CN105898216B - A kind of number method of counting carried out using unmanned plane - Google Patents
A kind of number method of counting carried out using unmanned plane Download PDFInfo
- Publication number
- CN105898216B CN105898216B CN201610231442.6A CN201610231442A CN105898216B CN 105898216 B CN105898216 B CN 105898216B CN 201610231442 A CN201610231442 A CN 201610231442A CN 105898216 B CN105898216 B CN 105898216B
- Authority
- CN
- China
- Prior art keywords
- unmanned plane
- frame
- video
- valid frame
- valid
- 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.)
- Active
Links
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- G05D1/08—Control of attitude, i.e. control of roll, pitch, or yaw
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06M—COUNTING MECHANISMS; COUNTING OF OBJECTS NOT OTHERWISE PROVIDED FOR
- G06M11/00—Counting of objects distributed at random, e.g. on a surface
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30242—Counting objects in image
Abstract
The present invention relates to a kind of number method of counting carried out using unmanned plane, using the unmanned plane for being mounted with camera vertically downward, unmanned plane flies over assembly place, by collected video by passing to ground handling platform, by the image procossing to video, the number on spacious place is counted.It can be used for density of personnel management and the security of city, square and large-scale assembly place in the express statistic closed with carry out number quantity under open environment.
Description
Technical field
The invention patent relates to field of image processings, and in particular to a kind of to carry out demographics using unmanned aerial vehicle onboard video
Method.
Background technique
With the fast development of Chinese society economy, people material requisite be continuously available meet while, to spirit
The demand of civilization is continuously increased, and various large-scale activities are constantly held.But it is trampled since a large amount of gathering of people is easy to bring
Safety problem, it is related also more and more important for the management work of density of personnel, it would be desirable to which that the number of assembly place is carried out
Statistics.
At present due to pattern-recognition and the rapid development of image processing techniques, fixed image capture device is set and passes through system
A large amount of application is had been obtained by the method for number in meter image, but those are substantially for counting a crossing or section
Point by number, it is for there is this method of the scenes such as the megastore of fixed entrance, station, museum also relatively more real
With.It is open for square, playground etc. without fixed entrance or be inconvenient to be arranged fixation monitoring device area we just
Crowd's quantity can not rapidly be counted.
With the development of unmanned air vehicle technique, unmanned air vehicle technique is all utilized in every field, there is electric inspection process, agricultural
Inspection, environmental observation etc., unmanned plane technology also have a large amount of application on mapping takes photo by plane and photographs.Small-sized rotor nobody
Machine is largely popularized with its human-oriented cost, and small-sized rotor wing unmanned aerial vehicle has light portable, the number such as height velocity
According to easy manipulation the advantages that.
Summary of the invention
The technical problem to be solved by the present invention is to provide a kind of number method of counting carried out using unmanned plane, Neng Gou
Closing and the express statistic that number quantity is carried out under open environment, can be used for the personnel of city, square and large-scale assembly place
Density management and security.
The present invention is in order to solve the above technical problems, adopt the following technical scheme that
A kind of number method of counting carried out using unmanned plane, it is characterised in that: using being mounted with camera vertically downward
Unmanned plane, control unmanned plane flies over place to be measured and acquires video in flight course, by collected video by wirelessly passing
Defeated to pass to ground handling platform, ground handling platform is by counting the number on place to be measured to video image processing;It is described nobody
Owner will by unmanned plane body part, include the video acquisition module of camera, wireless transport module, video figure vertically downward
As four part compositions of processing module, unmanned plane is connect by video wireless transmission module with computer host computer telecommunication.
In above-mentioned technical proposal, mainly include the following steps:
Step S1: by computer host computer, specified path of taking photo by plane, starting unmanned plane starts to leap to target place side to be measured
Edge;
Step S2: by computer host computer, setting the flying height and speed of unmanned plane, after selecting " starting to count ", nothing
It is man-machine to start to fly over target place to be measured and acquire video flowing, in flight course unmanned plane by wireless transmitting system will vertically to
The lower collected video of camera is streamed to computer host computer;
Step S3: on computers on the machine of position, the posture comparison using unmanned plane when extracting video guarantees to image vertically downward
Head pair direction it is consistent;By obtaining single human body corresponding size in video using drone flying height;Benefit
The number that the non-overlap part in valid frame counts unmanned plane flyover region is calculated with unmanned plane during flying speed.
In above-mentioned technical proposal, step S3 specifically includes the following steps:
Step S31: be delayed setting time afterwards for selection " starting to count ", after unmanned plane during flying is in stable condition, from video flowing
It is middle to start to select valid frame to be processed, wherein the 1st frame is the 0th frame valid frame;Then every the interval time of setting, by pair
The roll angle γ of unmanned plane when shooting the valid frameaAnd pitching angle thetaaAfter whether unanimously being judged with course angle and the 0th frame,
Valid frame after selection;
Step S32: it in order to avoid leakage statistics and repeats to count, in conjunction with the flying height h of unmanned plane, speed v is to having selected
Valid frame sheared, the width of the live part being cut out is original width width, a length of small two frame valid frame of unmanned plane it
Between the actual range that leaps be reconverted into the physical length λ in picturehv(ti-ti-1), wherein λhIt is real when flying height is h
Conversion coefficient at a distance from showing in border flying distance and video, v are the flying speed of unmanned plane, (ti-ti-1) it is that two frames are effective
Time difference between frame.
Step S33: image procossing, pattern-recognition are carried out to the live part of valid frame, identify the crowd in valid frame
Quantity, it is constantly accumulative, and show particular number, this quantity is exactly the crowd's quantity counted:
In conjunction with the flying height information of unmanned plane, single human body corresponding pixel size in video is calculated;To image
It carries out carrying out image binaryzation based on HSV color space, the black of prominent human hair weakens other colors;With single human body institute
Accounting for pixel size is scale, carries out closed operation, eliminates small black patch;By carrying out spot detection to image, identify that one has
Imitate the number in the live part of frame;It is cumulative to obtain the total number of persons in the target place region to be measured that unmanned plane is flown over.
In above-mentioned technical proposal, computer PC control unmanned plane flies over measured target, and unmanned plane automatically keeps fixed high, even
Fast rectilinear flight.
In above-mentioned technical proposal, unmanned plane is small-sized multi-rotor unmanned aerial vehicle.
In above-mentioned technical proposal, selecting " starting to count ", be delayed 3s afterwards, after unmanned plane during flying is in stable condition, from video flowing
It is middle to start to select valid frame to be processed.
The present invention carries out rally number counting using UAV flight's video capture device, instead of the irremovable, visual field
The advantages that limited fixed monitoring device makes this system be provided with unmanned plane convenient and efficient, light and flexible, controllability is strong,
The crowd's quantity that conveniently can effectively count certain piece spaciousness place, especially suitable for outdoor open large-scale assembly place
Demographics and counting.
In the present invention, unmanned plane automatically keep fixed high, unaccelerated flight can reduce operator workload it is same
When, the difficulty of video image processing can be greatly simplified;When extracting valid frame, first to the posture of unmanned plane, i.e. roll angle γaWith
Pitching angle thetaaWith course angleIt is compared with the posture of the 0th frame valid frame, keeps the posture of selected valid frame consistent, avoid
Shake bring when unmanned plane during flying influences.Multiple steps are all utilized unmanned plane and fly when carrying out image procossing to video flowing
Row data improve arithmetic speed, guarantee demographics using the algorithm complexity of existing data reduction video image processing
Real-time and accuracy.
An innovation of the invention is the flight attitude data by reading rotor wing unmanned aerial vehicle, ensure that image capture device
State when collecting valid frame is all identical, and it is unstable to avoid shaking etc. brought by ambulatory platform.
An innovative point of the invention is that rotor wing unmanned aerial vehicle flies in different height, and single human body is corresponding in video
Size is fixed, excessive and too small similar image can directly be judged as it is not human body, substantially increases detection speed
With accuracy.
An innovative point of the invention is height velocity's data by rotor wing unmanned aerial vehicle, and the number on a bulk of square is united
Meter problem is simply decomposed into the accumulative of number in the small picture of multiframe, avoids the splicing to multiple image,
Detailed description of the invention
Fig. 1 is unmanned plane structural schematic diagram of the present invention;
Fig. 2 is the schematic diagram for the HSV color space that the present invention is related to using the number method of counting that unmanned plane carries out;
Fig. 3 is the number method of counting overall procedure schematic diagram that invention is carried out using unmanned plane;
Fig. 4 is that image procossing involved in number method of counting step S3 that invention is carried out using unmanned plane shows
Meaning flow chart.
Specific implementation measure
The technical solution of patent in order to further illustrate the present invention, control attached drawing carry out specifically the invention patent
It is bright, but it should be stated that following facilities and equipments and embodiment are not limitation of the present invention, for implementing according to the present invention
Mode work done energy, method or equivalent transformation or substitution in structure, belong within protection scope of the present invention.
The unmanned plane that the present invention this time uses is small-sized quadrotor drone, and video image acquisition equipment uses GoPro
Hero4 motion cameras, graphic transmission equipment are big boundary DJI5.8G high definition simulation drawing biography, and computer host computer uses association Y450
Laptop, as shown in Figure 1, video camera is arranged on unmanned plane vertically downward, unmanned plane and computer host computer are wirelessly connected.
Experimental site is spacious outdoor congregation plaza.
Fig. 3 is the number method of counting overall procedure schematic diagram that invention is carried out using unmanned plane;Mainly include
Following steps:
Step S1: starting small-sized multi-rotor unmanned aerial vehicle, by computer host computer, controls small-sized multi-rotor unmanned aerial vehicle and flies to mesh
Mark square edge.
Step S2: by computer host computer, the flying height and speed of small-sized multi-rotor unmanned aerial vehicle are set, selection " starts
Statistics " after, small-sized multi-rotor unmanned aerial vehicle starts to fly over target square, and will be carried on rotor wing unmanned aerial vehicle by wireless transmitting system
The collected video of camera vertically downward be streamed to computer host computer.
Step S31: selection starts to count, and selects available video frame (namely valid frame) every a frame from video flowing:
Step S311: the 1st frame of selection is the 0th frame valid frame first.
Step S312: in addition to the 1st frame, at interval of a frame, posture when according to UAV Attitude at this time and 1 frame is compared, such as
Fruit posture is identical, then judges that this frame is effective, and posture difference is then spaced b frame and compares again, until there is valid frame, takes out valid frame
As the 2nd frame image, a, b are positive integer.
Step S13: effective (non-overlap) part in each frame is selected.
The whole frame of the S14: the 0 frame valid frame of step is not live part, the number M=M counted at this time0=0.
Step S32: valid frame later by with previous valid frame interval time and rotor wing unmanned aerial vehicle speed, calculate
The distance newly increased practical than previous valid frame in latter valid frame out, and the flying height of rotor wing unmanned aerial vehicle is combined, it selects
In this valid frame with the non-overlap part in former frame, the live part of this part valid frame thus.Live part is extracted first to exist
Some coordinate origin for live part is determined on the picture of valid frame, the coordinate origin of live part is scheduled on (0,300) here.
Then the width width for marking live part, height λ are providedhv(ti-ti-1), wherein λhFly to be practical when flying height is h
Conversion coefficient v at a distance from showing in row distance and video is the flying speed of small-sized multi-rotor unmanned aerial vehicle, (ti-ti-1) it is two
The time difference of frame valid frame.
Step S33: image procossing, pattern-recognition are carried out to the live part of valid frame, identify crowd's number in this frame
Amount, it is constantly accumulative, and show, this quantity is exactly the crowd's quantity counted.
Step S331: image processing process is first to carry out noise reduction herein, removes noise.
Step S332: the second step of image procossing is to carry out image binaryzation based on HSV color space herein, and people is from vertical view
See that the crown is black in figure, so being the standard of binaryzation according to the V component of HSV color space, that is, bright colors degree, such as Fig. 2 institute
Show.Here the calculation formula of V component are as follows:
V=max (R, G, B)/255
R, G, B, that is, primary colors red (red), green (green), blue three channels (blue) color value.
Step S333: the third step of image procossing is to carry out closed operation herein, and closed operation is a kind of according to mathematical morphology
A kind of image processing method that set enumeration tree develops is divided into expansion (dilate) and corrosion (erode) two step.
The mathematical definition of dilation operation may be expressed as:
S=X ⊕ B=x, y | Bxy∩X≠}
S is the image collection after operation in formula;B is corroded for doing, the structural element of dilation operation, it is by 0
With 1 composition, possess a central point, arbitrary shape, be chosen to be the circle close to number of people shape structural element here
Shape;BxyIt is the central point of structural element B;X is the image set for the live part that preceding step has already passed through valid frame after binaryzation
It closes.
The mathematical definition of erosion operation may be expressed as:
S is the later image collection of operation in formula;B is corroded for doing, the structural element of dilation operation, here we
Use circular structural element the same in dilation operation;BxyIt is the central point of structural element B;After X is used here as erosion operation
Image.
Closed operation can exclude small-sized black region, determine single people in picture based on the flying height of rotor wing unmanned aerial vehicle
In actual size, we can determine the parameter of closed operation based on this, a large amount of to exclude the block smaller than single human body simultaneously point
Cut open the connection between the single human body in part.
Step S334: the 4th step of image procossing is to extract each of image processed block herein,
It is completed in opencv using cvFindContours, this order effect passes through to extract all profiles
CvContourArea calculates i.e. each piece of area of each profile, area and the theoretically single human body of these blocks are answered
The area of the block of the generation is compared, and excludes area and single human body distinguishes very big part.Remaining area and single people
The quantity of similar piece of body top view size crowd's quantity M in valid frame thusi。
Crowd's display number that multi-rotor unmanned aerial vehicle has leapt part is(M is total number of persons, MiHave for i-th
The number counted in effect frame).
Claims (5)
1. a kind of number method of counting carried out using unmanned plane, it is characterised in that: using being mounted with camera vertically downward
Unmanned plane, control unmanned plane fly over place to be measured and acquire video in flight course, and collected video is passed through wireless transmission
Ground handling platform is passed to, ground handling platform is by counting the number on place to be measured to video image processing;The unmanned plane
Mainly by unmanned plane body part, include the video acquisition module of camera, wireless transport module, video image vertically downward
Four parts of processing module are constituted, and unmanned plane is connect by video wireless transmission module with computer host computer telecommunication;Mainly
Include the following steps:
Step S1: by computer host computer, specified path of taking photo by plane, starting unmanned plane starts to leap to target place edge to be measured;
Step S2: by computer host computer, setting the flying height and speed of unmanned plane, after selecting " starting to count ", unmanned plane
Start to fly over target place to be measured and acquire video flowing, the 1st frame of selection is the 0th frame valid frame first;In addition to the 1st frame, at interval of a
Frame, posture comparison when according to UAV Attitude at this time and 1 frame, if posture is identical, judges that this frame is effective, posture is different
It is then spaced b frame to compare again, until there is valid frame, takes out valid frame as the 2nd frame image, a, b are positive integer;
Valid frame to be processed is selected since video flowing, wherein the 1st frame is the 0th frame valid frame;Then every between setting
Every the time, by the roll angle γ to unmanned plane when shooting the valid frameaAnd pitching angle thetaaWith course angle and the 0th frame whether one
Valid frame after cause is judged, after choosing;Unmanned plane will be imaged vertically downward by wireless transmitting system in flight course
Collected video is streamed to computer host computer;
Step S3: on computers on the machine of position, the posture comparison using unmanned plane when extracting video guarantees camera institute vertically downward
Pair direction it is consistent;By obtaining single human body corresponding size in video using drone flying height;Utilize nothing
Man-machine flying speed calculates the number that the non-overlap part in valid frame counts unmanned plane flyover region.
2. the number method of counting according to claim 1 carried out using unmanned plane, it is characterised in that: step S3 is specifically wrapped
Include following steps:
Step S31: be delayed setting time afterwards for selection " starting to count ", after unmanned plane during flying is in stable condition, opens from video flowing
Beginning selects valid frame to be processed;
Step S32: it in order to avoid leakage statistics and repeats to count, in conjunction with the flying height h of unmanned plane, speed v has to what is selected
Effect frame is sheared, and the width for the live part being cut out is original width width, is flown between a length of small two frame valid frame of unmanned plane
Actual range more is reconverted into the physical length λ in picturehv(ti-ti-1), wherein λhFly to be practical when flying height is h
Conversion coefficient at a distance from showing in row distance and video, v are the flying speed of unmanned plane, (ti-ti-1) be two frame valid frames it
Between time difference,
Step S33: carrying out image procossing, pattern-recognition to the live part of valid frame, identify crowd's quantity in valid frame,
It is constantly accumulative, and show particular number, this quantity is exactly the crowd's quantity counted:
In conjunction with the flying height information of unmanned plane, single human body corresponding pixel size in video is calculated;Image is carried out
Image binaryzation is carried out based on HSV color space, the black of prominent human hair weakens other colors;With picture shared by single human body
Plain size is scale, carries out closed operation, eliminates small black patch;By carrying out spot detection to image, a valid frame is identified
Live part in number;It is cumulative to obtain the total number of persons in the target place region to be measured that unmanned plane is flown over.
3. the number method of counting according to claim 2 carried out using unmanned plane, it is characterised in that: computer host computer control
Unmanned plane processed flies over measured target, and unmanned plane automatically keeps fixed high, unaccelerated flight.
4. the number method of counting according to claim 1 carried out using unmanned plane, it is characterised in that: unmanned plane is small-sized
Multi-rotor unmanned aerial vehicle.
5. the number method of counting according to claim 3 carried out using unmanned plane, it is characterised in that: selection " starts to unite
Meter " be delayed 3s afterwards, after unmanned plane during flying is in stable condition, since video flowing select valid frame to be processed.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610231442.6A CN105898216B (en) | 2016-04-14 | 2016-04-14 | A kind of number method of counting carried out using unmanned plane |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610231442.6A CN105898216B (en) | 2016-04-14 | 2016-04-14 | A kind of number method of counting carried out using unmanned plane |
Publications (2)
Publication Number | Publication Date |
---|---|
CN105898216A CN105898216A (en) | 2016-08-24 |
CN105898216B true CN105898216B (en) | 2019-01-15 |
Family
ID=56704861
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610231442.6A Active CN105898216B (en) | 2016-04-14 | 2016-04-14 | A kind of number method of counting carried out using unmanned plane |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105898216B (en) |
Families Citing this family (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106940789B (en) * | 2017-03-10 | 2021-08-31 | 广东数相智能科技有限公司 | Quantity counting method, system and device based on video identification |
CN108513642B (en) * | 2017-07-31 | 2021-08-27 | 深圳市大疆创新科技有限公司 | Image processing method, unmanned aerial vehicle, ground console and image processing system thereof |
TWI632528B (en) * | 2017-09-15 | 2018-08-11 | 林永淵 | System and method for unmanned aircraft image analysis |
CN108388838B (en) * | 2018-01-26 | 2021-07-09 | 重庆邮电大学 | Unmanned aerial vehicle ground crowd monitoring system and monitoring method |
CN108877228B (en) * | 2018-08-31 | 2021-04-09 | 辽宁博昊土地科技发展有限公司 | A unmanned aerial vehicle for scenic spot guides |
CN109765931B (en) * | 2019-01-31 | 2021-03-16 | 交通运输部天津水运工程科学研究所 | Near-infrared video automatic navigation method suitable for breakwater inspection unmanned aerial vehicle |
CN109934148A (en) * | 2019-03-06 | 2019-06-25 | 华瑞新智科技(北京)有限公司 | A kind of real-time people counting method, device and unmanned plane based on unmanned plane |
CN110033452B (en) * | 2019-04-17 | 2021-06-08 | 北京云量互动科技有限公司 | Refrigerator shelf commodity statistical method and system based on image processing technology |
EP3748602A1 (en) * | 2019-06-03 | 2020-12-09 | Sony Corporation | Monitoring vehicle movement for traffic risk mitigation |
CN110235802A (en) * | 2019-07-08 | 2019-09-17 | 北京派克盛宏电子科技有限公司 | A kind of live pig amount of exercise statistical method |
CN110246174A (en) * | 2019-07-08 | 2019-09-17 | 北京派克盛宏电子科技有限公司 | A kind of farm live pig method of counting |
CN112327893A (en) * | 2019-08-05 | 2021-02-05 | 旭日蓝天(武汉)科技有限公司 | AI (Artificial intelligence) technology-based unmanned aerial vehicle flight control front-end people counting system and method |
CN111553904B (en) * | 2020-04-29 | 2022-11-22 | 厦门大学 | Unmanned aerial vehicle-based regional people counting method and system |
CN111696127A (en) * | 2020-06-02 | 2020-09-22 | 中国联合网络通信集团有限公司 | Locust plague analysis and early warning system and method |
TWI727906B (en) * | 2020-10-27 | 2021-05-11 | 台灣物聯網股份有限公司 | Method for calculating regional visitor information |
CN112381065B (en) * | 2020-12-07 | 2024-04-05 | 福建天创信息科技有限公司 | Face positioning method and terminal |
CN113947732B (en) * | 2021-12-21 | 2022-05-31 | 北京航空航天大学杭州创新研究院 | Aerial visual angle crowd counting method based on reinforcement learning image brightness adjustment |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102289817A (en) * | 2011-07-19 | 2011-12-21 | 中国科学院自动化研究所 | pedestrian counting method based on group context |
US20120307042A1 (en) * | 2011-06-02 | 2012-12-06 | Hon Hai Precision Industry Co., Ltd. | System and method for controlling unmanned aerial vehicle |
CN105404894A (en) * | 2015-11-03 | 2016-03-16 | 湖南优象科技有限公司 | Target tracking method used for unmanned aerial vehicle and device thereof |
CN105447459A (en) * | 2015-11-18 | 2016-03-30 | 上海海事大学 | Unmanned plane automation detection target and tracking method |
-
2016
- 2016-04-14 CN CN201610231442.6A patent/CN105898216B/en active Active
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20120307042A1 (en) * | 2011-06-02 | 2012-12-06 | Hon Hai Precision Industry Co., Ltd. | System and method for controlling unmanned aerial vehicle |
CN102289817A (en) * | 2011-07-19 | 2011-12-21 | 中国科学院自动化研究所 | pedestrian counting method based on group context |
CN105404894A (en) * | 2015-11-03 | 2016-03-16 | 湖南优象科技有限公司 | Target tracking method used for unmanned aerial vehicle and device thereof |
CN105447459A (en) * | 2015-11-18 | 2016-03-30 | 上海海事大学 | Unmanned plane automation detection target and tracking method |
Also Published As
Publication number | Publication date |
---|---|
CN105898216A (en) | 2016-08-24 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105898216B (en) | A kind of number method of counting carried out using unmanned plane | |
CN106203265B (en) | A kind of Construction Fugitive Dust Pollution source monitors automatically and coverage forecasting system and method | |
US10902566B2 (en) | Methods for agronomic and agricultural monitoring using unmanned aerial systems | |
CN104168455B (en) | A kind of space base large scene camera system and method | |
CN106657910B (en) | A kind of panoramic video monitoring method of electricity substation | |
CN103716594B (en) | Panorama splicing linkage method and device based on moving target detecting | |
CN107504957A (en) | The method that three-dimensional terrain model structure is quickly carried out using unmanned plane multi-visual angle filming | |
WO2019244153A1 (en) | Device, system, and method of computer vision, object tracking, image analysis, and trajectory estimation | |
WO2020221284A1 (en) | Unmanned aerial vehicle monitoring method and system for basin-wide flood scene | |
CN105527969B (en) | A kind of mountain garden belt investigation and monitoring method based on unmanned plane | |
CN106485655A (en) | A kind of taken photo by plane map generation system and method based on quadrotor | |
CN104036236B (en) | A kind of face gender identification method based on multiparameter exponential weighting | |
WO2019100219A1 (en) | Output image generation method, device and unmanned aerial vehicle | |
CN108154110A (en) | A kind of intensive people flow amount statistical method based on the detection of the deep learning number of people | |
CN109801301A (en) | A kind of automatic collection method of tile work progress msg based on BIM and computer vision | |
CN103729620B (en) | A kind of multi-view pedestrian detection method based on multi-view Bayesian network | |
CN106156714A (en) | The Human bodys' response method merged based on skeletal joint feature and surface character | |
CN108897342A (en) | For the positioning and tracing method and system of the civilian multi-rotor unmanned aerial vehicle fast moved | |
Shen et al. | Person tracking and frontal face capture with uav | |
CN108460361A (en) | A kind of crop monitoring device and method | |
CN107808402A (en) | Scaling method, multicamera system and the terminal device of multicamera system | |
CN108734655A (en) | The method and system that aerial multinode is investigated in real time | |
CN105892638A (en) | Virtual reality interaction method, device and system | |
KR102262382B1 (en) | large area space information system for image acquistion base Unmanned Aerial Vehicle | |
CN110378292A (en) | Three dimension location system and method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |