CN111639597A - Detection method of flag-raising touring event - Google Patents
Detection method of flag-raising touring event Download PDFInfo
- Publication number
- CN111639597A CN111639597A CN202010477150.7A CN202010477150A CN111639597A CN 111639597 A CN111639597 A CN 111639597A CN 202010477150 A CN202010477150 A CN 202010477150A CN 111639597 A CN111639597 A CN 111639597A
- Authority
- CN
- China
- Prior art keywords
- area
- flag
- pedestrian
- red flag
- event
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 48
- 238000000034 method Methods 0.000 claims abstract description 25
- 230000006399 behavior Effects 0.000 claims abstract description 10
- 238000013135 deep learning Methods 0.000 claims abstract description 7
- 238000012958 reprocessing Methods 0.000 claims abstract description 7
- 230000001788 irregular Effects 0.000 claims description 6
- 230000011218 segmentation Effects 0.000 claims description 2
- 238000012545 processing Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000001914 filtration Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 238000012795 verification Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/20—Movements or behaviour, e.g. gesture recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N20/00—Machine learning
-
- 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
- G06V20/42—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items of sport video content
-
- 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
- G06V20/44—Event detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V2201/00—Indexing scheme relating to image or video recognition or understanding
- G06V2201/07—Target detection
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Evolutionary Computation (AREA)
- Mathematical Physics (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Psychiatry (AREA)
- Social Psychology (AREA)
- Human Computer Interaction (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a method for detecting a flag-raising touring event, which comprises the following steps: 1) acquiring a video; executing the step 2; 2) carrying out pedestrian detection based on deep learning on each frame of image of the video to obtain a pedestrian region of each frame of image; executing the step 3; 3) reprocessing the pedestrian area to obtain an area needing red flag detection, and obtaining a red flag area in the area needing red flag detection; executing the step 4; 4) judging whether the red flag area has a pedestrian flag-raising behavior; if yes, executing step 6; otherwise, executing step 5; 5) the red flag area of the current frame is stored as a historical event; executing the step 5; 6) and uploading the flag tour event. According to the detection method of the flag-raising touring event, the judgment of the position of the pedestrian is completed by utilizing a deep learning detection technology, so that the accuracy is improved; the red flag detection module is used for obtaining information related to illegal meetings, so that normal meetings and illegal meetings can be distinguished.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a method for detecting a flag-raising touring event.
Background
At present, the gathering event of the downstream row in the scene is generally hidden, and the difference from the gathering of the pedestrian of a general non-gathering is small.
The related image processing techniques include the following methods:
CN201710405071.3A A natural language public sentiment analysis method based on big data sentiment analysis is used for estimating and preventing events such as tourist meetings and the like in the aspects of natural language and public sentiment monitoring, belongs to an indirect source from the viewpoint of data source, and the obtained conclusion cannot be directly applied, so that double verification with other data is required.
CN 201110139351 is a high density crowd flow statistical method, which predicts the event of tourist gathering by estimating the crowd flow through pedestrian detection, and the obtained result is also an indirect conclusion, which cannot be directly used and needs to be confirmed twice.
CN201410508575.4A A crowd gathering area detecting method and device, which uses the preset multi-scale rectangle window to traverse the gray scale map to obtain the approximate position, and uses the fine classification accurate position to obtain the position information accurately, but can not obtain whether the gathering is normal or illegal.
CN201510076652.8A the crowd gathering state detection method based on image can obtain more accurate crowd movement information by estimating crowd movement state with direct video data, but because there is no judgment of characteristic articles such as small red flag, it is impossible to accurately distinguish whether normal crowd movement or illegal tourist gathering event.
Accordingly, there is a need for improvements in the art.
Disclosure of Invention
The invention aims to solve the technical problem of providing a detection method of flag-raising touring events with high accuracy;
in order to solve the technical problem, the invention provides a method for detecting a flag-raising touring event, which comprises the following steps:
1) acquiring a video; executing the step 2;
2) carrying out pedestrian detection based on deep learning on each frame of image of the video to obtain a pedestrian region of each frame of image; executing the step 3;
3) reprocessing the pedestrian area to obtain an area needing red flag detection, and obtaining a red flag area in the area needing red flag detection; executing the step 4;
4) judging whether the red flag area has a pedestrian flag-raising behavior; if yes, executing step 6; otherwise, executing step 5;
5) the red flag area of the current frame is stored as a historical event;
6) and uploading the flag tour event.
As an improvement of the detection method of the flagging and touring event of the invention:
the reprocessing in step 3 includes:
3.1) properly expanding the single pedestrian area to obtain an expanded pedestrian area;
3.2) combining the repeated parts of the plurality of expanded pedestrian areas to obtain the pedestrian area.
As a further improvement of the detection method of the flagging and touring event of the invention:
step 4 comprises the following steps: and judging that the pedestrian flag raising behavior exists when the red flag areas exist in the three continuous frames of images.
As a further improvement of the detection method of the flagging and touring event of the invention:
step 4 comprises the following steps: combining the red flag area of the previous frame and the red flag area of the current frame, calculating the contact ratio of the areas, judging the same continuous event as a flag-raising travel event when the contact ratio is greater than a threshold value, and executing the step 6; otherwise, step 5 is executed.
As a further improvement of the detection method of the flagging and touring event of the invention:
in step 3, the exact position of the red flag is obtained using a segmentation and detection algorithm.
As a further improvement of the detection method of the flagging and touring event of the invention:
the red flag area is obtained by detecting the red flag area by utilizing the form and color information.
As a further improvement of the detection method of the flagging and touring event of the invention:
and judging the red flag pixel points by utilizing the color information, wherein if the red flag pixel area is an irregular area and is similar to a rectangle, the irregular area is the red flag area.
As a further improvement of the detection method of the flagging and touring event of the invention:
in step 6, generating the flag-holding travel event into structured information; and uploading the structured information to an event display platform.
The detection method of the flag-holding touring event has the technical advantages that:
1) the pedestrian position judgment is completed by utilizing the deep learning detection technology, and the accuracy is improved;
2) and the red flag detection module is utilized to obtain the information related to the illegal meetings, so that the normal meetings and the illegal meetings can be distinguished.
3) The method is suitable for more illegal types, and the effective early warning rate is greatly improved.
Drawings
The following describes embodiments of the present invention in further detail with reference to the accompanying drawings.
Fig. 1 is a schematic flow chart of a method for detecting a flag-holding travel event according to the present invention.
Detailed Description
The invention will be further described with reference to specific examples, but the scope of the invention is not limited thereto.
The embodiment 1 discloses a detection system for a flag-raising touring event, which comprises a pedestrian detection module, a red flag detection module, a flag-raising judgment module and a video front and rear frame filtering module.
A pedestrian detection module:
and adopting a target detection algorithm based on the existing deep learning technology, and carrying out pedestrian detection on the monitored image by using a pedestrian detection model to obtain a pedestrian region in the monitored image.
The red flag detection module:
the module obtains the area needing the red flag detection by reprocessing the single pedestrian area output before. And obtaining a red flag area by adopting methods such as red flag detection and the like for the area needing red flag detection.
The reprocessing includes: 1) properly expanding the single pedestrian area (for example, expanding the width and the height to 1.1 times) to obtain an expanded pedestrian area; 2) and combining the repeated parts of the plurality of expanded pedestrian areas.
The red flag detection method is to detect a red flag region by using morphological and color information, for example: and judging the red flag pixel points by utilizing the color information in the area, wherein if the red flag pixel area is an irregular area and is similar to a rectangle, the irregular area is the red flag area.
A flag raising judgment module:
the module filters the output of the red flag detection module, removes negative samples, and then judges whether the region has a pedestrian flag raising behavior or not by combining with a pedestrian region.
Judging as invalid data if the red flag area overlaps with the 1/4 area located near the central point of the pedestrian area; the area 1/4 near the center point is: the width and height of this region are both pedestrian regions 1/2, area 1/4, with the center points coinciding.
And (5) sequentially processing all pedestrian areas, and counting the number of people with flag raising behaviors. And when the total number of flag-raising pedestrians is higher than a set threshold value, determining that a flag-raising touring event occurs, otherwise, determining that normal behaviors do not carry out any treatment.
A video front and rear frame filtering module:
the characteristic that the difference of the front frame and the rear frame of the video is not large is utilized, a threshold value is set to be 3 generally, and the event is reported when continuous multiple frames occur simultaneously, so that the accuracy rate of the event is further improved.
The detection method of the flag-holding touring event comprises the following steps:
1) firstly, accessing a video; executing the step 2;
2) carrying out pedestrian detection based on deep learning on each frame of image of the accessed video to obtain a pedestrian region; executing the step 3;
3) the pedestrian area of the current frame is reprocessed to obtain an area needing red flag detection, and the area needing red flag detection obtains a red flag area by methods such as red flag detection and the like; executing the step 4;
4) judging whether the red flag area has a pedestrian flag-lifting behavior, and judging that a flag-lifting wandering event occurs when the pedestrian flag-lifting behavior occurs for a plurality of continuous frames at the same time;
combining the historical events (the red flag areas of the last two frames) and the red flag area of the current frame, calculating the contact ratio of the areas, judging the same continuous event as a flag-holding travel event when the contact ratio is greater than a threshold value, and executing the step 6; otherwise, executing step 5;
5) the red flag area of the current frame is stored as a historical event;
6) generating the flag-holding travel event into structured information; and uploading the structured information to an event display platform.
Finally, it is also noted that the above-mentioned lists merely illustrate a few specific embodiments of the invention. It is obvious that the invention is not limited to the above embodiments, but that many variations are possible. All modifications which can be derived or suggested by a person skilled in the art from the disclosure of the present invention are to be considered within the scope of the invention.
Claims (8)
1. The detection method of the flag-raising touring event is characterized by comprising the following steps: the method comprises the following steps:
1) acquiring a video; executing the step 2;
2) carrying out pedestrian detection based on deep learning on each frame of image of the video to obtain a pedestrian region of each frame of image; executing the step 3;
3) reprocessing the pedestrian area to obtain an area needing red flag detection, and obtaining a red flag area in the area needing red flag detection; executing the step 4;
4) judging whether the red flag area has a pedestrian flag-raising behavior; if yes, executing step 6; otherwise, executing step 5;
5) the red flag area of the current frame is stored as a historical event; executing the step 5;
6) and uploading the flag tour event.
2. The method of claim 1, wherein the method further comprises:
the reprocessing in step 3 includes:
3.1) properly expanding the single pedestrian area to obtain an expanded pedestrian area;
3.2) combining the repeated parts of the plurality of expanded pedestrian areas to obtain the pedestrian area.
3. The method of claim 2, wherein the method further comprises:
step 4 comprises the following steps: and judging that the pedestrian flag raising behavior exists when the red flag areas exist in the three continuous frames of images.
4. The method of claim 3, wherein the method comprises:
step 4 comprises the following steps: combining the red flag area of the previous frame and the red flag area of the current frame, calculating the contact ratio of the areas, judging the same continuous event as a flag-raising travel event when the contact ratio is greater than a threshold value, and executing the step 6; otherwise, step 5 is executed.
5. The method of claim 4, wherein the method comprises:
in step 3, the exact position of the red flag is obtained using a segmentation and detection algorithm.
6. The method of claim 5, wherein the method further comprises:
the red flag area is obtained by detecting the red flag area by utilizing the form and color information.
7. The method of claim 6, wherein the method comprises:
and judging the red flag pixel points by utilizing the color information, wherein if the red flag pixel area is an irregular area and is similar to a rectangle, the irregular area is the red flag area.
8. The method of claim 7, wherein the method further comprises:
in step 6, generating the flag-holding travel event into structured information; and uploading the structured information to an event display platform.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010477150.7A CN111639597A (en) | 2020-05-29 | 2020-05-29 | Detection method of flag-raising touring event |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010477150.7A CN111639597A (en) | 2020-05-29 | 2020-05-29 | Detection method of flag-raising touring event |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111639597A true CN111639597A (en) | 2020-09-08 |
Family
ID=72331662
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010477150.7A Pending CN111639597A (en) | 2020-05-29 | 2020-05-29 | Detection method of flag-raising touring event |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111639597A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112232316A (en) * | 2020-12-11 | 2021-01-15 | 科大讯飞(苏州)科技有限公司 | Crowd gathering detection method and device, electronic equipment and storage medium |
| CN113963316A (en) * | 2021-11-25 | 2022-01-21 | 上海闪马智能科技有限公司 | Target event determination method and device, storage medium and electronic device |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109446989A (en) * | 2018-10-29 | 2019-03-08 | 上海七牛信息技术有限公司 | Crowd massing detection method, device and storage medium |
| CN109558792A (en) * | 2018-10-11 | 2019-04-02 | 成都三零凯天通信实业有限公司 | Method and system for detecting Internet logo content based on samples and features |
| CN110502768A (en) * | 2018-05-18 | 2019-11-26 | 郑州大学 | The emulation mode and system of crowd movement in a kind of political rally scene |
-
2020
- 2020-05-29 CN CN202010477150.7A patent/CN111639597A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110502768A (en) * | 2018-05-18 | 2019-11-26 | 郑州大学 | The emulation mode and system of crowd movement in a kind of political rally scene |
| CN109558792A (en) * | 2018-10-11 | 2019-04-02 | 成都三零凯天通信实业有限公司 | Method and system for detecting Internet logo content based on samples and features |
| CN109446989A (en) * | 2018-10-29 | 2019-03-08 | 上海七牛信息技术有限公司 | Crowd massing detection method, device and storage medium |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112232316A (en) * | 2020-12-11 | 2021-01-15 | 科大讯飞(苏州)科技有限公司 | Crowd gathering detection method and device, electronic equipment and storage medium |
| CN112232316B (en) * | 2020-12-11 | 2021-03-26 | 科大讯飞(苏州)科技有限公司 | Crowd gathering detection method and device, electronic equipment and storage medium |
| CN113963316A (en) * | 2021-11-25 | 2022-01-21 | 上海闪马智能科技有限公司 | Target event determination method and device, storage medium and electronic device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112800860B (en) | High-speed object scattering detection method and system with coordination of event camera and visual camera | |
| Chen et al. | Vehicle detection, tracking and classification in urban traffic | |
| CN104392468B (en) | Based on the moving target detecting method for improving visual background extraction | |
| CN101470809B (en) | Moving object detection method based on expansion mixed gauss model | |
| CN110188807A (en) | Tunnel pedestrian target detection method based on cascade super-resolution network and improvement Faster R-CNN | |
| WO2008025092A1 (en) | Method and system of identifying one or more features represented in a plurality of sensor acquired data sets | |
| CN105744232A (en) | Method for preventing power transmission line from being externally broken through video based on behaviour analysis technology | |
| CN109615862A (en) | Road vehicle movement of traffic state parameter dynamic acquisition method and device | |
| CN111626170B (en) | Image recognition method for railway side slope falling stone intrusion detection | |
| JP6653361B2 (en) | Road marking image processing apparatus, road marking image processing method, and road marking image processing program | |
| CN107844772A (en) | A kind of motor vehicle automatic testing method based on movable object tracking | |
| CN111553214A (en) | Method and system for detecting smoking behavior of driver | |
| CN107909009A (en) | A kind of obstacle detection method and device based on road surface study | |
| CN109508576A (en) | A kind of abnormal driving behavioral value method, apparatus and electronic equipment | |
| CN111639597A (en) | Detection method of flag-raising touring event | |
| CN109684986A (en) | A kind of vehicle analysis method and system based on automobile detecting following | |
| CN111563468A (en) | Driver abnormal behavior detection method based on attention of neural network | |
| JP2019121356A (en) | Interference region detection apparatus and method, and electronic apparatus | |
| CN108960181B (en) | Black smoke vehicle detection method based on multi-scale block LBP and hidden Markov model | |
| CN116228756B (en) | Method and system for detecting bad points of camera in automatic driving | |
| CN112991769A (en) | Traffic volume investigation method and device based on video | |
| CN115719362B (en) | High-altitude parabolic detection method, system, equipment and storage medium | |
| CN116012822B (en) | Fatigue driving identification method and device and electronic equipment | |
| Muniruzzaman et al. | Deterministic algorithm for traffic detection in free-flow and congestion using video sensor | |
| Sridevi et al. | Automatic generation of traffic signal based on traffic volume |
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 |