CN106600965B - Traffic flow morning and evening peak period automatic identifying method based on sharpness - Google Patents
Traffic flow morning and evening peak period automatic identifying method based on sharpness Download PDFInfo
- Publication number
- CN106600965B CN106600965B CN201710041037.2A CN201710041037A CN106600965B CN 106600965 B CN106600965 B CN 106600965B CN 201710041037 A CN201710041037 A CN 201710041037A CN 106600965 B CN106600965 B CN 106600965B
- Authority
- CN
- China
- Prior art keywords
- data
- traffic flow
- morning
- sharpness
- evening
- 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
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
Landscapes
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Traffic Control Systems (AREA)
Abstract
The present invention proposes a kind of traffic flow morning and evening peak period automatic identifying method based on sharpness, comprising: obtains the acquisition time and traffic flow speed in traffic flow data；It is pre-processed；Draw m- speed curve diagram for the moment；When calculating in m- speed curve diagram each data point sharpness；Size sequence is carried out to the sharpness of all data points, takes wherein candidate collection of the sharpness numerical values recited in preceding a certain proportion of data point, as early evening peak extreme point；Retain the data point for just setting period morning and evening in candidate collection, obtains morning peak candidate subset and evening peak candidate subset；The smallest data point of traffic flow speed in morning peak candidate subset and evening peak candidate subset is searched for respectively, and the acquisition time of two data points recorded is shifted to an earlier date forward as origin using acquisition time and backward delay certain time obtains peak period morning and evening.The present invention can more effectively realize the peak period morning and evening automatic identification of road traffic, and method is simple, easy to implement.
Description
Technical field
The present invention relates to intelligent transportation fields, in particular to be a kind of peak period traffic flow morning and evening based on sharpness
Automatic identifying method.
Background technique
Multi-period timing controlled is to be divided into several periods for one day according to telecommunication flow information, and the different periods takes not
Same optimization control scheme.It is multi-period time-controlled on condition that traffic slot classifying rationally.Up to the present have a large amount of
Traffic slot division methods.
Most of traffic slot divisions are carried out based on clustering algorithm at this stage, belong to learning model, such as hierarchical clustering,
Genetic cluster, artificial immune clustering, fuzzy C-means clustering etc..The thought of clustering algorithm is that the similarity of homogeneous object is made to the greatest extent may be used
Can be high, the similarity of different objects is as low as possible, the automatic identificationization point of traffic slot may be implemented, but existing for traffic
There is also some problems for the clustering method of Time segments division.
For partition clustering method such as fuzzy C-means clustering, K mean cluster etc., clusters number need to be first set, but in reality
The number of cluster is clusters number that is unknown, being obtained according to subjective experience, has stronger subjectivity, is also easy to produce unreasonable
Time segments division；Although hierarchy clustering method does not have to that clusters number is first arranged, singular value can generate larger shadow to Clustering Effect
It rings, and algorithm complexity is higher, due to traffic flow data strong real-time, Outlier Data point is more, and data volume is big, so such method
It carries out slightly sensitive when traffic slot division.
Summary of the invention
Technical problem to be solved by the invention is to provide a kind of, and peak period traffic flow morning and evening based on sharpness is automatic
Recognition methods, the more effective peak period morning and evening automatic identification for realizing road traffic, has method simple, easy to implement etc. excellent
Point.
To solve the above problems, the present invention proposes a kind of traffic flow morning and evening peak period automatic identification side based on sharpness
Method, comprising the following steps:
S1: the acquisition time and traffic flow speed in traffic flow data are obtained, as initial data；
S2: pre-processing the initial data, obtains pretreated data；
S3: using the pretreated data as coordinate data, m- speed curve diagram for the moment is drawn；
S4: when calculating described in m- speed curve diagram each data point sharpness；
S5: it is descending to the sharpness of all data points to be ranked up, take wherein sharpness numerical values recited preceding certain
The data point of ratio, the candidate collection as early evening peak extreme point；
S6: retain the data point for just setting period morning and evening in the candidate collection, obtain morning peak candidate subset and evening
Peak candidate subset；
S7: the smallest data of traffic flow speed in the morning peak candidate subset and evening peak candidate subset are searched for respectively
Point, the acquisition time of two data points recorded, using acquisition time as origin forward in advance and backward delay certain time
The obtained period is peak period morning and evening.
According to one embodiment of present invention, in the step S1, GPS data or ground induction coil data are acquired as traffic
Flow data is stored, and reads the acquisition time in traffic flow data and traffic flow speed.
According to one embodiment of present invention, in the step S2, successively to the traffic flow speed data in initial data
Missing values completion, noise points deleting and smoothing processing are carried out, pretreated traffic flow speed, the pretreated friendship are obtained
Through flow velocity and original acquisition time are as the pretreated data.
According to one embodiment of present invention, it in the step S2, for the traffic flow speed data in initial data, adopts
Completion is carried out to data missing values with linear interpolation, followed by lower order polynomial expressions Fitted logistic according to noise reduction process is carried out, then
Data are carried out by moving average method smooth.
According to one embodiment of present invention, in the step S3, for pretreated data, using acquisition time as
Horizontal axis draws the when m- speed curve diagram under rectangular coordinate system using traffic flow speed as the longitudinal axis.
According to one embodiment of present invention, in the step S4, the data point P in access time-speed curve diagrami,
Take data point PiEach k number strong point in left and right is as supporting zone, with most left data point Pi-kTo data point PiAnd data point PiExtremely
Most right data point Pi+kFolded angle, θiAs data point PiStrut angle, PiPi-k、PiPi+kFor support arm, according to the following formula
(a) and (b) determines data point PiSharpness sharp (pi):
According to one embodiment of present invention, in step S5, by when m- speed curve diagram in all data points sharpness
It is descending to be ranked up, take sharpness numerical values recited to come preceding 10% data point, the candidate as early evening peak extreme point
Set.
According to one embodiment of present invention, in step S6, in candidate collection retention time between 6 points to 11 points
Data point, obtain morning peak candidate subset；In data point of in the candidate collection retention time between 15 points to 20 points, obtain
To evening peak candidate subset.
According to one embodiment of present invention, in step S7, it is minimum that traffic flow speed is searched for from morning peak candidate subset
Data point, record the corresponding acquisition time T of this data pointam, then the morning peak period is acquisition time TamRespectively forwardly in advance partly
Hour to backward delay half an hour；The smallest data point of traffic flow speed is searched for from evening peak candidate subset, records this data
The corresponding acquisition time T of pointpm, then the evening peak period is acquisition time TpmIt is small to backward delay half respectively forwardly to shift to an earlier date half an hour
When.
After adopting the above technical scheme, the present invention has the advantages that compared with prior art
By the way that image processing techniques to be incorporated in the analysis and processing of traffic data, innovation is proposed using at image for the first time
Sharpness index in reason identifies peak period traffic flow morning and evening；
Clusters number that no setting is required, and computation complexity is low, calculates at low cost, the easy easily realization of method, strong operability.
Detailed description of the invention
Fig. 1 is that the process of the traffic flow morning and evening peak period automatic identifying method based on sharpness of the embodiment of the present invention is shown
It is intended to；
Fig. 2 be being drawn according to pretreated data of the embodiment of the present invention made of when m- speed curve diagram；
Fig. 3 is the schematic illustration of the calculating sharpness of the embodiment of the present invention；
Fig. 4 is the when m- speed curve diagram for identifying peak period morning and evening of the embodiment of the present invention.
Specific embodiment
In order to make the foregoing objectives, features and advantages of the present invention clearer and more comprehensible, with reference to the accompanying drawing to the present invention
Specific embodiment be described in detail.
In the following description, numerous specific details are set forth in order to facilitate a full understanding of the present invention.But the present invention can be with
Much it is different from other way described herein to implement, those skilled in the art can be without prejudice to intension of the present invention the case where
Under do similar popularization, therefore the present invention is not limited to the specific embodiments disclosed below.
Referring to Fig. 1, the traffic flow morning and evening peak period automatic identifying method based on sharpness of the embodiment of the present invention includes
Following steps:
S1: the acquisition time and traffic flow speed in traffic flow data are obtained, as initial data；
S2: pre-processing the initial data, obtains pretreated data；
S3: using the pretreated data as coordinate data, m- speed curve diagram for the moment is drawn；
S4: when calculating described in m- speed curve diagram each data point sharpness；
S5: have to the sharpness of all data points arrive greatly it is small be ranked up, take wherein sharpness numerical values recited preceding certain
The data point of ratio, the candidate collection as early evening peak extreme point；
S6: retain the data point for just setting period morning and evening in the candidate collection, obtain morning peak candidate subset and evening
Peak candidate subset；
S7: the smallest data of traffic flow speed in the morning peak candidate subset and evening peak candidate subset are searched for respectively
Point, the acquisition time of two data points recorded, using acquisition time as origin forward in advance and backward delay certain time
The obtained period is peak period morning and evening.
The traffic flow morning and evening peak period automatic identifying method based on sharpness of the embodiment of the present invention can be applied to city
Control of traffic and road etc. is suitble to the traffic flow Time segments division of any the intensive traffic section, is especially suitable for that early evening peak can occur
In crowded section of highway.The traffic flow morning and evening peak period automatic identifying method based on sharpness of the embodiment of the present invention is carried out below
The description of specific embodiment.
In step sl, the acquisition time and traffic flow speed in traffic flow data are obtained, as initial data.Traffic flow
Data are the data or further statistics obtained by capture and detection apparatus for the driving conditions acquisition testing of the intensive traffic section
The data of acquisition, such as may include acquisition time, the magnitude of traffic flow, speed, density etc., specifically without limitation.In this implementation
In example, the acquisition time and traffic flow speed in traffic flow data are got, is not necessarily to other complex datas, it is convenient to obtain,
It calculates simple.
In one embodiment, in step sl, GPS data or ground induction coil data can be acquired as traffic flow data
It is stored, but without limitation, the traffic flow data that can also be obtained using other devices or other methods, these data can
To be ready-made acquired historical data, or collection in worksite can be gone to obtain latest data.Traffic flow data can store to
In the memory module of computer, in step S1 implementation procedure, by traffic flow data acquisition time and traffic flow speed read
Enter into memory, for subsequent step use.
Then step S2 is executed, is needed to guarantee the effect of subsequent processing to original speed to improve the quality of data
Degree pre-processes the initial data obtained in step S1, obtains pretreated data according to being pre-processed.
In one embodiment, in step s 2, missing values are successively carried out to the traffic flow speed data in initial data
Completion, noise points deleting and smoothing processing, obtain pretreated traffic flow speed, pretreated traffic flow speed with it is original
Acquisition time as pretreated data.
Specifically, in step s 2, for the traffic flow speed data in initial data, being lacked using linear interpolation to data
Mistake value carries out completion, followed by lower order polynomial expressions Fitted logistic according to noise reduction process is carried out, then passes through moving average method pair
Data carry out smooth.It is appreciated that can also be using other missing values completions, noise points deleting and smooth processing mode to friendship
Through flow velocity is pre-processed, and guarantees the effect of follow-up data processing.
Then execute step S3, for pretreated data, can carry out described point referring to Fig. 2 under the same coordinate system
It draws, after the data point drawn out is connected into curve, just delineates m- speed curve diagram when a width, convert data to
The image of usable image processing mode processing.
In one embodiment, in step S3, for pretreated data, using acquisition time as horizontal axis, with traffic
Flow velocity degree draws the when m- speed curve diagram under rectangular coordinate system as the longitudinal axis.For example, the traffic flow of an acquisition time
Speed and the acquisition time, should after navigating at corresponding coordinate respectively as the Y axis coordinate value and X axis coordinate value of rectangular coordinate system
Just a data point is formed at coordinate, and so whole acquisition times and traffic flow speed are all corresponded to and obtain one in coordinate system
Series data point, is depicted as curve.
Then step S4 is executed, the sharpness of each data point in when calculating m- speed curve diagram, for several before beginning
A point and last several points can not calculate its sharpness.Since the number of these points is generally smaller, this hair will not influence
The final calculated result of bright middle method.
Referring to Fig. 3, the local configuration enlarged drawing of m- rate curve, can be adapted for any non-end in curve graph when being
The data point of point.In one embodiment, in step s 4, the data point P in access time-speed curve diagrami, take the data
Point PiEach k number strong point in left and right is as supporting zone, with most left data point Pi-kTo data point PiAnd data point PiTo most right data
Point Pi+kFolded angle, θiAs data point PiStrut angle, peripheral solid line is contour line, and stain represents pixel, and dotted line is
Pi-k, Pi, Pi+kThis 3 points of circular arcs being fitted to, O point are the center of circle, PiPi-k、PiPi+kFor support arm, the value of k generally can be 3
~5, but without limitation, data point Pi, Pi-k, Pi+kIt is approximately 3 points on one section of circular arc, the interval between them is very small,
Thus may be assumed that | PiPi-k|=| PiPi+k|.(a) and (b) determines data point P according to the following formulaiSharpness sharp (pi):
sharp(pi) indicate strut angle acuity, sharp (pi) value is bigger, illustrate that the angle is more sharp.According to above formula
(a) when and (b) is calculated on m- speed curve diagram every bit sharpness.Point is carried out to the data point obtained according to coordinate data
Acutance calculates.
Then step S5 is executed, it is descending to the sharpness of all data points after obtaining the sharpness of all data points
It is ranked up, takes wherein Candidate Set of the sharpness numerical values recited in preceding a certain proportion of data point, as early evening peak extreme point
It closes.
In one embodiment, in step s 5, by when m- speed curve diagram in all data points sharpness by greatly to
It is small to be ranked up, take sharpness numerical values recited to come preceding 10% data point, the candidate collection as early evening peak extreme point.
Then step S6 is executed, the data interval for just setting period morning and evening is retained in candidate collection, obtains morning peak time
Select subset and evening peak candidate subset.Just setting period morning and evening can be and determine based on experience value, but the specific time is unrestricted.
In one embodiment, in step s 6, the data in candidate collection retention time between 6 points to 11 points
Point obtains morning peak candidate subset；In data point of in the candidate collection retention time between 15 points to 20 points, late height is obtained
Peak candidate subset.Due to data point image conversion, image can be carried out the processing such as replicating, thus successively image can be carried out
Interception obtains morning peak candidate subset and evening peak candidate subset respectively.
Then step S7 is executed, searches for traffic flow speed in morning peak candidate subset and evening peak candidate subset respectively most
Small data point, the acquisition time of two data points recorded, using acquisition time as origin forward in advance and backward delay
The period that certain time obtains is peak period morning and evening.The smallest data point of traffic flow speed shows that traffic is most crowded at this time,
For top time point, this point is nearby peak period morning and evening, thus is chosen after the data point to the timing that preceding extends back
Between can determine peak period morning and evening.
Referring to Fig. 4, in one embodiment, in the step s 7, traffic flow speed is searched for most from morning peak candidate subset
Small data point records the corresponding acquisition time T of this data pointam, then the morning peak period is acquisition time TamRespectively forwardly shift to an earlier date
Half an hour to backward delay half an hour；The smallest data point of traffic flow speed is searched for from evening peak candidate subset, records this number
The corresponding acquisition time T in strong pointpm, then the evening peak period is acquisition time TpmRespectively forwardly shift to an earlier date half an hour to backward delay half
Hour.
Although the present invention is disclosed as above with preferred embodiment, it is not for limiting claim, any this field
Technical staff without departing from the spirit and scope of the present invention, can make possible variation and modification, therefore of the invention
Protection scope should be subject to the range that the claims in the present invention are defined.
Claims (8)
1. a kind of traffic flow morning and evening peak period automatic identifying method based on sharpness, which comprises the following steps:
S1: the acquisition time and traffic flow speed in traffic flow data are obtained, as initial data；
S2: pre-processing the initial data, obtains pretreated data；
S3: using the pretreated data as coordinate data, m- speed curve diagram for the moment is drawn；
S4: when calculating described in m- speed curve diagram each data point sharpness；
S5: it is descending to the sharpness of all data points to be ranked up, take wherein sharpness numerical values recited in preceding certain proportion
Data point, the candidate collection as early evening peak extreme point；
S6: retain the data point for just setting period morning and evening in the candidate collection, obtain morning peak candidate subset and evening peak
Candidate subset；
S7: searching for the smallest data point of traffic flow speed in the morning peak candidate subset and evening peak candidate subset respectively,
The acquisition time for recording two obtained data points is obtained with backward delay certain time forward in advance using acquisition time as origin
Period be peak period morning and evening；
Wherein, in the step S4, the data point P in access time-speed curve diagrami, take data point PiEach k number in left and right
Strong point is as supporting zone, with most left data point Pi-kTo data point PiAnd data point PiTo most right data point Pi+kFolded angle
θiAs data point PiStrut angle, PiPi-k、PiPi+kFor support arm, (a) and (b) determines data point P according to the following formulaiPoint
Acutance sharp (pi):
2. the traffic flow morning and evening peak period automatic identifying method based on sharpness as described in claim 1, which is characterized in that
In the step S1, acquires GPS data or ground induction coil data are stored as traffic flow data, and read traffic flow data
In acquisition time and traffic flow speed.
3. the traffic flow morning and evening peak period automatic identifying method based on sharpness as described in claim 1, which is characterized in that
In the step S2, missing values completion, noise points deleting and smooth are successively carried out to the traffic flow speed data in initial data
Processing obtains pretreated traffic flow speed, and the pretreated traffic flow speed and original acquisition time are as institute
State pretreated data.
4. the traffic flow morning and evening peak period automatic identifying method based on sharpness as claimed in claim 3, which is characterized in that
In the step S2, for the traffic flow speed data in initial data, completion is carried out to data missing values using linear interpolation,
Followed by lower order polynomial expressions Fitted logistic according to noise reduction process is carried out, then data are carried out by moving average method smooth.
5. the traffic flow morning and evening peak period automatic identifying method based on sharpness as described in claim 1, which is characterized in that
In the step S3, for pretreated data, using acquisition time as horizontal axis, using traffic flow speed as the longitudinal axis, draw
When m- speed curve diagram under rectangular coordinate system.
6. the traffic flow morning and evening peak period automatic identifying method based on sharpness as described in claim 1, which is characterized in that
In step S5, by when m- speed curve diagram in all data points sharpness it is descending be ranked up, take sharpness numerical value big
It is small come preceding 10% data point, the candidate collection as early evening peak extreme point.
7. the traffic flow morning and evening peak period automatic identifying method based on sharpness as described in claim 1, which is characterized in that
In step S6, in data point of in the candidate collection retention time between 6 points to 11 points, morning peak candidate subset is obtained；?
Data point of the retention time between 15 points to 20 points, obtains evening peak candidate subset in candidate collection.
8. the traffic flow morning and evening peak period automatic identifying method based on sharpness as described in claim 1, which is characterized in that
In step S7, the smallest data point of traffic flow speed is searched for from morning peak candidate subset, records the corresponding acquisition of this data point
Time Tam, then the morning peak period is acquisition time TamRespectively forwardly shift to an earlier date half an hour to backward delay half an hour；It is waited from evening peak
The search the smallest data point of traffic flow speed in subset is selected, the corresponding acquisition time T of this data point is recordedpm, then evening peak period
For acquisition time TpmRespectively forwardly shift to an earlier date half an hour to backward delay half an hour.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710041037.2A CN106600965B (en) | 2017-01-19 | 2017-01-19 | Traffic flow morning and evening peak period automatic identifying method based on sharpness |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710041037.2A CN106600965B (en) | 2017-01-19 | 2017-01-19 | Traffic flow morning and evening peak period automatic identifying method based on sharpness |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106600965A CN106600965A (en) | 2017-04-26 |
| CN106600965B true CN106600965B (en) | 2018-12-14 |
Family
ID=58585388
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710041037.2A Active CN106600965B (en) | 2017-01-19 | 2017-01-19 | Traffic flow morning and evening peak period automatic identifying method based on sharpness |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106600965B (en) |
Families Citing this family (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109871412A (en) * | 2018-12-26 | 2019-06-11 | 航天科工广信智能技术有限公司 | Lane flow analysis method based on K-Means cluster |
| CN109697849B (en) * | 2018-12-26 | 2020-06-16 | 航天科工广信智能技术有限公司 | Intelligent traffic time interval dividing method based on moving average algorithm |
| CN113593262B (en) * | 2019-11-14 | 2022-09-27 | 北京百度网讯科技有限公司 | Traffic signal control method, traffic signal control device, computer equipment and storage medium |
| CN112819325B (en) * | 2021-01-29 | 2024-07-05 | 北京嘀嘀无限科技发展有限公司 | Rush hour determination method, apparatus, electronic device, and storage medium |
| CN113399784B (en) * | 2021-07-09 | 2022-08-12 | 武汉武重机床有限公司 | Workpiece processing control method, device, equipment and storage medium |
Family Cites Families (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP3580475B2 (en) * | 1998-09-14 | 2004-10-20 | 矢崎総業株式会社 | Perimeter monitoring device |
| JP4075585B2 (en) * | 2002-11-25 | 2008-04-16 | 株式会社デンソー | Vehicle speed control device and program |
| CN102985956B (en) * | 2010-06-29 | 2017-03-08 | 本田技研工业株式会社 | Traffic congestion prediction display method |
| US8803956B2 (en) * | 2011-12-20 | 2014-08-12 | King Fahd University Of Petroleum And Minerals | Computer vision-based valve control system and method |
| CN102890866B (en) * | 2012-09-17 | 2015-01-21 | 上海交通大学 | Traffic flow speed estimation method based on multi-core support vector regression machine |
| CN103761430B (en) * | 2014-01-10 | 2017-07-07 | 安徽科力信息产业有限责任公司 | A kind of road network peak period recognition methods based on Floating Car |
| CN106295683A (en) * | 2016-08-01 | 2017-01-04 | 上海理工大学 | A kind of outlier detection method of time series data based on sharpness |
-
2017
- 2017-01-19 CN CN201710041037.2A patent/CN106600965B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN106600965A (en) | 2017-04-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106600965B (en) | Traffic flow morning and evening peak period automatic identifying method based on sharpness | |
| CN110298404B (en) | Target tracking method based on triple twin Hash network learning | |
| CN111460984B (en) | Global lane line detection method based on key points and gradient equalization loss | |
| CN109766868B (en) | Real scene pedestrian-sheltered detection network based on body key point detection and detection method thereof | |
| CN111738055B (en) | Multi-category text detection system and bill form detection method based on same | |
| Chen et al. | Learning-based spatio-temporal vehicle tracking and indexing for transportation multimedia database systems | |
| Lee et al. | Place recognition using straight lines for vision-based SLAM | |
| CN103164694A (en) | Method for recognizing human motion | |
| CN107862275A (en) | Human bodys' response model and its construction method and Human bodys' response method | |
| WO2013091369A1 (en) | Multi-target segmentation and tracking method based on depth image | |
| CN104463117A (en) | Sample collection method and system used for face recognition and based on video | |
| CN111104903A (en) | Depth perception traffic scene multi-target detection method and system | |
| CN111027377B (en) | Double-flow neural network time sequence action positioning method | |
| CN107633226A (en) | A kind of human action Tracking Recognition method and system | |
| CN109670462A (en) | Continue tracking across panorama based on the aircraft of location information | |
| CN116030396B (en) | Accurate segmentation method for video structured extraction | |
| CN114677323A (en) | Semantic vision SLAM positioning method based on target detection in indoor dynamic scene | |
| CN112509009B (en) | Target tracking method based on natural language information assistance | |
| CN112287906B (en) | Template matching tracking method and system based on depth feature fusion | |
| CN105513083A (en) | PTAM camera tracking method and device | |
| CN112861931A (en) | Multi-level change detection method based on difference attention neural network | |
| CN111275732B (en) | Foreground object image segmentation method based on depth convolution neural network | |
| CN114037666B (en) | Shadow detection method assisted by data set expansion and shadow image classification | |
| Xu et al. | License plate recognition system based on deep learning | |
| CN105404682B (en) | A kind of book retrieval method based on digital image content |
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 |