CN113269127A - Face recognition and pedestrian re-recognition monitoring method and system for real-time automatic database building - Google Patents
Face recognition and pedestrian re-recognition monitoring method and system for real-time automatic database building Download PDFInfo
- Publication number
- CN113269127A CN113269127A CN202110648829.2A CN202110648829A CN113269127A CN 113269127 A CN113269127 A CN 113269127A CN 202110648829 A CN202110648829 A CN 202110648829A CN 113269127 A CN113269127 A CN 113269127A
- Authority
- CN
- China
- Prior art keywords
- pedestrian
- face
- recognition
- real
- time
- 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
- 238000012544 monitoring process Methods 0.000 title claims abstract description 34
- 238000000034 method Methods 0.000 title claims abstract description 26
- 238000001514 detection method Methods 0.000 claims abstract description 34
- 230000005540 biological transmission Effects 0.000 claims description 3
- 230000001360 synchronised effect Effects 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 4
- 230000002159 abnormal effect Effects 0.000 abstract description 2
- 238000012795 verification Methods 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 238000000605 extraction Methods 0.000 description 2
- 238000003780 insertion Methods 0.000 description 2
- 230000037431 insertion Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012549 training 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
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
- G06V20/53—Recognition of crowd images, e.g. recognition of crowd congestion
Abstract
The invention relates to a face recognition and pedestrian re-recognition monitoring method and system for automatically building a bank in real time, wherein the method comprises the following steps: step S1: decoding the video stream transmitted by the camera in real time into image frames; step S2: simultaneously carrying out face detection and pedestrian detection on the image; and step S3: and matching the corresponding pedestrian according to the face. The invention can automatically and effectively monitor and manage the personnel in a complex scene for a long time without manually marking and building a library, and can effectively determine the ID and early warning strangers in the scene with variable angles of the personnel and variable characteristics of the personnel in the monitoring scene; the workload of manual checking and monitoring can be reduced to a great extent, abnormal personnel can be tracked and chased in real time, the safety of the area is guaranteed, and a certain deterrent effect is achieved.
Description
Technical Field
The invention relates to the technical field of security and protection monitoring, in particular to a face recognition and pedestrian re-recognition monitoring method and system for automatically building a warehouse in real time.
Background
The face recognition and pedestrian re-recognition are the application of the computer vision technology in the security protection field, and the face recognition is as follows: detecting a face in the image, identifying face features, outputting a corresponding ID, and identifying pedestrians: and (4) searching the pictures of the pedestrians to be inquired in other monitoring.
Patent document 2 discloses a pedestrian re-identification method, a pedestrian re-identification device, and computer equipment, and the technical scheme is as follows: determining the category information of the pedestrian picture to be inquired; the category information is used for representing the category of the human body part contained in the pedestrian picture to be inquired; selecting comparison matched with the category information from comparison libraries corresponding to all pre-established designated human body parts as a target comparison library; and calculating the similarity between the pedestrian picture to be inquired and the target data in the target comparison library, and carrying out pedestrian re-identification on the pedestrian picture to be inquired according to the calculated similarity. However, this solution has the following disadvantages: the comparison library corresponding to each designated human body part needs to be established in advance, manual marking is needed, and features of monitoring scenes with long periods, such as residential quarter residents, clothes hairstyles of people and the like, can be randomly changed, so that the pedestrian re-identification library needs to be manually updated.
Documents of the prior art
Patent document
Patent document 1: CN109359548A gazette
Patent document 2: CN111832361A gazette
Disclosure of Invention
Problems to be solved by the invention
The invention aims at the following problems in the prior art: the existing personnel monitoring scheme is characterized in that personnel are often monitored through a single face recognition method or a pedestrian re-recognition method, the face recognition method can only recognize the face exposure condition, the pedestrian re-recognition method needs manual auxiliary labeling because the pedestrian features are not unique, and the re-recognition only has a short-term effect due to the fact that the appearance features of the same pedestrian can change along with the change of time, so that the face recognition and pedestrian re-recognition monitoring method for real-time automatic warehouse building is provided.
Means for solving the problems
In order to achieve the above object, the present invention provides a method for monitoring face recognition and pedestrian re-recognition by automatically building a bank in real time, which comprises the following steps:
step S1: decoding the video stream transmitted by the camera in real time into image frames;
step S2: simultaneously carrying out face detection and pedestrian detection on the image; and
step S3: and matching the corresponding pedestrian according to the face.
Preferably, the step S3 includes:
and if the face is detected, searching the closest pedestrian detection result, wherein the searching rule is to calculate the intersection ratio of the face frame and all detected pedestrian frames, and find out the pedestrian with the maximum intersection ratio and the face frame positioned at the top of the pedestrian frames as the matched pedestrian.
Preferably, the step S3 includes:
and judging whether the maximum similarity face ID is larger than a first preset threshold value.
Preferably, the step S3 includes:
and recording the ID, the camera position and the time into a database, extracting the pedestrian feature, and replacing the existing feature vector node of the ID in the pedestrian feature library with the current feature vector node.
Preferably, the step S3 includes:
and recording the position and time of the camera corresponding to the ID to a database, and adding the characteristics to a face database.
Preferably, the step S3 includes:
and for the detected pedestrians, eliminating the pedestrian frames capable of detecting the human faces, sending the remaining pedestrian detection results into a pedestrian re-identification model, extracting the characteristics of the pedestrians, and searching the characteristics and the ID corresponding to the maximum similarity in a pedestrian re-identification library.
Preferably, the step S3 includes:
and judging whether the maximum similarity ID is larger than a second preset threshold value.
Preferably, the step S3 includes:
recording the ID, the position of the camera and the time to a database, adding an unidentified cache queue, early warning strangers, recording the time and the camera, and searching again after waiting for updating the pedestrian re-identification database.
Preferably, the step S3 includes:
and if the pedestrian re-identification library is updated, searching the maximum similarity of the updating part of the pedestrian re-identification library, and recording the ID, the camera position and the time to the database if the maximum similarity is greater than a first preset threshold.
The invention also provides a face recognition and pedestrian re-recognition monitoring system for automatically building a bank in real time, which comprises:
the real-time transmission module decodes the video stream transmitted by the camera in real time into image frames;
the synchronous detection module is used for simultaneously carrying out face detection and pedestrian detection on the image; and
and the face matching module is used for matching the corresponding pedestrian according to the face.
ADVANTAGEOUS EFFECTS OF INVENTION
The invention adopts the modes of face recognition and pedestrian re-recognition to carry out detection and tracking, and can play an effective monitoring role aiming at the situations of face fuzzy shielding and back-to-back under the monitoring scene.
The invention adopts a mode of combining pedestrian re-recognition and face recognition to simultaneously detect faces and pedestrians, builds a library by face features and endows IDs to the pedestrians corresponding to the faces so as to track the corresponding pedestrians, forms a pedestrian re-recognition library with the same ID and multiple angles, sends the pedestrian re-recognition library into a pedestrian re-recognition model to perform real-time training and then updates the pedestrian re-recognition model, does not need to manually mark pedestrian pictures, and can flexibly update the pedestrian re-recognition library for the condition of pedestrian feature change.
Drawings
FIG. 1 is a flow chart of a method for monitoring face recognition and pedestrian re-recognition by real-time automatic library building.
Fig. 2 is a schematic diagram of a face recognition and pedestrian re-recognition monitoring system for real-time automatic library building.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without inventive effort based on the embodiments of the present invention, are within the scope of the present invention. It should be further emphasized here that the following embodiments provide preferred embodiments, and that the various aspects (embodiments) may be used in combination or cooperation with each other.
As shown in fig. 1, the present invention provides a method for monitoring face recognition and pedestrian re-recognition by automatically building a bank in real time, which comprises the following steps:
step S1: decoding the video stream transmitted by the camera in real time into image frames;
step S2: performing face detection and pedestrian detection on the image at the same time, wherein a Retinaface face detection model is used for face detection, a Yolov5 detection model is used for pedestrian detection, and an hnsw algorithm is adopted for library construction, searching and insertion replacement;
step S3: matching corresponding pedestrians according to the faces;
the step S3 includes the following substeps:
step S31: if the face is detected, searching (matching) the closest pedestrian detection result, wherein the searching rule is to calculate the intersection and comparison of a face frame and all detected pedestrian frames, find out the pedestrian with the largest intersection and comparison and the face frame positioned at the top of the pedestrian frames as the matched pedestrian, extract the face features by using an Arcface algorithm model, the feature vector dimension is 512 dimensions, and search the face ID with the largest similarity in the existing face feature library;
step S32: judging whether the maximum similarity face ID is larger than a first preset threshold value, wherein the first preset threshold value is set to be 0.4, if so, going to the step S33, otherwise, going to the step S34;
step S33: the method comprises the steps of considering a person corresponding to the ID, recording the ID, the position of a camera and time to a database, and extracting pedestrian features, wherein an AGW _ R50-ibn model in fast-reiD is adopted in the step, the features corresponding to the ID in a pedestrian feature library are updated, the dimension of a feature vector is 2048, and the specific method is to replace the existing feature vector node of the ID in the pedestrian feature library with the current feature vector node;
step S34: recording the position and time of a camera corresponding to the ID into a database, and adding features into a face library, wherein the specific mode is to insert feature vector nodes into the existing face library, warn strangers, extract pedestrian features, add the pedestrian features into a pedestrian feature library, and the specific mode is to insert the feature vector nodes into the existing pedestrian library;
step S35: for the detected pedestrian, eliminating pedestrian frames which can detect the human face in S33 and S34, sending the remaining pedestrian detection results into a pedestrian re-identification model, extracting the characteristics of the pedestrian, and searching the characteristics and the ID corresponding to the maximum similarity in a pedestrian re-identification library, wherein the ID is equivalent to the human face ID, and the same ID refers to the same person;
step S36: judging whether the maximum similarity ID is greater than a second preset threshold, wherein the second preset threshold may be set to 0.7, if so, proceeding to step S38, and if not, proceeding to step S39;
step S37: recording the ID, the camera position and the time to a database;
step S38: adding an unidentified cache queue, early warning strangers, recording time and a camera, and searching again after waiting for updating a pedestrian re-identification library;
step S39: if the pedestrian re-recognition library has been updated in step S34, the maximum similarity of the updated portion of the pedestrian re-recognition library is searched, and if it is greater than the first preset threshold, the ID, the camera position, and the time are recorded in the database.
The invention also provides a face recognition and pedestrian re-recognition monitoring system for automatically building a bank in real time, wherein the system 1 comprises:
a real-time transmission module 11, which decodes the video stream transmitted by the camera in real time into image frames;
the synchronous detection module 12 is used for simultaneously carrying out face detection and pedestrian detection on the image, wherein a Retinaface face detection model is used for face detection, a Yolov5 detection model is used for pedestrian detection, and an hnsw algorithm is adopted for library construction, searching and insertion replacement; and
a face matching module 13 for matching the corresponding pedestrian according to the face;
the face matching module comprises the following sub-modules:
the first search module searches the maximum similarity face ID in the existing face feature library, specifically, if a face is detected, the closest pedestrian detection result is searched (matched), the search rule is to solve the intersection and comparison between a face frame and all detected pedestrian frames, find out the pedestrian with the largest intersection and comparison and the face frame positioned at the top of the pedestrian frames as the matched pedestrian, extract the face feature by using an Arcface algorithm model, the feature vector dimension is 512 dimensions, and search the maximum similarity face ID in the existing face feature library;
the first judgment module is used for judging whether the face ID with the maximum similarity is larger than a first preset threshold value, and the first preset threshold value is set to be 0.4;
the first feature extraction module is used for replacing the existing feature vector node of the ID in the pedestrian feature library with the current feature vector node, and specifically comprises the following steps: recording the ID, the position of a camera and the time of the camera to a database, extracting the characteristics of the pedestrian, updating the characteristics of the ID corresponding to the pedestrian characteristic library by adopting an AGW _ R50-ibn model in fast-reiD, wherein the dimension of the characteristic vector is 2048, and replacing the existing characteristic vector node of the ID in the pedestrian characteristic library with the current characteristic vector node;
the second feature extraction module inserts the feature vector nodes into the existing pedestrian library, and specifically comprises: recording the position and time of a camera corresponding to the ID into a database, adding features into a face library, inserting feature vector nodes into the existing face library, early warning strangers, extracting pedestrian features, adding the pedestrian features into a pedestrian feature library, and inserting the feature vector nodes into the existing pedestrian library;
the maximum similarity searching module searches the features and the ID corresponding to the maximum similarity in the pedestrian re-identification library, and specifically comprises the following steps: for detected pedestrians, eliminating the pedestrian frames capable of detecting the faces, sending the remaining pedestrian detection results into a pedestrian re-identification model, extracting the characteristics of the pedestrians, searching the characteristics and the ID corresponding to the maximum similarity in a pedestrian re-identification library, wherein the ID is equivalent to the face ID, and the same ID refers to the same person;
a second judgment module, configured to judge whether the maximum similarity ID is greater than a second preset threshold, where the second preset threshold may be set to 0.7;
a first recording module: it records the ID, camera position and time to a database;
and the re-searching module is added into the unidentified cache queue, pre-warns strangers, records time and a camera, and performs re-searching after waiting for updating the pedestrian re-identification library.
Compared with the prior art, the invention can automatically and effectively monitor and manage the personnel in a complex scene for a long time without manually marking and building a library, and can effectively determine the ID and early warning strangers in the scene with variable angles of the personnel and variable characteristics of the pedestrians in the monitoring scene; the workload of manual checking and monitoring can be reduced to a great extent, abnormal personnel can be tracked and chased in real time, the safety of the area is guaranteed, and a certain deterrent effect is achieved.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A face recognition and pedestrian re-recognition monitoring method for automatically building a library in real time is characterized by comprising the following steps:
step S1: decoding the video stream transmitted by the camera in real time into image frames;
step S2: simultaneously carrying out face detection and pedestrian detection on the image; and
step S3: and matching the corresponding pedestrian according to the face.
2. The method for monitoring face recognition and pedestrian re-recognition in real-time automatic database building according to claim 1, wherein the step S3 includes:
and if the face is detected, searching the closest pedestrian detection result, wherein the searching rule is to calculate the intersection ratio of the face frame and all detected pedestrian frames, and find out the pedestrian with the maximum intersection ratio and the face frame positioned at the top of the pedestrian frames as the matched pedestrian.
3. The method for monitoring face recognition and pedestrian re-recognition in real-time automatic database building according to claim 2, wherein the step S3 includes:
and judging whether the maximum similarity face ID is larger than a first preset threshold value.
4. The method for monitoring face recognition and pedestrian re-recognition in real-time automatic database building according to claim 3, wherein the step S3 includes:
and recording the ID, the camera position and the time into a database, extracting the pedestrian feature, and replacing the existing feature vector node of the ID in the pedestrian feature library with the current feature vector node.
5. The method for monitoring face recognition and pedestrian re-recognition in real-time automatic database building according to claim 4, wherein the step S3 includes:
and recording the position and time of the camera corresponding to the ID to a database, and adding the characteristics to a face database.
6. The method for monitoring face recognition and pedestrian re-recognition in real-time automatic database building according to claim 5, wherein the step S3 includes:
and for the detected pedestrians, eliminating the pedestrian frames capable of detecting the human faces, sending the remaining pedestrian detection results into a pedestrian re-identification model, extracting the characteristics of the pedestrians, and searching the characteristics and the ID corresponding to the maximum similarity in a pedestrian re-identification library.
7. The method for monitoring face recognition and pedestrian re-recognition in real-time automatic database building according to claim 6, wherein the step S3 includes:
and judging whether the maximum similarity ID is larger than a second preset threshold value.
8. The method for monitoring face recognition and pedestrian re-recognition in real-time automatic database building according to claim 7, wherein the step S3 includes:
recording the ID, the position of the camera and the time to a database, adding an unidentified cache queue, early warning strangers, recording the time and the camera, and searching again after waiting for updating the pedestrian re-identification database.
9. The method for monitoring face recognition and pedestrian re-recognition in real-time automatic database building according to claim 8, wherein the step S3 includes:
and if the pedestrian re-identification library is updated, searching the maximum similarity of the updating part of the pedestrian re-identification library, and recording the ID, the camera position and the time to the database if the maximum similarity is greater than a first preset threshold.
10. The utility model provides a real-time face identification and pedestrian who builds storehouse heavily discern monitored control system which characterized in that includes:
the real-time transmission module decodes the video stream transmitted by the camera in real time into image frames;
the synchronous detection module is used for simultaneously carrying out face detection and pedestrian detection on the image; and
and the face matching module is used for matching the corresponding pedestrian according to the face.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110648829.2A CN113269127B (en) | 2021-06-10 | 2021-06-10 | Face recognition and pedestrian re-recognition monitoring method and system for real-time automatic database establishment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110648829.2A CN113269127B (en) | 2021-06-10 | 2021-06-10 | Face recognition and pedestrian re-recognition monitoring method and system for real-time automatic database establishment |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113269127A true CN113269127A (en) | 2021-08-17 |
| CN113269127B CN113269127B (en) | 2024-04-02 |
Family
ID=77234777
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110648829.2A Active CN113269127B (en) | 2021-06-10 | 2021-06-10 | Face recognition and pedestrian re-recognition monitoring method and system for real-time automatic database establishment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113269127B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113780284A (en) * | 2021-09-17 | 2021-12-10 | 焦点科技股份有限公司 | Logo detection method based on target detection and metric learning |
| CN114067421A (en) * | 2022-01-17 | 2022-02-18 | 广东中运信息科技有限公司 | Personnel duplicate removal identification method, storage medium and computer equipment |
| CN114743155A (en) * | 2022-03-10 | 2022-07-12 | 慧之安信息技术股份有限公司 | Mall pedestrian recognition method based on combination of face recognition and pedestrian re-recognition |
| CN113780284B (en) * | 2021-09-17 | 2024-04-19 | 焦点科技股份有限公司 | Logo detection method based on target detection and metric learning |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109344787A (en) * | 2018-10-15 | 2019-02-15 | 浙江工业大学 | A kind of specific objective tracking identified again based on recognition of face and pedestrian |
| CN110609920A (en) * | 2019-08-05 | 2019-12-24 | 华中科技大学 | Pedestrian hybrid search method and system in video monitoring scene |
| CN111523383A (en) * | 2020-03-19 | 2020-08-11 | 创新奇智(北京)科技有限公司 | Non-perception face recognition system and method based on pedestrian ReID |
| CN112801018A (en) * | 2021-02-07 | 2021-05-14 | 广州大学 | Cross-scene target automatic identification and tracking method and application |
-
2021
- 2021-06-10 CN CN202110648829.2A patent/CN113269127B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109344787A (en) * | 2018-10-15 | 2019-02-15 | 浙江工业大学 | A kind of specific objective tracking identified again based on recognition of face and pedestrian |
| CN110609920A (en) * | 2019-08-05 | 2019-12-24 | 华中科技大学 | Pedestrian hybrid search method and system in video monitoring scene |
| CN111523383A (en) * | 2020-03-19 | 2020-08-11 | 创新奇智(北京)科技有限公司 | Non-perception face recognition system and method based on pedestrian ReID |
| CN112801018A (en) * | 2021-02-07 | 2021-05-14 | 广州大学 | Cross-scene target automatic identification and tracking method and application |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113780284A (en) * | 2021-09-17 | 2021-12-10 | 焦点科技股份有限公司 | Logo detection method based on target detection and metric learning |
| CN113780284B (en) * | 2021-09-17 | 2024-04-19 | 焦点科技股份有限公司 | Logo detection method based on target detection and metric learning |
| CN114067421A (en) * | 2022-01-17 | 2022-02-18 | 广东中运信息科技有限公司 | Personnel duplicate removal identification method, storage medium and computer equipment |
| CN114067421B (en) * | 2022-01-17 | 2022-04-19 | 广东中运信息科技有限公司 | Personnel duplicate removal identification method, storage medium and computer equipment |
| CN114743155A (en) * | 2022-03-10 | 2022-07-12 | 慧之安信息技术股份有限公司 | Mall pedestrian recognition method based on combination of face recognition and pedestrian re-recognition |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113269127B (en) | 2024-04-02 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11704936B2 (en) | Object tracking and best shot detection system | |
| CN106991395B (en) | Information processing method and device and electronic equipment | |
| JP6854881B2 (en) | Face image matching system and face image search system | |
| CN109558810B (en) | Target person identification method based on part segmentation and fusion | |
| US10037467B2 (en) | Information processing system | |
| US8266174B2 (en) | Behavior history retrieval apparatus and behavior history retrieval method | |
| CN109934104A (en) | The pedestrian retrieval method and system across camera lens identified again based on pedestrian | |
| US20180081908A1 (en) | Image Search Apparatus, System, and Method | |
| CN106355154B (en) | Method for detecting frequent passing of people in surveillance video | |
| CN113269127A (en) | Face recognition and pedestrian re-recognition monitoring method and system for real-time automatic database building | |
| CN107230267A (en) | Intelligence In Baogang Kindergarten based on face recognition algorithms is registered method | |
| CN113887445A (en) | Method and system for identifying standing and loitering behaviors in video | |
| CN103514429A (en) | Method for detecting specific part of object and image processing equipment | |
| US20110280442A1 (en) | Object monitoring system and method | |
| CN108830204B (en) | Method for detecting abnormality in target-oriented surveillance video | |
| CN111914649A (en) | Face recognition method and device, electronic equipment and storage medium | |
| US20240111835A1 (en) | Object detection systems and methods including an object detection model using a tailored training dataset | |
| CN110826390A (en) | Video data processing method based on face vector characteristics | |
| US10902249B2 (en) | Video monitoring | |
| CN116419059A (en) | Automatic monitoring method, device, equipment and medium based on behavior label | |
| CN111814690A (en) | Target re-identification method and device and computer readable storage medium | |
| KR102016262B1 (en) | Image processing method and system for smoky environment | |
| KR20200062065A (en) | Method and apparatus for prodicting moving path | |
| CN115272967A (en) | Cross-camera pedestrian real-time tracking and identifying method, device and medium | |
| Seidenari et al. | Non-parametric anomaly detection exploiting space-time features |
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 | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: Room 711c, 7 / F, block a, building 1, yard 19, Ronghua Middle Road, Beijing Economic and Technological Development Zone, Daxing District, Beijing 102600 Applicant after: Beijing Zhongke Flux Technology Co.,Ltd. Address before: Room 711c, 7 / F, block a, building 1, yard 19, Ronghua Middle Road, Beijing Economic and Technological Development Zone, Daxing District, Beijing 102600 Applicant before: Beijing Ruixin high throughput technology Co.,Ltd. |
|
| GR01 | Patent grant | ||
| GR01 | Patent grant |