, the interim storage area of civil explosive material and powder charge warning region;
In the present embodiment, explosion waiting area, civil explosive material loading and unloading area, the interim storage area of civil explosive material and dress are obtained
The collected realtime graphic of camera of medicine warning region installation.
Step S202: it is examined using realtime graphic of the Mask R-CNN model after training to the explosion waiting area
It surveys, obtains the quantity and corresponding work post information of the explosion waiting area borehole blasting operating personnel;
It is understood that the bursting work personnel of all work posts wait to be waited in explosion before bursting work starts
Region is to be checked.After the present embodiment gets the realtime graphic in the explosion waiting area, Mask R-CNN mould after training is utilized
Type identifies realtime graphic, acquires in the explosion waiting area quantity of current bursting work personnel and corresponding
Dressing color, safety cap wearing information further obtain the corresponding work post information of the bursting work personnel.
In the present embodiment, in the quantity and corresponding work post information for obtaining the explosion waiting area borehole blasting operating personnel
Later, the quantity and corresponding work post information are sent to supervision side, so that supervision side judges current bursting work personnel's
Whether configuration meets default bursting work specification.For example, if providing that each bursting work project is matched in default bursting work specification
Standby 2 dynamiters, and regulation dynamiter wears yellow clothes, after being detected to current acquired image, if recognizing certain
A bursting work project is only equipped with 1 dynamiter, or current dynamiter does not wear yellow clothes, then determines current bursting work
The configuration of personnel does not meet default bursting work specification.
Step S203: using Mask R-CNN model after training to the civil explosive material loading and unloading area, the civil explosive material
The realtime graphic of interim storage area and the powder charge warning region is detected, and the work post letter of corresponding bursting work personnel is obtained
Breath.
In the present embodiment, bursting work personnel enter the civil explosive material loading and unloading area, the civil explosive material is temporarily stored
Behind region and the powder charge warning region, using Mask R-CNN model after training to the civil explosive material loading and unloading area, described
The realtime graphic of the interim storage area of civil explosive material and the powder charge warning region is identified that detection obtains current each region
The dressing color of interior explosion operating personnel, safety cap wear condition, further obtain the work post of current region bursting work personnel
Information.
It is understood that after getting the work post information of the bursting work personnel, the present embodiment is by the explosion
The work post information of operating personnel is sent to supervision side, so that supervision side judges whether the work post information of the bursting work personnel accords with
Close corresponding working region.For example, using Mask R-CNN model after training to the reality of the interim storage area of the civil explosive material
When image identified, according to corresponding guarantor should be equipped in the interim storage area of civil explosive material described in default bursting work specification
Guan Yuan if regulation administrator wears blue clothes, and currently recognizes the administrator in the interim storage area of the civil explosive material
Blue clothes are not worn, then can determine that the work post information of current bursting work personnel does not meet corresponding working region.
In the present embodiment, using Mask R-CNN model after training to explosion waiting area, civil explosive material loading and unloading area, the people
The realtime graphic of the interim storage area of quick-fried equipment and powder charge warning region is detected, and be will test result and be sent to supervision side,
So that supervision side determines whether to meet default bursting work specification, the detection to bursting work personnel is realized.
One kind of bursting work personnel's detection method based on Mask R-CNN provided by the embodiment of the present invention is specific
In embodiment, this method further include:
The judging result of the supervision side is sent to the display screen of the corresponding default pickup area.
It is understood that whether being met pre- after supervision side judges the configuration of current bursting work personnel
If bursting work specification as a result, by corresponding judging result it is corresponding be sent to explosion waiting area, civil explosive material loading and unloading area,
Display screen in the interim storage area of civil explosive material and powder charge warning region, it is quick-fried to prompt bursting work personnel whether can start
Broken operation.
In addition, in addition to the mode for the display screen that judging result is sent to default pickup area disclosed in above-described embodiment,
Judging result can also be carried out voice broadcast by the present invention, to achieve the purpose that prompt bursting work personnel.It is, of course, also possible to adopt
In other ways, herein the present invention without limitation.
One kind of bursting work personnel's detection method based on Mask R-CNN provided by the embodiment of the present invention is specific
In embodiment, the training process of the Mask R-CNN model is further elaborated, process includes:
Before the Mask R-CNN model training, Faster is set by the hyper parameter of the Mask R-CNN model
The parameter value of R-CNN model, and pre-training is carried out to the hyper parameter using ResNet50, ResNet101, FPN network;Into one
Step is trained the Mask R-CNN model using the history picture sample largely comprising bursting work personnel essential information,
Obtain Mask R-CNN model.
In the present embodiment, after training obtains Mask R-CNN model, using the test sample of preset quantity to described
Mask R-CNN model is tested, to verify the accuracy of the Mask R-CNN model.
One kind of bursting work personnel's detection method based on Mask R-CNN provided by the embodiment of the present invention is specific
In embodiment, this method further include:
The result that Mask R-CNN model inspection obtains after the training is saved into distributed data base, to utilize institute
Distributed data base is stated to be updated Mask R-CNN model after the training.
It is understood that the present embodiment by the result that Mask R-CNN model inspection obtains after the training save to point
In cloth database, it can use the distributed data base and Mask R-CNN model after the training regularly updated,
So that the accuracy of Mask R-CNN model is higher and higher after the training.
Further, the system of rewards and penalties can also be arranged in the present embodiment, and the inspection in distributed data base described in periodic analysis
It surveys as a result, punishing the bursting work personnel for violating the default bursting work specification, to realize to bursting work people
The supervision role of member.
Bursting work personnel's detection device to provided in an embodiment of the present invention based on Mask R-CNN is introduced below,
It is described below based on the bursting work personnel detection device of Mask R-CNN with above-described based on the quick-fried of Mask R-CNN
Broken operating personnel's detection method can correspond to each other reference.
Fig. 3 is the structural frames of bursting work personnel's detection device based on Mask R-CNN provided by the embodiment of the present invention
Figure, referring to shown in Fig. 3, bursting work personnel's detection device based on Mask R-CNN includes:
Image collection module 100, for obtaining the collected realtime graphic of default pickup area;
Image detection module 200 is obtained for being detected using Mask R-CNN model after training to the realtime graphic
To the essential information of the default pickup area borehole blasting operating personnel;Wherein, the essential information is for judging the explosion
Whether the configuration of operating personnel meets default bursting work specification;Mask R-CNN model is to utilize to include explosion after the training
The history picture sample of operating personnel's essential information is trained to obtain to the blank model constructed based on Mask R-CNN algorithm
Model.
The present embodiment is based on Mask based on the bursting work personnel detection device of Mask R-CNN for realizing above-mentioned
Bursting work personnel's detection method of R-CNN, thus it is specific in bursting work personnel's detection device based on Mask R-CNN
The implementation section of the visible hereinbefore bursting work personnel's detection method based on Mask R-CNN of implementation method, no longer carries out herein
It repeats.
Further, the embodiment of the invention also discloses a kind of servers, shown in Figure 4, and the server includes depositing
Reservoir 11 and processor 12, wherein realized when the processor 12 executes the computer program saved in the memory 11 with
Lower step:
Obtain the default collected realtime graphic of pickup area;
The realtime graphic is detected using Mask R-CNN model after training, is obtained in the default pickup area
The essential information of bursting work personnel;Wherein, the essential information is for judging whether the configuration of the bursting work personnel accords with
Close default bursting work specification;Mask R-CNN model is to utilize going through comprising bursting work personnel's essential information after the training
The model that history picture sample is trained the blank model constructed based on Mask R-CNN algorithm.
In the present embodiment, when the processor 12 executes the computer subprogram saved in the memory 11, can have
Body is performed the steps of to be detected using realtime graphic of the Mask R-CNN model after training to the explosion waiting area,
Obtain the quantity and corresponding work post information of the explosion waiting area borehole blasting operating personnel.
In the present embodiment, when the processor 12 executes the computer subprogram saved in the memory 11, can have
Body, which is performed the steps of, is sent to supervision side for the quantity and corresponding work post information, so that supervision side judges current explosion
Whether the configuration of operating personnel meets default bursting work specification.
In the present embodiment, when the processor 12 executes the computer subprogram saved in the memory 11, can have
Body, which is performed the steps of, faces the civil explosive material loading and unloading area, the civil explosive material using Mask R-CNN model after training
When storage area and the realtime graphic of the powder charge warning region detected, obtain the work post letter of corresponding bursting work personnel
Breath.
In the present embodiment, when the processor 12 executes the computer subprogram saved in the memory 11, can have
Body, which is performed the steps of, is sent to supervision side for the work post information of the bursting work personnel so that supervision side judge it is described quick-fried
Whether the work post information of broken operating personnel meets corresponding working region.
In the present embodiment, when the processor 12 executes the computer subprogram saved in the memory 11, can have
Body performs the steps of the display screen that the judging result of the supervision side is sent to the corresponding default pickup area.
In the present embodiment, when the processor 12 executes the computer subprogram saved in the memory 11, can have
Body, which performs the steps of, saves the result that Mask R-CNN model inspection obtains after the training into distributed data base,
To be updated using the distributed data base to Mask R-CNN model after the training.
Further, bursting work personnel's detection system based on Mask R-CNN that the embodiment of the invention also discloses a kind of
System, wherein the detection system includes aforementioned disclosed server.
Further, the embodiment of the invention also discloses a kind of computer readable storage mediums, for storing computer journey
Sequence, wherein the computer program realizes the aforementioned disclosed bursting work people based on Mask R-CNN when being executed by processor
Member's detection method, about this method specific steps can with reference to corresponding contents disclosed in previous embodiment, herein no longer into
Row repeats.
The present invention detects the collected realtime graphic of default pickup area using Mask R-CNN model after training,
It avoids wrong leakage detection caused by neglecting in manual inspection method due to administrative staff's work and interrogates topic, ensured that bursting work is applied
The safety of work work, and human resources are further saved, improve detection efficiency.
Each embodiment in this specification is described in a progressive manner, the highlights of each of the examples are with it is other
The difference of embodiment, same or similar part may refer to each other between each embodiment.For being filled disclosed in embodiment
For setting, since it is corresponded to the methods disclosed in the examples, so being described relatively simple, related place is referring to method part
Explanation.
Professional further appreciates that, unit described in conjunction with the examples disclosed in the embodiments of the present disclosure
And algorithm steps, can be realized with electronic hardware, computer software, or a combination of the two, in order to clearly demonstrate hardware and
The interchangeability of software generally describes each exemplary composition and step according to function in the above description.These
Function is implemented in hardware or software actually, the specific application and design constraint depending on technical solution.Profession
Technical staff can use different methods to achieve the described function each specific application, but this realization is not answered
Think beyond the scope of this invention.
The step of method described in conjunction with the examples disclosed in this document or algorithm, can directly be held with hardware, processor
The combination of capable software module or the two is implemented.Software module can be placed in random access memory (RAM), memory, read-only deposit
Reservoir (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technology
In any other form of storage medium well known in field.
Finally, it is to be noted that, herein, relational terms such as first and second and the like be used merely to by
One entity or operation are distinguished with another entity or operation, without necessarily requiring or implying these entities or operation
Between there are any actual relationship or orders.Moreover, the terms "include", "comprise" or its any other variant meaning
Covering non-exclusive inclusion, so that the process, method, article or equipment for including a series of elements not only includes that
A little elements, but also including other elements that are not explicitly listed, or further include for this process, method, article or
The intrinsic element of equipment.In the absence of more restrictions, the element limited by sentence "including a ...", is not arranged
Except there is also other identical elements in the process, method, article or apparatus that includes the element.
Above to a kind of bursting work personnel detection method based on Mask R-CNN provided by the present invention, device, be
System, storage medium and a kind of server are described in detail, and specific case used herein is to the principle of the present invention and reality
The mode of applying is expounded, and the above description of the embodiment is only used to help understand the method for the present invention and its core ideas;Together
When, for those of ordinary skill in the art, according to the thought of the present invention, have in specific embodiments and applications
Change place, in conclusion the contents of this specification are not to be construed as limiting the invention.
Claims (12)
1. a kind of bursting work personnel's detection method based on Mask R-CNN characterized by comprising
Obtain the default collected realtime graphic of pickup area;
The realtime graphic is detected using Mask R-CNN model after training, obtains the default pickup area borehole blasting
The essential information of operating personnel;Wherein, the essential information is pre- for judging whether the configuration of the bursting work personnel meets
If bursting work specification;Mask R-CNN model is to utilize the history figure comprising bursting work personnel's essential information after the training
The model that piece sample is trained the blank model constructed based on Mask R-CNN algorithm.
2. bursting work personnel's detection method according to claim 1 based on Mask R-CNN, which is characterized in that described
Default pickup area includes explosion waiting area, civil explosive material loading and unloading area, the interim storage area of civil explosive material and powder charge warning
Region.
3. bursting work personnel's detection method according to claim 2 based on Mask R-CNN, which is characterized in that described
The realtime graphic is detected using Mask R-CNN model after training, obtains the default pickup area borehole blasting operation
The essential information of personnel, comprising:
It is detected using realtime graphic of the Mask R-CNN model after training to the explosion waiting area, obtains the explosion
The quantity of waiting area borehole blasting operating personnel and corresponding work post information.
4. bursting work personnel's detection method according to claim 3 based on Mask R-CNN, which is characterized in that described
The realtime graphic is detected using Mask R-CNN model after training, obtains the default pickup area borehole blasting operation
After the essential information of personnel, further includes:
The quantity and corresponding work post information are sent to supervision side, so that supervision side judges matching for current bursting work personnel
It sets and whether meets default bursting work specification.
5. bursting work personnel's detection method according to claim 2 based on Mask R-CNN, which is characterized in that described
The realtime graphic is detected using Mask R-CNN model after training, obtains the default pickup area borehole blasting operation
The essential information of personnel, comprising:
Using Mask R-CNN model after training to the civil explosive material loading and unloading area, the interim storage area of the civil explosive material and
The realtime graphic of the powder charge warning region is detected, and the work post information of corresponding bursting work personnel is obtained.
6. bursting work personnel's detection method according to claim 5 based on Mask R-CNN, which is characterized in that described
The realtime graphic is detected using Mask R-CNN model after training, obtains the default pickup area borehole blasting operation
After the essential information of personnel, further includes:
The work post information of the bursting work personnel is sent to supervision side, so that supervision side judges the bursting work personnel's
Whether work post information meets corresponding working region.
7. bursting work personnel's detection method according to claim 6 based on Mask R-CNN, which is characterized in that also wrap
It includes:
The judging result of the supervision side is sent to the display screen of the corresponding default pickup area.
8. bursting work personnel's detection method according to any one of claims 1 to 7 based on Mask R-CNN, feature
It is, further includes:
The result that Mask R-CNN model inspection obtains after the training is saved into distributed data base, to utilize described point
Cloth database is updated Mask R-CNN model after the training.
9. a kind of bursting work personnel's detection device based on Mask R-CNN characterized by comprising
Image collection module, for obtaining the collected realtime graphic of default pickup area;
Image detection module is obtained described for being detected using Mask R-CNN model after training to the realtime graphic
The essential information of default pickup area borehole blasting operating personnel;Wherein, the essential information is for judging the bursting work people
Whether the configuration of member meets default bursting work specification;Mask R-CNN model is to utilize to include bursting work people after the training
The model that the history picture sample of member's essential information is trained the blank model constructed based on Mask R-CNN algorithm.
10. a kind of server characterized by comprising
Memory, for storing computer program;
Processor is realized when for executing the computer program and is based on Mask R-CNN as described in any one of claim 1 to 8
Bursting work personnel's detection method the step of.
11. a kind of bursting work personnel's detection system based on Mask R-CNN, which is characterized in that including such as claim 10 institute
The server stated.
12. a kind of computer readable storage medium, which is characterized in that for storing computer program, the computer program quilt
Bursting work personnel's detection method as described in any one of claim 1 to 8 based on Mask R-CNN is realized when processor executes
The step of.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811332421.9A CN109447176A (en) | 2018-11-09 | 2018-11-09 | Bursting work personnel detection method, device, system, medium and server |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811332421.9A CN109447176A (en) | 2018-11-09 | 2018-11-09 | Bursting work personnel detection method, device, system, medium and server |
Publications (1)
Publication Number | Publication Date |
---|---|
CN109447176A true CN109447176A (en) | 2019-03-08 |
Family
ID=65551737
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811332421.9A Pending CN109447176A (en) | 2018-11-09 | 2018-11-09 | Bursting work personnel detection method, device, system, medium and server |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109447176A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110425008A (en) * | 2019-08-02 | 2019-11-08 | 精英数智科技股份有限公司 | A kind of pair of bursting work operates method, system and the storage medium to exercise supervision |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106302758A (en) * | 2016-08-19 | 2017-01-04 | 广东振声科技股份有限公司 | Supervision department's management informationalized to demolition site system and using method thereof |
CN106372662A (en) * | 2016-08-30 | 2017-02-01 | 腾讯科技(深圳)有限公司 | Helmet wearing detection method and device, camera, and server |
WO2017201676A1 (en) * | 2016-05-24 | 2017-11-30 | Intel Corporation | Self-adaptive window mechanism |
CN107563693A (en) * | 2017-08-31 | 2018-01-09 | 福建广汇信息技术有限公司 | Civilian explosive device uses closed loop Visualization Management Platform and method |
CN108038424A (en) * | 2017-11-27 | 2018-05-15 | 华中科技大学 | A kind of vision automated detection method suitable for working at height |
CN108052900A (en) * | 2017-12-12 | 2018-05-18 | 成都睿码科技有限责任公司 | A kind of method by monitor video automatic decision dressing specification |
CN108270999A (en) * | 2018-01-26 | 2018-07-10 | 中南大学 | A kind of object detection method, image recognition server and system |
CN108268832A (en) * | 2017-12-20 | 2018-07-10 | 广州供电局有限公司 | Electric operating monitoring method, device, storage medium and computer equipment |
-
2018
- 2018-11-09 CN CN201811332421.9A patent/CN109447176A/en active Pending
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2017201676A1 (en) * | 2016-05-24 | 2017-11-30 | Intel Corporation | Self-adaptive window mechanism |
CN106302758A (en) * | 2016-08-19 | 2017-01-04 | 广东振声科技股份有限公司 | Supervision department's management informationalized to demolition site system and using method thereof |
CN106372662A (en) * | 2016-08-30 | 2017-02-01 | 腾讯科技(深圳)有限公司 | Helmet wearing detection method and device, camera, and server |
CN107563693A (en) * | 2017-08-31 | 2018-01-09 | 福建广汇信息技术有限公司 | Civilian explosive device uses closed loop Visualization Management Platform and method |
CN108038424A (en) * | 2017-11-27 | 2018-05-15 | 华中科技大学 | A kind of vision automated detection method suitable for working at height |
CN108052900A (en) * | 2017-12-12 | 2018-05-18 | 成都睿码科技有限责任公司 | A kind of method by monitor video automatic decision dressing specification |
CN108268832A (en) * | 2017-12-20 | 2018-07-10 | 广州供电局有限公司 | Electric operating monitoring method, device, storage medium and computer equipment |
CN108270999A (en) * | 2018-01-26 | 2018-07-10 | 中南大学 | A kind of object detection method, image recognition server and system |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110425008A (en) * | 2019-08-02 | 2019-11-08 | 精英数智科技股份有限公司 | A kind of pair of bursting work operates method, system and the storage medium to exercise supervision |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US9471749B2 (en) | Healthcare verification system and method | |
US10692339B2 (en) | Personalized emergency evacuation plan | |
US20170004427A1 (en) | Selection of emergency responders | |
CN102622549B (en) | Electronic seal implementation system and method | |
CN104200159B (en) | Configure the method and device of the authority of application program | |
CN104040550A (en) | Integrating security policy and event management | |
US10043353B2 (en) | Wearer role-based visually modifiable garment | |
CN105893989A (en) | Dynamic calibration method for ultrasonic fingerprint identification, device and electronic device | |
CN112016520B (en) | Traffic violation credential generation method and device based on AI, terminal and storage medium | |
CN110146035A (en) | Built-in fitting detection method, device, equipment and the system of component production line | |
CN110443907A (en) | Patrol task processing method and patrol task processing server | |
CN105095753B (en) | Broadcast safe detection method, device | |
CN109447176A (en) | Bursting work personnel detection method, device, system, medium and server | |
CN107871344B (en) | Anti-substitute-printing attendance checking method for intelligent monitoring management | |
CN111639879B (en) | Intelligent security personnel information management method, device and system, storage medium and server | |
CN112884905A (en) | Virtual scene construction method and device, computer equipment and storage medium | |
CN114071060B (en) | Intelligent remote comment supervision system | |
KR20160046767A (en) | Server for assessing personal information protection | |
Cheng et al. | Contextual geotracking service of incident markers in disaster search-and-rescue operations | |
CN106156610B (en) | A kind of process path acquisition methods, device and electronic equipment | |
CN109036523A (en) | Inquire method, apparatus, equipment and the storage medium of case ID diagnostic message | |
CN105208329B (en) | A kind of image data selecting method and relevant apparatus | |
CN114071062B (en) | Video recording method and device in remote comment video conference process | |
CN109347846A (en) | A kind of website clearance method, apparatus, equipment and readable storage medium storing program for executing | |
CN114071061B (en) | Evaluation method and device for evaluation expert behavior in remote evaluation video conference process |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190308 |
|
RJ01 | Rejection of invention patent application after publication |