CN109697426A - Flight based on multi-detector fusion shuts down berth detection method - Google Patents
Flight based on multi-detector fusion shuts down berth detection method Download PDFInfo
- Publication number
- CN109697426A CN109697426A CN201811584147.4A CN201811584147A CN109697426A CN 109697426 A CN109697426 A CN 109697426A CN 201811584147 A CN201811584147 A CN 201811584147A CN 109697426 A CN109697426 A CN 109697426A
- Authority
- CN
- China
- Prior art keywords
- berth
- aircraft
- detection
- shut down
- ranging
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/25—Fusion techniques
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Multimedia (AREA)
- Optical Radar Systems And Details Thereof (AREA)
Abstract
The present invention relates to a kind of flights based on multi-detector fusion to shut down berth detection method, comprising: the object ranging for shut down berth calculates distance measurement result, obtains ranging and detects the lower confidence level shut down and have aircraft on berth;Shooting image is analyzed in the shoot on location for shut down berth, obtains the confidence level shut down under image detection and have aircraft on berth;There are the confidence calculations of aircraft to shut down the objective degrees of confidence for having aircraft on berth on berth according to shutting down under the confidence level and image detection for having aircraft on the priori accuracy rate of image detection, the priori accuracy rate of ranging detection, the lower shutdown berth of ranging detection; when objective degrees of confidence is greater than the judgment threshold of setting; confirmation, which is shut down on berth, aircraft; when objective degrees of confidence is less than the judgment threshold of setting, confirmation, which is shut down on berth, does not have aircraft.The present invention uses modelling method of probabilistic, merge ranging detection and image detection as a result, avoid single method there are the shortcomings that, to obtain more preferably detection effect.
Description
Technical field
The present invention relates to a kind of flights based on multi-detector fusion to shut down berth detection method.
Background technique
Aircraft shuts down one that berth is particularly important in airport control, and flight, which shuts down berth state, influences the utilization of seat in the plane
Rate, and then directly affect the handling capacity on airport.Current main mode is artificial detection, furthermore there is the detection based on laser ranging
With the detection based on video analysis.
The statistics of most of airport aircrafts shutdown berth times is still by the way of artificial at present, not only at high cost and effect
Rate is low.A small number of airports are to shut down berth installation laser ranging system to be detected, and principle is by measurement light source point and front
The distance of object judges the presence or absence of aircraft, this method have strong real-time, do not influenced by Changes in weather, that equipment is cheap etc. is excellent
Point, but rate of false alarm is serious, requires harshness to decorating position, and many airports early stages arrange this equipment as auxiliary detection, but because
Rate of false alarm height is only put into the auxiliary system of practical application, is still required manual intervention.On the other hand, almost each in airport
Monitor camera is all installed in seat in the plane, stops situation by video come observation airplane.Recently, which it is automatic to be also used for shutdown berth
Detection detects aircraft parked state using video analysis method, such as the algorithm based on moving object detection, the algorithm speed compared with
Fastly, detection effect is preferable under good image-forming condition, but has the disadvantage in that and can not differentiate whether target detected is winged
Machine may cause wrong report when occurring other moving targets (such as oversize vehicle) in image;Moving target can only be detected, aircraft is worked as
It remains static down, can not judge its parked state;The setting of algorithm parameter is more sensitive to conditions such as weather, illumination, to winged
The berth accuracy rate of identification of machine is still unable to reach intelligent requirements, it is also necessary to human assistance inspection.
Summary of the invention
In order to solve the above technical problems, the present invention provides a kind of flights based on multi-detector fusion to shut down berth detection
Method is desirably to obtain the higher object detection results of accuracy.
The technical scheme is that a kind of flight based on multi-detector fusion shuts down berth detection method, including
The object ranging for shut down berth, calculates distance measurement result, and obtaining lower shut down on berth of ranging detection has
The confidence level of aircraft,
Shooting image is analyzed in the shoot on location for shut down berth, and obtaining to shut down under image detection on berth has
The confidence level of aircraft,
The objective degrees of confidence shut down and have aircraft on berth is calculated according to following equation:
When objective degrees of confidence is greater than the judgment threshold of setting, confirmation, which is shut down on berth, aircraft, when objective degrees of confidence is small
When the judgment threshold of setting, confirmation, which is shut down on berth, does not have aircraft,
Wherein,
PfuseTo shut down the objective degrees of confidence for having aircraft on berth, PridlFor the priori accuracy rate of image detection, PrilaFor
The priori accuracy rate of ranging detection, PlaThe confidence level for having aircraft on lower shutdown berth, P are detected for rangingdlTo stop under image detection
There is the confidence level of aircraft on machine berth.
It can carry out shutting down the object ranging in berth using laser ranging system.
The confidence level for having aircraft on the lower shutdown berth of ranging detection can be calculated according to following equation:
Wherein, L is the detected object distance of current ranging, Ls be detection when having an aircraft stop on shutdown berth away from
From standard value, Lm is the detecting distance standard value shut down when not having aircraft stop on berth.
The Ls and Lm can be predefined.
The Ls and Lm can be obtained under good ranging testing conditions by experiment.Aircraft is rested in and is shut down on berth
Common location, away from ranging is carried out under the various states such as ranging detection device proximal most position and highest distance position, obtain corresponding real
Object distance is surveyed, to survey average value or the minimum value of object distance as Ls;There is no aircraft and other influences on shutting down berth
Ranging is carried out in the state of the object of distance measurement result, corresponding actual measurement object distance is obtained, to survey the average value of object distance
Or maximum value is Lm.
(it can also not flown according to ranging detection device away from the actual range for shutting down berth and away from front object correlation
Machine berthing time is away from the object detected) actual range determine Ls and Lm.
The priori accuracy rate of ranging detection can be determined by experiment.
Carry out many experiments for example, having under aircraft and the not two states of aircraft on shutting down berth, with practical survey
Ranging detection is carried out away from identical mode is detected, and artificial nucleus are correctly secondary with testing result to the correctness of ranging testing result
For number divided by the total degree of detection, the quotient of acquisition is the priori accuracy rate of ranging detection.
It is preferred that being analyzed using YOLO algorithm shooting image, obtaining to shut down under image detection on berth has setting for aircraft
Reliability.
When multiple cells and/or bounding box detect target aircraft, it is preferred to use non-maxima suppression algorithm obtains
Maximum bounding box, using the respective confidence of maximum bounding box obtained as image detection under shut down berth on have
The confidence level of aircraft.
The priori accuracy rate of image detection can be determined by experiment.
Carry out many experiments for example, having under aircraft and the not two states of aircraft on shutting down berth, with practical figure
Image detection is carried out as detecting identical mode, and artificial nucleus are correctly secondary with testing result to the correctness of image detection result
For number divided by the total degree of detection, the quotient of acquisition is the priori accuracy rate of image detection.
The beneficial effects of the present invention are: obtaining target by using modelling method of probabilistic with two methods respectively and existing
Then confidence level is merged using new probability formula, obtain finally judging algorithm, thus avoid single method there are the shortcomings that.
Selection for visible sensation method, it is contemplated that the defect of moving object detection algorithm, by the deep learning method application of current mainstream
It berths detection in aircraft, more can robustly detect the interested aircraft of user (aircraft) this specific objective, no matter fly
Machine is in movement or stationary state and can all be detected, in combination with laser distance measurement method not by the factors shadow such as weather, illumination
Loud advantage, to finally obtain more preferably detection effect.
Detailed description of the invention
Fig. 1 is a kind of flow diagram of the invention.
Specific embodiment
Referring to Fig. 1, the present invention blends physical method and visible sensation method, it is intended to avoid single method there are the shortcomings that.
Selection for visible sensation method, it is contemplated that the defect of moving object detection algorithm, by the deep learning method application of current mainstream
It berths detection in aircraft, one side this method more can robustly detect the interested specific objective-aircraft of user no matter
Aircraft be in movement or stationary state can all be detected, on the other hand combine laser distance measurement method not by weather, illumination etc. because
The advantages of element influences, to finally obtain more preferably detection effect.
The invention solves main problem be two methods fusion.Using modelling method of probabilistic, respectively with two kinds of sides
Method obtains confidence level existing for target, is then merged using new probability formula, and obtaining finally judging to shut down on berth is
The no objective degrees of confidence for having aircraft.
The present invention includes the following steps:
1) laser detection
Laser Distance Measuring Equipment is by optical receiver, optical transmitting set and timer composition, and the course of work is sent out by optical transmitting set
Laser is penetrated, reflective object is encountered and is received by optical receiver, the time of timer and light velocity quadrature are obtained into target range laser
The distance of transmitter, the cost performance of this Laser Distance Measuring Equipment is high, speed is fast, has very high practical value.
Space laser ranging of the present invention is different from cooperative satellite laser ranging (SLR), because flying as measured target
The corner reflector for making laser backtracking is fitted without on machine, it can only be anti-by incident laser by the diffusing characteristic diffuser of target itself
It is mapped on receiver, thereby also reduces the accuracy of laser equipment.
The present invention designs the probabilistic model about laser detection using the characteristic of laser ranging.If Ls is to have aircraft to rest in
Distance threshold of the aircraft away from laser pickoff when seat in the plane, Lm are the distance threshold that laser acquisition is arrived when no aircraft is stopped, the two
Value be previously set.If the object detection distance that present laser ranging obtains is L, defining under laser detection has aircraft on seat in the plane
Confidence level PlaIt is as follows:
PlaValue between zero and one, value is bigger, represent shut down berth on there are the confidence level of aircraft is higher.
2) image recognition of deep learning algorithm
The algorithm of deep learning is just to start to rise in recent years, and the airport overwhelming majority all uses conventional method at present, is based on
The research of the airport target identification of deep learning algorithm is fewer.Deep learning is a kind of algorithm with multi-level fuzzy judgment, is adopted
Model parameter is automatically adjusted with back-propagation algorithm, learns the abstract expression of data out from mass data automatically, by training
Model the position of aircraft in the picture can be identified from input picture automatically, compared to conventional method it is more stable.
The YOLOv3 algorithm that the present invention uses Joseph Redmon and Ali Farhadi to propose[1]As basic frame, have
The advantages that speed is fast, multi-scale prediction, low, versatile background false detection rate.Object detector passes through Microsoft COCO data set[2]
Training obtains, and is able to detect the frequent goals such as aircraft, pedestrian.
Input picture is divided into several cells by YOLO, and then each cell, which is responsible for detecting those central points, falls
Target in the grid.Each cell can predict the confidence level of several bounding boxes and bounding box.Confidence level is mainly wrapped
Containing two aspects, first is that a possibility that this bounding box contains target;Second is that the accuracy of this bounding box.The former contains bounding box
There is a possibility that target to be denoted as Pr, the Pr=0 when the bounding box is background, that is, does not include target;And when the bounding box includes mesh
Pr=1 when mark.The accuracy of bounding box can be with the friendship of prediction block and actual frames and than IOU (Intersection Over
Union it) characterizes, is denoted asTherefore corresponding confidence level can be defined as follows[1]:
Since YOLO has divided the image into several cells, therefore detection target may be cut into many blocks, just
Good wherein to have several cells to detect the target, this, which just will appear a target, multiple bounding boxes, that is, has multiple confidences
Degree.The present invention is to solve a target using non-maxima suppression algorithm NMS (Non Maximum Suppression) master
The problem of by repeated detection.A certain frame is selected first as initial block, then hands over and compares respectively with remaining frame, if handing over simultaneously ratio
Greater than certain threshold value (registration is excessively high), then just rejecting the frame, otherwise maximum frame option is replaced.To all detection block weights
The multiple above process, until obtain unique maximum frame, and by the confidence level P of the framedlAs the confidence level that whether there is target.
Equally, PdlValue between zero and one, value is bigger, and it is higher to represent confidence level existing for aircraft.
3) multi-detector merges
Multi-detector Fusion Model merges laser ranging with the objective degrees of confidence that two methods of deep learning obtain,
Firstly the need of two methods of the priori accuracy rate of estimation, two methods of the objective degrees of confidence of mixed model formula fusion is recycled,
It obtains final aircraft and berths to determine result.
(1) priori accuracy rate is estimated
(a) the priori accuracy rate for estimating laser distance measurement method can carry out n times test, and artificial nucleus are to correct in N test
The number M for detecting target obtains the priori accuracy rate of this methodN should be sufficiently large to guarantee priori standard
The precision of the numerical value of exactness.
(b) the in kind priori accuracy rate Pri of estimating depth learning algorithmdl。
(2) multi-detector fusion detection
(a) to current detection, objective degrees of confidence P is calculated with laser distance measurement method and deep learning method respectivelylaAnd Pdl。
(b) objective degrees of confidence merged using following formula.
(5) judge that flight parked state judges.Work as PfuseWhen > Thr, there is aircraft in seat in the plane, occupied;Otherwise without aircraft, machine
Position is idle, and wherein threshold value Thr should be set as 0.5 under normal circumstances.
Aircraft, flight alleged by the present invention refer to various aircrafts, including commonly called aircraft and flight.
It is disclosed by the invention it is each preferably with optional technological means, unless otherwise indicated and one preferably or can selecting technology hand
Section is that further limiting for another technological means is outer, can form several different technical solutions in any combination.
Bibliography
[1]Joseph Redmon and Ali Farhadi,“YOLOv3:An Incremental Improvement”,
Technical report 2018.
[2]Tsungyi Lin et al.,“Microsoft COCO:Common Objects in Context”,ECCV
2015.
Claims (10)
1. a kind of flight based on multi-detector fusion shuts down berth detection method, including
The object ranging for shut down berth, calculates distance measurement result, and obtaining lower shut down on berth of ranging detection has aircraft
Confidence level,
Shooting image is analyzed in the shoot on location for shut down berth, and obtaining to shut down under image detection on berth has aircraft
Confidence level,
The objective degrees of confidence shut down and have aircraft on berth is calculated according to following equation:
When objective degrees of confidence is greater than the judgment threshold of setting, confirmation, which is shut down on berth, aircraft, sets when objective degrees of confidence is less than
When fixed judgment threshold, confirmation, which is shut down on berth, does not have aircraft,
Wherein,
PfuseTo shut down the objective degrees of confidence for having aircraft on berth, PridlFor the priori accuracy rate of image detection, PrilaFor ranging
The priori accuracy rate of detection, PlaThe confidence level for having aircraft on lower shutdown berth, P are detected for rangingdlTo shut down pool under image detection
There is the confidence level of aircraft on position.
2. the method as described in claim 1, it is characterised in that shut down using laser ranging system the object ranging in berth.
3. the method as described in claim 1, it is characterised in that calculating lower shut down on berth of ranging detection according to following equation has
The confidence level of aircraft:
Wherein, L is the detected object distance of current ranging, and Ls is the detecting distance mark shut down when having aircraft stop on berth
Quasi- value, Lm are the detecting distance standard value shut down when not having aircraft stop on berth.
4. method as claimed in claim 3, it is characterised in that the Ls and Lm passes through experiment under good ranging testing conditions
It obtains, or the actual range according to ranging detection device away from the actual range for shutting down berth and away from front object correlation is true
It is fixed.
5. the method as described in claim 1, it is characterised in that be determined by experiment the priori accuracy rate of ranging detection.
6. method as claimed in claim 5, it is characterised in that have aircraft and the not two states of aircraft on shutting down berth
Lower carry out many experiments carry out ranging detection in a manner of identical with practical ranging detection, and artificial nucleus are to ranging testing result
Correctness, with the correct number of testing result divided by the total degree of detection, the quotient of acquisition is the priori accuracy rate of ranging detection.
7. the method as described in claim 1-6 is any, it is characterised in that analyzed using YOLO algorithm shooting image, obtained
Obtain the confidence level shut down under image detection and have aircraft on berth.
8. the method for claim 7, it is characterised in that when multiple cells and/or bounding box detect target aircraft
When, maximum bounding box is obtained using non-maxima suppression algorithm, with the corresponding confidence of maximum bounding box obtained
It spends as the confidence level for having aircraft under image detection on shutdown berth.
9. the method for claim 7, it is characterised in that be determined by experiment the priori accuracy rate of image detection.
10. method as claimed in claim 9, it is characterised in that have aircraft and the not two states of aircraft on shutting down berth
Lower carry out many experiments carry out image detection in a manner of identical with real image detection, and artificial nucleus are to image detection result
Correctness, with the correct number of testing result divided by the total degree of detection, the quotient of acquisition is the priori accuracy rate of image detection.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811584147.4A CN109697426B (en) | 2018-12-24 | 2018-12-24 | Flight based on multi-detector fusion shuts down berth detection method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201811584147.4A CN109697426B (en) | 2018-12-24 | 2018-12-24 | Flight based on multi-detector fusion shuts down berth detection method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109697426A true CN109697426A (en) | 2019-04-30 |
CN109697426B CN109697426B (en) | 2019-10-18 |
Family
ID=66232766
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201811584147.4A Active CN109697426B (en) | 2018-12-24 | 2018-12-24 | Flight based on multi-detector fusion shuts down berth detection method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109697426B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110309735A (en) * | 2019-06-14 | 2019-10-08 | 平安科技(深圳)有限公司 | Exception detecting method, device, server and storage medium |
CN110544396A (en) * | 2019-08-12 | 2019-12-06 | 南京莱斯信息技术股份有限公司 | Airplane berth guiding equipment |
CN111427374A (en) * | 2020-02-25 | 2020-07-17 | 深圳市镭神智能系统有限公司 | Airplane berth guiding method, device and equipment |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20120155712A1 (en) * | 2010-12-17 | 2012-06-21 | Xerox Corporation | Method for automatic license plate recognition using adaptive feature set |
CN103076877A (en) * | 2011-12-16 | 2013-05-01 | 微软公司 | Interacting with a mobile device within a vehicle using gestures |
CN103390173A (en) * | 2013-07-24 | 2013-11-13 | 佳都新太科技股份有限公司 | Plate number character vote algorithm based on SVM (support vector machine) confidence |
CN104290730A (en) * | 2014-06-20 | 2015-01-21 | 郑州宇通客车股份有限公司 | Radar and video information fusing method applied to advanced emergency brake system |
CN107850895A (en) * | 2015-05-13 | 2018-03-27 | 优步技术公司 | The automatic driving vehicle of operation is assisted by guiding |
CN108009494A (en) * | 2017-11-30 | 2018-05-08 | 中山大学 | A kind of intersection wireless vehicle tracking based on unmanned plane |
-
2018
- 2018-12-24 CN CN201811584147.4A patent/CN109697426B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20120155712A1 (en) * | 2010-12-17 | 2012-06-21 | Xerox Corporation | Method for automatic license plate recognition using adaptive feature set |
CN103076877A (en) * | 2011-12-16 | 2013-05-01 | 微软公司 | Interacting with a mobile device within a vehicle using gestures |
CN103390173A (en) * | 2013-07-24 | 2013-11-13 | 佳都新太科技股份有限公司 | Plate number character vote algorithm based on SVM (support vector machine) confidence |
CN104290730A (en) * | 2014-06-20 | 2015-01-21 | 郑州宇通客车股份有限公司 | Radar and video information fusing method applied to advanced emergency brake system |
CN107850895A (en) * | 2015-05-13 | 2018-03-27 | 优步技术公司 | The automatic driving vehicle of operation is assisted by guiding |
CN108009494A (en) * | 2017-11-30 | 2018-05-08 | 中山大学 | A kind of intersection wireless vehicle tracking based on unmanned plane |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110309735A (en) * | 2019-06-14 | 2019-10-08 | 平安科技(深圳)有限公司 | Exception detecting method, device, server and storage medium |
CN110544396A (en) * | 2019-08-12 | 2019-12-06 | 南京莱斯信息技术股份有限公司 | Airplane berth guiding equipment |
CN111427374A (en) * | 2020-02-25 | 2020-07-17 | 深圳市镭神智能系统有限公司 | Airplane berth guiding method, device and equipment |
Also Published As
Publication number | Publication date |
---|---|
CN109697426B (en) | 2019-10-18 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109697426B (en) | Flight based on multi-detector fusion shuts down berth detection method | |
CN109544932B (en) | Urban road network flow estimation method based on fusion of taxi GPS data and gate data | |
WO2023124383A1 (en) | Vehicle speed measurement method, collision early-warning method, and electronic device | |
US11092444B2 (en) | Method and system for recording landmarks in a traffic environment of a mobile unit | |
CN110501700A (en) | A kind of personnel amount method of counting based on millimetre-wave radar | |
CN108389421B (en) | Parking lot accurate induction system and method based on image re-identification | |
CN109635666A (en) | A kind of image object rapid detection method based on deep learning | |
CN110443196A (en) | Fire-fighting road occupying detection method based on SSIM algorithm | |
CN110490125A (en) | A kind of fueling area service quality detection system detected automatically based on gesture | |
CN109740676A (en) | Object detection moving method based on similar purpose | |
CN109426800A (en) | A kind of method for detecting lane lines and device | |
WO2019198076A1 (en) | Real-time raw data- and sensor fusion | |
CN112927565B (en) | Method, device and system for improving accuracy of comprehensive track monitoring data of apron | |
CN104616502A (en) | License plate identification and positioning system based on combined type vehicle-road video network | |
CN106327880B (en) | A kind of speed recognition methods and its system based on monitor video | |
CN107909601A (en) | A kind of shipping anti-collision early warning video detection system and detection method suitable for navigation mark | |
CN108052887A (en) | A kind of doubtful illegal land automatic recognition system and method for merging SLAM/GNSS information | |
CN115797736B (en) | Training method, device, equipment and medium for target detection model and target detection method, device, equipment and medium | |
CN114089330B (en) | Indoor mobile robot glass detection and map updating method based on depth image restoration | |
CN108229473A (en) | Vehicle annual inspection label detection method and device | |
CN116573017A (en) | Urban rail train running clearance foreign matter sensing method, system, device and medium | |
CN116993681A (en) | Substation inspection defect detection method and system | |
CN114170487A (en) | Vision-based water surface oil stain detection method | |
CN112699748B (en) | Human-vehicle distance estimation method based on YOLO and RGB image | |
KR102616571B1 (en) | System and method for providing road traffic information based on image analysis using artificial intelligence |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |