CN108921137A - A kind of unmanned plane and storage medium - Google Patents
A kind of unmanned plane and storage medium Download PDFInfo
- Publication number
- CN108921137A CN108921137A CN201810865016.7A CN201810865016A CN108921137A CN 108921137 A CN108921137 A CN 108921137A CN 201810865016 A CN201810865016 A CN 201810865016A CN 108921137 A CN108921137 A CN 108921137A
- Authority
- CN
- China
- Prior art keywords
- image
- density
- portrait
- obtains
- stop
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
- G06V20/53—Recognition of crowd images, e.g. recognition of crowd congestion
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of unmanned plane and storage medium, which includes memory and processor, and processor calls the executable program code of memory storage to execute following steps:Image is acquired in specified region by unmanned plane, obtains the first image;By to the first image carry out crowd density analysis, obtain the first image in density element, when density element be more than threshold value, obtain the second image;The second image is persistently obtained, continuous second image is obtained;By carrying out crowd density analysis to the first image, the stop element in continuous second image is obtained, is more than threshold value when stopping element, obtains third image.The present invention in unmanned plane by being arranged two image capture modules to Image Acquisition number information, collected information is compared with preset threshold, when collected information is more than preset threshold, just acquired image is sent back into press center, greatly improves the convenience and accuracy and the timeliness released news of acquisition hot news.
Description
Technical field
The present invention relates to unmanned plane tracking field more particularly to a kind of unmanned planes and storage medium.
Background technique
With the rapid development of unmanned air vehicle technique, unmanned plane starts to be put to more and more fields.Unmanned plane is most opened
Beginning is to be applied to army, very convenient as one kind, and the unmanned machine of low-risk, is usually used to and does unmanned plane investigation, is supervised
Depending on the work that combats terrorism, for example " predator " in the U.S., Chinese " pterosaur " etc. are all military unmanned air vehicles.When nobody
After machine enters civil field, welcome the climax of Development of UAV, as China Shenzhen great Jiang Science and Technology Ltd. develop it is big
Boundary unmanned plane is even more the fame whole world, has also driven the development of unmanned plane type, there is agricultural watering unmanned plane, and forest fire monitors nothing
It is man-machine, unmanned plane of taking photo by plane etc., express delivery with take out industry all in the corresponding unmanned plane of exploitation.
In today of information age, mobile terminal reading news has become the habit of people, traditional news media
Mode starts to decline, and just to news industry, more stringent requirements are proposed for this, and timely social focus news is not only wanted also to have acquisition
The convenient channel of news.
The news collection mode in general urban district has reporter street to set foot-point, and citizen proactive contact reporter contacts with public security department
Etc. a variety of modes for obtaining news, but due to reporter's limited amount, it cannot many times obtain accurate and timely believe in time
Breath, will result in news lose, out-of-date or valueless news etc..What news was particular about is timeliness and accuracy, the prior art
Acquisition mode it is not good enough in this respect, it is also inconvenient, be unfavorable for being engaged in news report enterprise reduce acquisition news at
This.
Summary of the invention
The purpose of the present invention is in view of the above-mentioned drawbacks of the prior art, providing a kind of unmanned plane and storage medium.
The technical solution adopted by the present invention is that a kind of method of unmanned plane hot news tracking is provided, the method includes:
Unmanned plane acquires image in specified region, obtains the first image;
Unmanned plane carries out crowd density analysis to the first image, the density element in the first image is obtained, when described
Density element is more than threshold value, is the first image configuration identifier, obtains the second image;
Unmanned plane persistently obtains second image, obtains continuous second image;
Unmanned plane identifies continuous second image by the second image recognition, obtains in continuous second image
Stop element for the continuous second image configurations mark, obtain third image when the stop element is more than threshold value;
The third image is sent press center by unmanned plane.
Preferably, the first image is identified as number identification, and the density element is density of human number;
The unmanned plane carries out crowd density analysis to the first image, obtains the density element packet in the first image
It includes:
The number in identification the first image is identified by the first image, obtains number information;
According to the number information recognized, the density of human number in the first image is calculated.
The density of human number that the first image is identified using the first image passes through the density of human number energy
Fuzzy Judgment goes out the first image and whether meets the density of human number requirement of hot news, if meeting density of human number requirement just for institute
The first image tagged is stated into next link, is otherwise judged as valueless news, continues to acquire next the first image, unmanned plane
Some valueless news have been filtered, the accuracy of acquisition news is improved;The density of human number requires to preset in advance, institute
Stating density of human number requirement is that a hot spot density of human number grade, the hot spot are counted according to previous hot news density of human number
Density of human number grade is also referred to as hot spot density of human number threshold value.
Preferably, second image recognition is coordinate identification, and the stop element is residence time and the people of people
Stop quantity;
It is described that continuous second image is identified by the second image recognition, it obtains in continuous second image
Stopping element includes:
The default continuous time for continuously acquiring second image;
The portrait position in second image is identified by second image recognition, obtains portrait in second figure
Location information as in;
According to location information of the portrait in second image, the residence time of people and the stop quantity of people are obtained.
By the default continuous time for continuously acquiring the second image, obtained on the basis of the first image of the label
Continuous second image is got, because there is interval in the continuous time, it is easy to the residence time of the people can be obtained, it is desirable to
The stop quantity of people in certain period of time is obtained, only need to continuous second image in the period be compared and can be thought
The result wanted.
Preferably, the portrait position identified in second image by second image recognition obtains portrait
Location information in second image includes:
Obtain fuzzy coordinate of the portrait in second image;
By comparing the variation of the fuzzy coordinate, a stop threshold value is set, the variation of the fuzzy coordinate is stopping
In threshold value, it is denoted as stop;The stop threshold value is a changes in coordinates range.
Using coordinate setting can clear everyone relative position in tag image, by being obtained in continuous second picture
The fuzzy coordinate of every second image is got, the variation of the fuzzy coordinate on each the second image is compared, judges everyone
Whether fuzzy changes in coordinates range meets preset stop threshold value, if the fuzzy changes in coordinates of most people, which meets, stops threshold
Value, just enters next link for continuous second image tagged, and this judgment mode can largely solve acquisition
The accuracy and timeliness of news.
A kind of unmanned plane hot news tracing system is also provided, the system is used for unmanned plane, the system comprises:
First acquisition module acquires image in specified region for unmanned plane, obtains the first image;
First picture recognition module, unmanned plane identify the first image by the first picture recognition module, obtain first
Density element in image, when the density element obtains the second image more than threshold value for the first image configuration identifier;
Second acquisition module persistently obtains second image for unmanned plane, obtains continuous second image;
Second picture recognition module identifies continuous second figure by the second picture recognition module for unmanned plane
Picture obtains the stop element in continuous second image, is described continuous second when the stops element is more than threshold value
Image configurations mark, obtains third image;
The third image is sent press center by sending module, unmanned plane.
Preferably, the first image identification module includes:Number identifies submodule, density computational submodule;
The number identification submodule is used to obtain number information by the number in identification the first image;
The density computational submodule is used to calculate the people in the first image according to the number information recognized
Number density.
Preferably, second picture recognition module includes:Acquisition time submodule, position identification submodule, time meter
Operator module and quantity calculating submodule;
The acquisition time submodule is for acquiring the continuous time for continuously acquiring second image;
The position identification submodule is used to obtain portrait described by identifying the portrait position in second image
Location information in second image;
The time computational submodule be used for according to portrait in second image residence time, obtain the stop of people
Time;
The quantity calculating submodule obtains the stop of people for the location information according to portrait in second image
Quantity.
Preferably, the position identification submodule includes coordinate acquiring unit, and the coordinate acquiring unit is for obtaining people
As the fuzzy coordinate in second image;
The quantity calculating submodule includes coordinate comparison unit, by comparing the variation of the fuzzy coordinate, setting one
The variation of a stop threshold value, the fuzzy coordinate is stopping in threshold value, is denoted as stop.
A kind of unmanned plane is also provided, the unmanned plane includes processor and memory, is stored at least in the memory
One instruction, at least one section of program, code set or instruction set, at least one instruction, at least one section of program, the generation
Code collection or described instruction collection are loaded by the processor and execute following steps:
Image is acquired in specified region, obtains the first image;
Crowd density analysis is carried out to the first image, the density element in the first image is obtained, when the density is wanted
Element is more than threshold value, is the first image configuration identifier, obtains the second image;
Second image is persistently obtained, continuous second image is obtained;
Residence Time Analysis is carried out to second image, obtains the stop element in continuous second image, when
The stop element is more than threshold value, for the continuous second image configurations mark, obtains third image;
Press center is sent by the third image.
Preferably, the density element is density of human number;
The processor carries out crowd density analysis to the first image, obtains the side of the density element in the first image
Formula includes:
By being identified to the first image, identifies the number in the first image, obtain number information;
According to the number information recognized, the density of human number in the first image is calculated.
Preferably, the stop quantity for stopping residence time and people that element is people;
The processor carries out residence Time Analysis to second image, obtains and stops in continuous second image
The mode for staying element includes:
The default continuous time for continuously acquiring second image;
By being identified to second image, identifies the portrait position in second image and label, obtain people
As the location information in second image;
According to location information of the portrait in second image, the residence time of people and the stop quantity of people are obtained.
Preferably, the processor identifies the portrait in second image by identifying to second image
Position simultaneously marks, and the mode for obtaining location information of the portrait in second image includes:
By portrait position's label in second image, fuzzy coordinate of the portrait in second image is obtained;
A stop threshold value is set, by comparing the variation of the fuzzy coordinate, if the variation of the fuzzy coordinate is stopping
It stays in threshold value, is denoted as stop.
A kind of computer readable storage medium is also provided, at least one instruction, at least one are stored in the storage medium
Duan Chengxu, code set or instruction set, at least one instruction, at least one section of program, the code set or the described instruction
Collection is loaded by processor and executes following steps:
Image is acquired in specified region, obtains the first image;
Crowd density analysis is carried out to the first image, the density element in the first image is obtained, when the density is wanted
Element is more than threshold value, is the first image configuration identifier, obtains the second image;
Second image is persistently obtained, continuous second image is obtained;
Residence Time Analysis is carried out to second image, obtains the stop element in continuous second image, when
The stop element is more than threshold value, for the continuous second image configurations mark, obtains third image;
Press center is sent by the third image.
Preferably, the density element is density of human number;
The processor carries out crowd density analysis to the first image, obtains the side of the density element in the first image
Formula includes:
By being identified to the first image, identifies the number in the first image, obtain number information;
According to the number information recognized, the density of human number in the first image is calculated.
Preferably, the stop quantity for stopping residence time and people that element is people;
The processor carries out residence Time Analysis to second image, obtains and stops in continuous second image
The mode for staying element includes:
The default continuous time for continuously acquiring second image;
By being identified to second image, identifies the portrait position in second image and label, obtain people
As the location information in second image;
According to location information of the portrait in second image, the residence time of people and the stop quantity of people are obtained.
Preferably, the processor identifies the portrait in second image by identifying to second image
Position simultaneously marks, and the mode for obtaining location information of the portrait in second image includes:
By portrait position's label in second image, fuzzy coordinate of the portrait in second image is obtained;
A stop threshold value is set, by comparing the variation of the fuzzy coordinate, if the variation of the fuzzy coordinate is stopping
It stays in threshold value, is denoted as stop.
Compared with prior art, the present invention at least has the advantages that:The present invention in unmanned plane by being arranged two
A image capture module is compared to number of people in image information collection, by collected information with preset threshold, when collected
Information is more than preset threshold, just passes captured image back press center, greatly improves the convenience of acquisition hot news
Property with accuracy and the timeliness released news.
Detailed description of the invention
Fig. 1 is a kind of unmanned plane hot news method for tracing flow chart of the embodiment of the present invention;
Fig. 2 is a kind of unmanned plane hot news tracing system module map of the embodiment of the present invention;
Fig. 3 is the details module map of 22 parts in Fig. 2 of the embodiment of the present invention;
Fig. 4 is the details module map of 24 parts in Fig. 2 of the embodiment of the present invention;
Fig. 5 is a kind of structural schematic diagram of unmanned plane provided in an embodiment of the present invention.
Specific embodiment
The present invention will be further described with reference to the accompanying drawings and examples.
As shown in Figure 1, the unmanned plane can be more the invention proposes a kind of method of unmanned plane hot news tracking
Rotor wing unmanned aerial vehicle, fixed-wing unmanned plane and unmanned helicopter etc., the present invention does not do specifically the product type of the unmanned plane
It limits.The unmanned plane is equipped with high-resolution video camera, and HD image can also be able to maintain higher pixel after amplification, in this way
The information that can be easier in identification described image.
In embodiments of the present invention, the unmanned plane selects quadrotor drone, and the quadrotor drone is provided with can
To shoot high-resolution video camera high-definition, in quadrotor drone, it is additionally provided with video memory and image recognition fills
It sets.
The method includes:
11, the first image is obtained, unmanned plane acquires image in specified region, obtains the first image;
Specifically, ground press center control unmanned plane flies to target area, by the way that adopting on the unmanned plane is arranged in
Collect vision facilities and obtain the first image, the acquisition vision facilities can be the equipment that video camera etc. has image collecting function.
Further, in order to obtain clearer first image, the acquisition vision facilities is high-resolution video camera, from the
One image solves interference of the fuzzy objective to image recognition result.
12, the first image recognition, unmanned plane carry out crowd density analysis to the first image, obtain in the first image
Density element;First image recognition is function, and effect is the density element identified on the first image, the density element
By region number and the density of human number calculated according to region area.
Specifically, the first image is identified as thermal imaging scan identification, the first image is being obtained simultaneously, thermal imaging scan
Instrument scans the heat radiation of outdoor scene number, counts the number in outdoor scene, obtains number information in the first image;According to recognizing
The number information, the density of human number in the first image is calculated by density of population calculation formula.
Certainly, the first image identification is also possible to portrait scanning recognition, does not limit to and a kind of mode.
13, according to the density of human number in the first image to determine whether meeting default density of human number threshold value, if institute
Density of human number is stated more than the density of human number threshold value, is just the first image configuration identifier, obtains the second image;If described
Density of human number in first image does not reach the density threshold, just by the first image storage that this is acquired, as with reference to figure
Picture, video camera reacquire the first new image.
14, the second image is continuously acquired;First the second image is obtained according to the first image of the label, in this base
Fixed time interval is set to continuously acquire the second image, the method for obtaining the second image is above-mentioned steps 11- on plinth
13。
Further, continuous second image that will acquire is sequentially arranged, and is reduced due to two band of sequence disorder
The data interference come.
15, the second image recognition, the second image described in unmanned plane carry out residence Time Analysis, obtain described continuous second
Stop element in image;Further information identification, second image recognition are carried out according to continuous second image of acquisition
For the portrait position in the second image of identification, location information of the portrait in second image is obtained, each continuous the
Two images can get the location information of corresponding portrait, can count portrait by the variation of the continuous position information
The stop quantity of residence time and within a certain period of time portrait is as stop element.
Specifically, the second image recognition mode is coordinate identification, each of image portrait position can be by
One group of fuzzy coordinate representation, portrait position's information is to obtain the portrait coordinate information in obtaining second image.
16, when whether determine portrait is to stop, a residence time and a stop threshold value, that is, coordinate information are first preset
The coordinate information of identical portrait in continuous second image within the residence time is compared, the seat by variation range
Mark information change range is less than stop threshold value, and then the portrait is denoted as stop;The residence time is changed for portrait by coordinate information small
In stop threshold value to the time for being greater than stop threshold value;The portrait quantity that stops is the portrait number that stop is denoted as in a period of time
Amount.
Further, it after most portraits are denoted as stop, meets preset number and stops threshold value, the group is continuous
The second image configurations label, obtain third image.
17, by continuous second image of label as third image, the third image has figure as one group of atlas
As quantity is big, the features such as image recognition data are complicated, the third compression of images for needing will acquire thus and encryption, encryption be for
Guarantee that data are not intercepted when transmission data;Compressed image is converted into digital signal by digital analog converter, is convenient for
It is transmitted with the data of console.
18, the third image digital signal is sent back into press center by sending device, completes hot news tracking
Final step.
As shown in Figures 2 to 4, a kind of unmanned plane hot news tracing system is additionally provided, the unmanned plane can be more
Rotor wing unmanned aerial vehicle, fixed-wing unmanned plane and unmanned helicopter etc., the present invention does not do specifically the product type of the unmanned plane
It limits.The unmanned plane is equipped with high-resolution video camera, and HD image can also be able to maintain higher pixel after amplification, in this way
The information that can be easier in identification described image.
The system comprises:
First acquisition module 21, for obtaining the first image;First module 21 is provided with high-resolution video camera.
First picture recognition module 22, for identification number information of the first image;In the first image identification module
It is provided with number identification submodule 31 and density computational submodule 32.
In embodiments of the present invention, the number identification submodule 31 is thermal imaging scan instrument, the thermal imaging scan instrument
It is to obtain density of human number, personal infrared radiant energy magnitude less, is considered as fixed value by everyone infrared radiant energy magnitude fluctuation, can be with
Number is calculated according to total temperature is radiated;The number and outdoor scene area are made portrait densitometer and calculated by the density submodule 32,
Obtain density of human number;The portrait density calculation formula is:(portrait density)=(number) ÷ (outdoor scene area).
Second acquisition module 23, for obtaining the second image;Fixed Time Interval is specially first preset, to first the
After one image configurations mark, it is spaced the preset time interval and continuously acquires the second image.
Second picture recognition module 24, for identification the portrait quantity on continuous second image and portrait position;It is described
Second picture recognition module include the acquisition time submodule 41 for acquiring the continuous second image spacing time, identification described in
Portrait position in second image obtains the position identification submodule 42 of location information of the portrait in second image, adopts
Collect the quantity that number in the second image of time computational submodule 43 and acquisition of portrait residence time in the second image stops quantity
Computational submodule 44.
The acquisition time submodule 41 is to acquire continuous second image according to the time label on the second image
Time interval;The position identification submodule 42 is the portrait position's information identified according to the coordinate being arranged on the image, because
It is two-dimensional image for the second image, so coordinate setting is also two-dimensional coordinate,;The time computational submodule 43 is the company of calculating
Counting period time of the residence time of portrait in the second continuous image, the time computational submodule 43 are less than continuous second
The time interval of image more accurately calculates the portrait residence time;There are two the functions that the quantity calculating submodule 44 is,
One be statistics every second image in portrait quantity, for portrait quantity variation comparison, further increase hot news
Accuracy, the other is identification portrait stops, by the coordinate comparison unit that is arranged in the quantity calculating submodule 44 come
It realizes, compares the variation of the portrait coordinate, set a stop threshold value, the variation of the coordinate is being stopped in threshold value, is being denoted as
It stops.
Sending module 25 is for sending third image;Described ground sending module 25 will acquire third elementary area and hair
Elementary area is sent to be combined into one, the acquisition third elementary area is the second compression of images that will be marked, encrypts and be converted into number
Word signal, the transmission elementary area is then that third image digital signal is sent to press center.
The embodiment of the present invention also provides a kind of unmanned plane, can be used for executing the unmanned plane news heat of previous embodiment offer
The method of point tracking.As shown in figure 5, the unmanned plane at least may include memory 100 and at least one processor 200, such as
CPU (Central Processing Unit, central processing unit), wherein memory 100 and processor 200 can pass through bus
It is communicatively coupled.It will be understood by those skilled in the art that the structure of unmanned plane shown in Fig. 5 is not constituted to of the invention real
The restriction of example is applied, it is also possible to hub-and-spoke configuration either busbar network, can also include more more or fewer than illustrating
Component perhaps combines certain components or different component layouts.
Wherein, memory 100 can be high speed RAM memory, be also possible to non-labile memory (non-
Volatile memory), a for example, at least magnetic disk storage.It is remote that memory 100 optionally can also be that at least one is located at
Storage device from aforementioned processor 200.At least one instruction, at least one section of program, code are can store in memory 100
Collection or instruction set, the embodiment of the present invention are not construed as limiting.
In unmanned plane shown in Fig. 5, processor 200 can be used for calling above-mentioned at least one stored in memory 100
Item instruction, at least one section of program, code set or instruction set, execute following steps:
Image is acquired in specified region, obtains the first image;
Crowd density analysis is carried out to the first image, obtains the density element in the first image, when density element is more than threshold
Value identifies for the first image configurations, obtains the second image;
The second image is persistently obtained, continuous second image is obtained;
Residence Time Analysis is carried out to the second image, obtains the stop element in continuous second image, when stop element
More than threshold value, is identified for continuous second image configurations, obtain third image;
Press center is sent by third image.
Optionally, which is density of human number;
Processor 200 carries out crowd density analysis to the first image, and the mode for obtaining the density element in the first image can
To include:
By being identified to the first image, identifies the number in the first image, obtain number information;
According to the number information recognized, the density of human number in the first image is calculated.
Optionally, which is the residence time of people and the stop quantity of people;
Processor 200 carries out residence Time Analysis to the second image, obtains the stop element in continuous second image
Mode may include:
The default continuous time for continuously acquiring the second image;
By being identified to the second image, identifies the portrait position in the second image and label, obtain portrait second
Location information in image;
According to location information of the portrait in the second image, the residence time of people and the stop quantity of people are obtained.
Optionally, processor 200 identifies the portrait position in the second image and mark by identifying to the second image
Note, the mode for obtaining location information of the portrait in the second image may include:
By portrait position's label in the second image, fuzzy coordinate of the portrait in the second image is obtained;
A stop threshold value is set, by comparing the variation of fuzzy coordinate, if the variation of fuzzy coordinate is stopping in threshold value,
It is denoted as stop.
Unmanned plane shown in fig. 5 will be collected by two image capture modules of setting to number of people in image information collection
Information is compared with preset threshold, when collected information be more than preset threshold, just captured image is passed back in news
The heart greatly improves the convenience and accuracy and the timeliness released news of acquisition hot news.
The embodiment of the present invention also provides a kind of computer readable storage medium, can store at least one in the storage medium
Item instruction, at least one section of program, code set or instruction set, above-mentioned at least one instruction, at least one section of program, code set or instruction
Collection can be loaded by the processor in unmanned plane and execute following steps:
Image is acquired in specified region, obtains the first image;
Crowd density analysis is carried out to the first image, obtains the density element in the first image, when density element is more than threshold
Value identifies for the first image configurations, obtains the second image;
The second image is persistently obtained, continuous second image is obtained;
Residence Time Analysis is carried out to the second image, obtains the stop element in continuous second image, when stop element
More than threshold value, is identified for continuous second image configurations, obtain third image;
Press center is sent by third image.
Optionally, which is density of human number;
Processor carries out crowd density analysis to the first image, and the mode for obtaining the density element in the first image can wrap
It includes:
By being identified to the first image, identifies the number in the first image, obtain number information;
According to the number information recognized, the density of human number in the first image is calculated.
Optionally, which is the residence time of people and the stop quantity of people;
Processor carries out residence Time Analysis to the second image, obtains the mode of the stop element in continuous second image
May include:
The default continuous time for continuously acquiring the second image;
By being identified to the second image, identifies the portrait position in the second image and label, obtain portrait second
Location information in image;
According to location information of the portrait in the second image, the residence time of people and the stop quantity of people are obtained.
Optionally, processor identifies the portrait position in the second image and label, obtains by identifying to the second image
The mode for taking location information of the portrait in the second image may include:
By portrait position's label in the second image, fuzzy coordinate of the portrait in the second image is obtained;Setting one
Threshold value is stopped, by comparing the variation of fuzzy coordinate, if the variation of fuzzy coordinate is stopping in threshold value, is denoted as stop.
Above-described embodiment is merely to illustrate a specific embodiment of the invention.It should be pointed out that for the general of this field
For logical technical staff, without departing from the inventive concept of the premise, several deformations and variation can also be made, these deformations and
Variation all should belong to protection scope of the present invention.
Claims (8)
1. a kind of unmanned plane, which is characterized in that the unmanned plane includes processor and memory, be stored in the memory to
Few an instruction, at least one section of program, code set or instruction set, it is at least one instruction, at least one section of program, described
Code set or described instruction collection are loaded by the processor and execute following steps:
Image is acquired in specified region, obtains the first image;
Crowd density analysis is carried out to the first image, obtains the density element in the first image, when the density element is super
Threshold value is crossed, is the first image configuration identifier, obtains the second image;
Second image is persistently obtained, continuous second image is obtained;
Residence Time Analysis is carried out to second image, the stop element in continuous second image is obtained, when described
Stopping element is more than threshold value, for the continuous second image configurations mark, obtains third image;
Press center is sent by the third image.
2. unmanned plane as described in claim 1, which is characterized in that the density element is density of human number;
The processor carries out crowd density analysis to the first image, obtains the mode packet of the density element in the first image
It includes:
By being identified to the first image, identifies the number in the first image, obtain number information;
According to the number information recognized, the density of human number in the first image is calculated.
3. unmanned plane as claimed in claim 1 or 2, which is characterized in that described to stop the residence time and people that element is people
Stop quantity;
The processor carries out residence Time Analysis to second image, and the stop obtained in continuous second image is wanted
Element mode include:
The default continuous time for continuously acquiring second image;
By being identified to second image, identifies the portrait position in second image and label, obtain portrait and exist
Location information in second image;
According to location information of the portrait in second image, the residence time of people and the stop quantity of people are obtained.
4. unmanned plane as claimed in claim 3, which is characterized in that the processor is by knowing second image
Not, portrait position and the label in second image are identified, the side of location information of the portrait in second image is obtained
Formula includes:
By portrait position's label in second image, fuzzy coordinate of the portrait in second image is obtained;
A stop threshold value is set, by comparing the variation of the fuzzy coordinate, if the variation of the fuzzy coordinate is stopping threshold
In value, it is denoted as stop.
5. a kind of computer readable storage medium, which is characterized in that be stored at least one instruction, at least in the storage medium
One section of program, code set or instruction set, at least one instruction, at least one section of program, the code set or the finger
Collection is enabled to be loaded by processor and execute following steps:
Image is acquired in specified region, obtains the first image;
Crowd density analysis is carried out to the first image, obtains the density element in the first image, when the density element is super
Threshold value is crossed, is the first image configuration identifier, obtains the second image;
Second image is persistently obtained, continuous second image is obtained;
Residence Time Analysis is carried out to second image, the stop element in continuous second image is obtained, when described
Stopping element is more than threshold value, for the continuous second image configurations mark, obtains third image;
Press center is sent by the third image.
6. computer readable storage medium as claimed in claim 5, which is characterized in that the density element is density of human number;
The processor carries out crowd density analysis to the first image, obtains the mode packet of the density element in the first image
It includes:
By being identified to the first image, identifies the number in the first image, obtain number information;
According to the number information recognized, the density of human number in the first image is calculated.
7. such as computer readable storage medium described in claim 5 or 6, which is characterized in that the element that stops is stopping for people
Stay time and the stop quantity of people;
The processor carries out residence Time Analysis to second image, and the stop obtained in continuous second image is wanted
Element mode include:
The default continuous time for continuously acquiring second image;
By being identified to second image, identifies the portrait position in second image and label, obtain portrait and exist
Location information in second image;
According to location information of the portrait in second image, the residence time of people and the stop quantity of people are obtained.
8. computer readable storage medium as claimed in claim 7, which is characterized in that the processor passes through to described second
Image is identified, identifies the portrait position in second image and label, obtains position of the portrait in second image
Confidence breath mode include:
By portrait position's label in second image, fuzzy coordinate of the portrait in second image is obtained;
A stop threshold value is set, by comparing the variation of the fuzzy coordinate, if the variation of the fuzzy coordinate is stopping threshold
In value, it is denoted as stop.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810865016.7A CN108921137A (en) | 2018-08-01 | 2018-08-01 | A kind of unmanned plane and storage medium |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810865016.7A CN108921137A (en) | 2018-08-01 | 2018-08-01 | A kind of unmanned plane and storage medium |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108921137A true CN108921137A (en) | 2018-11-30 |
Family
ID=64393337
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810865016.7A Pending CN108921137A (en) | 2018-08-01 | 2018-08-01 | A kind of unmanned plane and storage medium |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108921137A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111291597A (en) * | 2018-12-07 | 2020-06-16 | 杭州海康威视数字技术股份有限公司 | Image-based crowd situation analysis method, device, equipment and system |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105447458A (en) * | 2015-11-17 | 2016-03-30 | 深圳市商汤科技有限公司 | Large scale crowd video analysis system and method thereof |
CN105809178A (en) * | 2014-12-31 | 2016-07-27 | 中国科学院深圳先进技术研究院 | Population analyzing method based on human face attribute and device |
CN106845344A (en) * | 2016-12-15 | 2017-06-13 | 重庆凯泽科技股份有限公司 | Demographics' method and device |
CN106933867A (en) * | 2015-12-30 | 2017-07-07 | 杭州华为企业通信技术有限公司 | A kind of image inquiry method and device |
CN107256225A (en) * | 2017-04-28 | 2017-10-17 | 济南中维世纪科技有限公司 | A kind of temperature drawing generating method and device based on video analysis |
-
2018
- 2018-08-01 CN CN201810865016.7A patent/CN108921137A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105809178A (en) * | 2014-12-31 | 2016-07-27 | 中国科学院深圳先进技术研究院 | Population analyzing method based on human face attribute and device |
CN105447458A (en) * | 2015-11-17 | 2016-03-30 | 深圳市商汤科技有限公司 | Large scale crowd video analysis system and method thereof |
CN106933867A (en) * | 2015-12-30 | 2017-07-07 | 杭州华为企业通信技术有限公司 | A kind of image inquiry method and device |
CN106845344A (en) * | 2016-12-15 | 2017-06-13 | 重庆凯泽科技股份有限公司 | Demographics' method and device |
CN107256225A (en) * | 2017-04-28 | 2017-10-17 | 济南中维世纪科技有限公司 | A kind of temperature drawing generating method and device based on video analysis |
Non-Patent Citations (1)
Title |
---|
叶青青: "《网络与新媒体财经报道》", 31 July 2018 * |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111291597A (en) * | 2018-12-07 | 2020-06-16 | 杭州海康威视数字技术股份有限公司 | Image-based crowd situation analysis method, device, equipment and system |
CN111291597B (en) * | 2018-12-07 | 2023-10-13 | 杭州海康威视数字技术股份有限公司 | Crowd situation analysis method, device, equipment and system based on image |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
KR101984983B1 (en) | System and method for monitoring wild animals | |
WO2020215961A1 (en) | Personnel information detection method and system for indoor climate control | |
WO2020206850A1 (en) | Image annotation method and device employing high-dimensional image | |
CN110532970B (en) | Age and gender attribute analysis method, system, equipment and medium for 2D images of human faces | |
KR20200098875A (en) | System and method for providing 3D face recognition | |
CN104537339B (en) | Information identifying method and information identification system | |
CN105279480A (en) | Method of video analysis | |
CN101300588A (en) | Determining a particular person from a collection | |
CN106548148A (en) | The recognition methodss of unknown face and system in video | |
CN105022999A (en) | Man code company real-time acquisition system | |
DE102016203882A1 (en) | Recognition of tumorous breast tissue on a thermal image | |
DE102016203881A1 (en) | Software interface tool for breast cancer investigation | |
CN107566790A (en) | With reference to REID and the real time monitoring apparatus and method of Video Supervision Technique | |
Schumann et al. | An image processing pipeline for long range UAV detection | |
CN103020655A (en) | Remote identity authentication method based on single training sample face recognition | |
CN109087515A (en) | Unmanned plane expressway road conditions cruise method and system | |
Petso et al. | Individual animal and herd identification using custom YOLO v3 and v4 with images taken from a uav camera at different altitudes | |
JP6396682B2 (en) | Surveillance camera system | |
CN108921137A (en) | A kind of unmanned plane and storage medium | |
Kashihara et al. | Supportive information to find victims from aerial video in search and rescue operation | |
CN109063646A (en) | Method, system, unmanned plane and the storage medium of unmanned plane hot news tracking | |
CN109080819A (en) | A kind of unmanned plane and storage medium for news tracking | |
Ntouskos et al. | Saliency prediction in the coherence theory of attention | |
Peng et al. | [Retracted] Helmet Wearing Recognition of Construction Workers Using Convolutional Neural Network | |
Qi et al. | Birdseyeview: aerial view dataset for object classification and detection |
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 | ||
WD01 | Invention patent application deemed withdrawn after publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20181130 |