CN110598551B - Method, device, equipment and medium for improving pedestrian identity recognition efficiency - Google Patents
Method, device, equipment and medium for improving pedestrian identity recognition efficiency Download PDFInfo
- Publication number
- CN110598551B CN110598551B CN201910732754.9A CN201910732754A CN110598551B CN 110598551 B CN110598551 B CN 110598551B CN 201910732754 A CN201910732754 A CN 201910732754A CN 110598551 B CN110598551 B CN 110598551B
- Authority
- CN
- China
- Prior art keywords
- information
- pedestrian
- description information
- camera
- identity
- 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
- 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
-
- 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
- G06V40/161—Detection; Localisation; Normalisation
- G06V40/166—Detection; Localisation; Normalisation using acquisition arrangements
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Collating Specific Patterns (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to the technical field of image recognition, in particular to a method, a device, equipment and a medium for improving the pedestrian identity recognition efficiency, wherein the method comprises the steps of acquiring video data acquired by a common camera and an AI camera; detecting pedestrians from video data acquired by an AI camera and acquiring first description information of the detected pedestrians, and detecting pedestrians from video data acquired by a common camera and acquiring second description information of the detected pedestrians; correspondingly storing the pedestrian and the first description information thereof to form an identity information search library; calculating the similarity between the first description information and the second description information, judging whether the similarity is larger than a preset degree, and if so, associating the second description information with the first description information; and updating the identity information search library according to the first description information and the second description information of the associated pedestrians. Thus, the required pedestrian identity information can be quickly and accurately searched from the identity information search library.
Description
Technical Field
The invention relates to the technical field of image recognition, in particular to a method, a device, equipment and a medium for improving pedestrian identity recognition efficiency.
Background
In the security field, the application of face recognition is mainly realized by newly building an AI camera with face recognition functions such as face snapshot on a sensing node or transmitting a video stream of a common camera to a back-end server.
The occupancy of AI cameras is relatively low compared to tens of millions of public security monitoring cameras across the country. And the construction cost is high and the period is long. And the video stream is transmitted to the back end for realizing, for face recognition application, the monitoring camera has a plurality of old devices. Due to the technical factors and the erection height in the past, the quality of the collected face images is low, the resolution ratio of the face images is low, and the efficiency of the face images for identifying pedestrians is poor. In the field of pedestrian identity recognition, the human face which is required to be snapped is high in definition, and a common camera is difficult to meet the requirements. However, the AI camera with the face recognition function can meet the requirements, but the popularization of the AI camera in the prior art is small, the security monitoring demand cannot be met, and the AI camera construction period is long.
Disclosure of Invention
In order to solve among the prior art because AI camera arrangement rate is low, and ordinary camera is because shooting definition is not enough, consequently based on the technical problem of the unable quick discernment trip personal identification information of current fewest AI camera and ordinary camera, this application provides following technical scheme:
a method for improving pedestrian identity recognition efficiency, which is applied to pedestrian identity recognition equipment, wherein the pedestrian identity recognition equipment is respectively in communication connection with a plurality of common cameras and at least one AI camera, and the method comprises the following steps:
acquiring video data collected by the common camera and the AI camera;
detecting pedestrians from video data acquired by the AI camera and acquiring first description information of the detected pedestrians, and detecting pedestrians from video data acquired by the common camera and acquiring second description information of the detected pedestrians;
correspondingly storing the pedestrian and the first description information thereof to form an identity information search library;
calculating the similarity between the first description information and the second description information, judging whether the similarity is larger than a preset degree, and if so, associating the second description information with the first description information;
and updating the identity information search library according to the first description information and the second description information of the associated pedestrians.
The calculating the similarity between the first description information and the second description information, judging whether the similarity is greater than a preset degree, if so, associating the second description information with the first description information includes:
and respectively acquiring the feature vectors of the first descriptive information and the second descriptive information, calculating whether the Euclidean distance between the two feature vectors is larger than a preset value, if so, considering that the similarity of the first descriptive information and the second descriptive information is larger than a preset degree, and associating the second descriptive information with the pedestrian corresponding to the first descriptive information.
Further, the method further comprises the following steps: and clearing the identity information search library after a preset time period.
Wherein the first descriptive information and the second descriptive information each include a plurality of information elements;
the updating the identity information search library according to the first descriptive information and the second descriptive information of the associated pedestrians comprises the following steps:
for each information element in the second description information, judging whether the first description information contains the information element or not in sequence, if so, replacing the information element corresponding to the pedestrian in the identity information search library with the information element in the second description information; if not, adding the information elements in the information search library.
The first descriptive information and the second descriptive information at least comprise face information and pedestrian characteristics.
An apparatus for improving pedestrian identity recognition efficiency, equipped with a pedestrian identity recognition device, the identity recognition device being respectively in communication connection with a plurality of normal cameras and at least one AI camera, the apparatus comprising:
the data acquisition module is used for acquiring video data acquired by the common camera and the AI camera;
the pedestrian detection module is used for detecting pedestrians from video data acquired by the AI camera and the common camera respectively and acquiring first description information and second description information of the detected pedestrians respectively;
the library management module is used for correspondingly storing the pedestrians and the first description information thereof to form an identity information search library;
the pedestrian association module is used for calculating the similarity between the first description information and the second description information, judging whether the similarity is larger than a preset degree, and associating the second description information with the first description information if the similarity is larger than the preset degree;
and the updating module is used for updating the identity information search library according to the first description information and the second description information of the associated pedestrians.
Further, the device also comprises a feature extraction module for extracting feature vectors of the first description information and the second description information respectively;
and the pedestrian association module is used for calculating whether the Euclidean distance between the feature vectors is larger than a preset value, if so, the similarity between the first description information and the second description information is considered to be larger than a preset degree, and the second description information and the first description information are associated.
The library management module is further configured to clear the identity information search library after a preset time period.
An apparatus for improving pedestrian identity recognition efficiency, the apparatus being in communication connection with a plurality of normal cameras and at least one AI camera, respectively, the identity recognition apparatus comprising a memory and a processor, the memory having stored therein computer program instructions which, when read and run by the processor, perform a method for improving pedestrian identity recognition efficiency as described above.
A computer readable storage medium comprising a program executable by a processor to implement a method of improving pedestrian identification efficiency as described above.
According to the method, the device or the equipment for improving the pedestrian identity recognition efficiency, a small number of AI cameras and a large number of common cameras can be used for respectively acquiring video data and acquiring first description information of pedestrians from the acquired video data, the pedestrian description information analyzed from the video data adopted by the AI cameras is used as reference information of an identity information search library, second description information of the pedestrians acquired from the video data acquired by the common cameras is associated with the first description information of the pedestrians, and meanwhile, the identity information search library is updated according to the associated first description information and second description information of the pedestrians, so that the description information of the pedestrians in the identity information search library is richer, and meanwhile, the pedestrians can be updated in time, and therefore, the required identity information of the pedestrians can be searched from the identity information search library quickly and accurately.
Drawings
FIG. 1 is a flowchart of a method for improving pedestrian identification efficiency according to an embodiment of the present application;
FIG. 2 is a block diagram of a device for improving pedestrian identification efficiency according to an embodiment of the present application;
fig. 3 is a schematic diagram of an arrangement manner of a device for improving pedestrian identity recognition efficiency according to an embodiment of the present application.
Detailed Description
The invention will be described in further detail below with reference to the drawings by means of specific embodiments. Wherein like elements in different embodiments are numbered alike in association. In the following embodiments, numerous specific details are set forth in order to provide a better understanding of the present application. However, one skilled in the art will readily recognize that some of the features may be omitted, or replaced by other elements, materials, or methods in different situations. In some instances, some operations associated with the present application have not been shown or described in the specification to avoid obscuring the core portions of the present application, and may not be necessary for a person skilled in the art to describe in detail the relevant operations based on the description herein and the general knowledge of one skilled in the art.
Furthermore, the described features, operations, or characteristics of the description may be combined in any suitable manner in various embodiments. Also, various steps or acts in the method descriptions may be interchanged or modified in a manner apparent to those of ordinary skill in the art. Thus, the various orders in the description and drawings are for clarity of description of only certain embodiments, and are not meant to be required orders unless otherwise indicated.
Example 1
The embodiment provides a method for improving pedestrian identity recognition efficiency, which is applied to pedestrian identity recognition equipment, wherein the pedestrian identity recognition equipment is respectively in communication connection with a plurality of common cameras and an AI camera which are arranged in a scene, and the AI camera adopts an intelligent camera in order to clearly capture the face information of a pedestrian, as shown in fig. 1, and the method comprises the following steps:
step 101: acquiring video data acquired by a common camera and an AI camera; for example, for a certain arrangement scene, the AI camera can be arranged at a bayonet, such as a highway exit and an entrance, so as to acquire the identity information video of a pedestrian entering the arrangement scene, wherein the AI camera has the intelligent photographing function, so that a high-resolution and clear face image can be acquired, while the common camera cannot intelligently photograph due to fixed positions, so that the acquired face image is low in quality and low in resolution, and therefore, the recognition precision and recognition efficiency are relatively poor when the AI camera is used for face recognition.
Step 102: detecting pedestrians from video data acquired by an AI camera and acquiring first description information of the detected pedestrians, and detecting pedestrians from video data acquired by a common camera and acquiring second description information of the detected pedestrians; the first description information of the pedestrian detected from the video data acquired by the AI camera mainly comprises face information and the characteristics of the pedestrian, wherein the pedestrian characteristics and the face characteristics are high-dimensional characteristic vectors. The pedestrian is characterized by pedestrian re-identification, reID. This feature may take the entire figure as input, find the same person, or very similar persons. In the first description, two images of a human face and a pedestrian are acquired, and two high-dimensional features are obtained. In the second description, high-dimensional features and unreliable face features of pedestrians (faces cannot be found most of the time) can be obtained. By pedestrian feature matching, it is possible to in the second descriptionThe pedestrian is found in the first description. The identity information of the person may be determined by association. Because the resolution of the video data acquired by the AI camera is high, at least clear face information can be acquired from the video data acquired by the AI camera, and for convenience in storing description information, for example, for a certain bayonet, the video data captured by a certain entering AI camera is marked as C A1 . AI camera obtains face { F through its built-in face snapshot module and pedestrian snapshot module A1 Sum of pedestrian { P } A1 And the face recognition module and the pedestrian re-recognition module of the server fed into the background are used for extracting the features of the faces and the pedestrians, and the obtained face information and pedestrians are respectively marked as: { f A1 Sum of pedestrian { p } A1 }. Through the face and pedestrian retrieval module, the face ID is assigned 1 And pedestrian PID 1 And performs identity association { PID } 1 ，ID 1 By PID 1 Can obtain ID 1 . The second description information analyzed by the video data acquired by the normal camera includes a plurality of information elements such as face information and characteristics of pedestrians, but since the normal camera does not have an intelligent recognition function, the acquired video sometimes cannot clearly acquire all the above information, such as sometimes unreliable face information, so that the face information in the video sometimes is not available.
Step 103: correspondingly storing the pedestrian and the first description information thereof to form an identity information search library; for example, the obtained pedestrian and the first description information thereof are added into an identity information search library through a library management module, the state is set to be effective (monitoring is effective), the warehousing time is recorded, and a user can obtain the corresponding identity information by searching the identity information search library at any time.
Step 104: calculating the similarity between the first description information and the second description information, judging whether the similarity is larger than a preset degree, and if so, associating the second description information with the first description information; and calculating feature vectors of the first descriptive information and the second descriptive information respectively, calculating Euclidean distance between the two feature vectors, and if the Euclidean distance between the two feature vectors is smaller than a preset value, considering that the similarity between the first descriptive information and the second descriptive information is larger than a preset degree, and associating the second descriptive information with the first descriptive information.
Step 105: and updating the identity information search library according to the first descriptive information and the second descriptive information of the associated pedestrian. Specifically, after associating the first description information with the second description information, updating the identity information search library according to the associated first description information and second description information of the pedestrian includes: for each information element in the second description information, judging whether the first description information contains the information element or not in sequence, if so, replacing the information corresponding to the pedestrian in the information search library with the information element in the second description information; if not, adding the information element in the information search library. For example, if the first description information only includes face information and the second description information includes clothing information, the clothing information is added to the description information corresponding to the pedestrian in the information search library, and the clothing information mainly refers to the type and color of clothing, such as red down jackets. In some cases, the first description information includes clothing information, for example, red overcoat, but if the pedestrian changes the clothing midway, the clothing information acquired through the common camera is black down coat, and the clothing information corresponding to the pedestrian is replaced by the clothing information in the second description information, so that the information search library can be updated timely, and the most prepared pedestrian identity information can be ensured to be searched.
Further, the identity information search library is cleared after a preset time period, data storage is obtained again, a new information search library is constructed, and the identity information search library is cleared once every 24 hours, for example.
By the method, a small number of AI cameras and a large number of common cameras can be used for respectively acquiring video data and acquiring first description information of pedestrians from the acquired video data, the pedestrian description information analyzed from the video data adopted by the AI cameras is used as reference information of an identity information search library, second description information of the pedestrians acquired from the video data acquired by the common cameras is associated with the first description information of the pedestrians, and meanwhile the identity information search library is updated according to the first description information and the second description information of the pedestrians, so that the description information of the pedestrians in the identity information search library is richer, and meanwhile, the pedestrians can be updated timely, and therefore the required identity information of the pedestrians can be searched from the identity information search library quickly and accurately.
Example 2
The present embodiment provides a device for improving pedestrian identity recognition efficiency, which is provided in a pedestrian identity recognition apparatus, and the identity recognition apparatus is respectively in communication connection with a plurality of normal cameras and at least one AI camera arranged in a scene, as shown in fig. 2, and the device includes:
the data acquisition module 201 is used for acquiring video data acquired by the common camera and the AI camera;
the pedestrian detection module 202 is configured to detect a pedestrian from video data collected by the AI camera and the normal camera, and obtain first description information and second description information of the detected pedestrian, respectively; after the pedestrian detection module 202 detects a pedestrian, the pedestrian re-recognition module extracts the high-dimensional features of the pedestrian for comparison and retrieval. The first descriptive information and the second descriptive information at least comprise face information, clothing information, height information and special mark information, wherein the special mark information is used for example that a pedestrian has a tattoo and the like. For example, for a certain bayonet, denoted as bayonet 1, video data captured by a certain incoming AI camera is denoted as C A1 . AI camera obtains face { F through its built-in face snapshot module and pedestrian snapshot module A1 Sum of pedestrian { P } A1 And the face recognition module and the pedestrian re-recognition module of the server fed into the background are used for extracting the features of the faces and the pedestrians, and the obtained face information and pedestrians are respectively marked as: { f A1 Sum of pedestrian { p } A1 }. At the common camera, marked as C vi The pedestrian detection module of the background server samples the video stream every s seconds (s takes 1-5), and then performs face detection and pedestrian detection. Face detection is performed mainly because there is a possibility that some detected faces are recognized. The detected faces and pedestrians are recorded as follows: f (F) vi And P vi Firstly, the human face feature extraction module and the pedestrian feature extraction module are fed intoAnd the module is taken for feature extraction. If only the face or the pedestrian is detected at this time (because the actual scene is complex, the pedestrian detection module cannot detect the situation), only the face or the pedestrian feature is extracted. By f vi And p vi Feature vectors representing the extracted first and second description information, respectively.
The library management module 203 is configured to store the pedestrian and the first description information thereof in correspondence to form an identity information search library; through the face and pedestrian retrieval module, the face ID is assigned 1 And pedestrian PID 1 And performs identity association { PID } 1 ，ID 1 By PID 1 Can obtain ID 1 . And adding the features into an identity information search library through a library management module, setting the state as effective (monitoring effective), and recording the warehousing time.
And the pedestrian association module 204 is configured to calculate a similarity between the first description information and the second description information, determine whether the similarity is greater than a preset degree, and if yes, associate the second description information with the first description information. Specifically, the pedestrian association module is configured to calculate whether the euclidean distance between feature vectors is greater than a preset value, if yes, consider that the similarity between the first description information and the second description information is greater than a preset degree, associate the second description information with the first description information, and if the pedestrian PID is retrieved i Without retrieving the face ID i Then use the established association relation { PID ] i ，ID i Identity given to the pedestrian feature is ID i . If the face information in the identity information search library is searched through the face information and the associated ID of the pedestrian information obtained through the pedestrian feature search is inconsistent, the association ID matched with the face is used as the reference, and meanwhile, the library management module records two inconsistent face identity IDs and the pedestrian ID and stores corresponding video screenshot to carry out subsequent auditing. If the two are consistent, the identity ID of the pedestrian characteristic is directly given, the detected pedestrian characteristic is added into an identity information search library of the pedestrian, and the identity information search library is enriched in { p } A1 、p A2 、p vi }。
Update module 205, forAnd updating the identity information search library according to the first descriptive information and the second descriptive information of the associated pedestrian. After the first descriptive information and the second descriptive information are associated, the identity information search library can be updated through the second descriptive information, specifically, for each information element in the second descriptive information, whether the first descriptive information contains the information element or not is judged in sequence, if so, the information corresponding to the pedestrian in the information search library is replaced by the information element in the second descriptive information; if not, adding the information element in the information search library. Therefore, the description information of the pedestrian in the identity information search library is more abundant, and meanwhile, the pedestrian can be updated in time, so that the required identity information of the pedestrian can be quickly and accurately searched from the identity information search library. For example, when a pedestrian who leaves the entrance 1 enters another entrance, denoted as entrance 2, and a certain AI camera is denoted as C A2 And (5) taking a snapshot. After feature extraction and search comparison, if the similarity of the description information obtained through the two AI cameras is larger than a preset value, enriching a search library { f A1 、f A2 Sum of pedestrian { p } A1 、p A2 }. AI camera C of the above embodiment using only bayonet 1, 2 A1 、C A1 Common camera C vi For example, the invention is not limited to the several cameras described above, multiple bayonets, multiple AI cameras, and a large number of common cameras in a time-directed surveillance scene.
Further, the library management module 203 is further configured to clear the identity information search library after a preset period of time. The method ensures that the expired information is cleared in time, the memory is released, and new pedestrian information is stored. Each pedestrian information in the expired pedestrian identity information search library is cleared, for example, at zero points every day.
Through the device of the embodiment, a small number of AI cameras and a large number of common cameras can be used for respectively acquiring video data and acquiring first description information of pedestrians from the acquired video data, the pedestrian description information analyzed from the video data adopted by the AI cameras is used as reference information of an identity information search library, second description information of the pedestrians acquired from the video data acquired by the common cameras is associated with the first description information of the pedestrians, and meanwhile, the identity information search library is updated according to the associated first description information and second description information of the pedestrians, so that the description information of the pedestrians in the identity information search library is richer, and meanwhile, the pedestrians can be updated timely, and therefore, the required identity information of the pedestrians can be searched from the identity information search library quickly and accurately.
Example 3
The present embodiment provides a device for improving the efficiency of pedestrian identification, as shown in fig. 3, where the device is respectively connected to a plurality of common cameras and at least one AI camera in a communication manner, where the AI cameras are generally disposed at the positions of the gates of the urban arrangement scene, and the common cameras are disposed along the lines, and the identity device includes a memory and a processor, where the memory stores computer program instructions, and when the computer program instructions are read and executed by the processor, the method for improving the efficiency of pedestrian identification provided in the embodiment 1 is executed.
Example 4
The present embodiment provides a computer-readable storage medium including a program executable by a processor to implement the method of improving pedestrian identification efficiency as provided in embodiment 1.
The foregoing description of the invention has been presented for purposes of illustration and description, and is not intended to be limiting. Several simple deductions, modifications or substitutions may also be made by a person skilled in the art to which the invention pertains, based on the idea of the invention.
Claims (5)
1. A method for improving the efficiency of pedestrian identification, which is characterized by being applied to equipment for improving the efficiency of pedestrian identification, wherein the equipment for improving the efficiency of pedestrian identification is respectively in communication connection with a plurality of common cameras and at least one AI camera, and the method comprises the following steps:
acquiring video data acquired by the common camera and the AI camera;
detecting pedestrians from video data acquired by the AI camera and acquiring first description information of the detected pedestrians, detecting pedestrians from video data acquired by the common camera and acquiring second description information of the detected pedestrians, wherein the first description information and the second description information comprise various information elements;
correspondingly storing the pedestrian and the first description information thereof to form an identity information search library;
respectively acquiring feature vectors of the first descriptive information and the second descriptive information, calculating whether the Euclidean distance between the two feature vectors is larger than a preset value, if so, considering that the similarity of the first descriptive information and the second descriptive information is larger than a preset degree, and associating the second descriptive information with pedestrians corresponding to the first descriptive information;
updating the identity information search library according to the first description information and the second description information of the associated pedestrians, wherein the updating comprises the following steps: judging whether the first descriptive information contains the information element or not for each information element in the second descriptive information, if so, replacing the information element corresponding to the pedestrian in the identity information search library with the information element in the second descriptive information; if not, adding the information element in the identity information search library;
the method further comprises the steps of: and clearing the identity information search library after a preset time period.
2. The method for improving pedestrian identification efficiency of claim 1 wherein the first and second description information include at least face information and pedestrian characteristics.
3. An apparatus for improving pedestrian identity recognition efficiency, which is characterized by being equipped with a device for improving pedestrian identity recognition efficiency, wherein the device for improving pedestrian identity recognition efficiency is respectively in communication connection with a plurality of common cameras and at least one AI camera, and the apparatus comprises:
the data acquisition module is used for acquiring video data acquired by the common camera and the AI camera;
the pedestrian detection module is used for detecting pedestrians from video data acquired by the AI camera and the common camera respectively and acquiring first description information and second description information of the detected pedestrians respectively, wherein the first description information and the second description information comprise various information elements;
the library management module is used for correspondingly storing the pedestrians and the first description information thereof to form an identity information search library, and clearing the identity information search library after a preset time period;
the feature extraction module is used for extracting feature vectors of the first description information and the second description information respectively;
the pedestrian association module is used for calculating whether the Euclidean distance between the feature vector of the first descriptive information and the feature vector of the second descriptive information is larger than a preset value, if so, the similarity of the first descriptive information and the second descriptive information is considered to be larger than a preset degree, and the second descriptive information and the first descriptive information are associated;
the updating module is used for updating the identity information search library according to the first description information and the second description information of the associated pedestrians, and specifically comprises the following steps: judging whether the first descriptive information contains the information element or not for each information element in the second descriptive information, if so, replacing the information element corresponding to the pedestrian in the identity information search library with the information element in the second descriptive information; if not, adding the information element in the identity information search library.
4. An apparatus for improving pedestrian identification efficiency, wherein the apparatus for improving pedestrian identification efficiency is in communication connection with a plurality of normal cameras and at least one AI camera, respectively, the apparatus for improving pedestrian identification efficiency comprising a memory and a processor, the memory having stored therein computer program instructions that, when read and executed by the processor, perform the method for improving pedestrian identification efficiency as claimed in any one of claims 1-2.
5. A computer-readable storage medium comprising a program executable by a processor to implement the method of improving pedestrian identification efficiency of any one of claims 1-2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910732754.9A CN110598551B (en) | 2019-08-09 | 2019-08-09 | Method, device, equipment and medium for improving pedestrian identity recognition efficiency |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910732754.9A CN110598551B (en) | 2019-08-09 | 2019-08-09 | Method, device, equipment and medium for improving pedestrian identity recognition efficiency |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110598551A CN110598551A (en) | 2019-12-20 |
| CN110598551B true CN110598551B (en) | 2023-07-28 |
Family
ID=68853834
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910732754.9A Active CN110598551B (en) | 2019-08-09 | 2019-08-09 | Method, device, equipment and medium for improving pedestrian identity recognition efficiency |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110598551B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111368101B (en) * | 2020-03-05 | 2021-06-18 | 腾讯科技（深圳）有限公司 | Multimedia resource information display method, device, equipment and storage medium |
| CN111797691A (en) * | 2020-06-03 | 2020-10-20 | 力引万物(深圳)科技有限公司 | Method for improving face recognition accuracy and processing subsystem |
| CN111898515A (en) * | 2020-07-24 | 2020-11-06 | 北京灵汐科技有限公司 | Method and device for identifying identity of pedestrian |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108540755B (en) * | 2017-03-01 | 2021-03-05 | 中国电信股份有限公司 | Identity recognition method and device |
| CN108230700A (en) * | 2017-12-29 | 2018-06-29 | 上海大汉三通无线通信有限公司 | A kind of method for controlling traffic signal lights, system and traffic control system |
| CN109660509A (en) * | 2018-10-29 | 2019-04-19 | 北京旷视科技有限公司 | Login method, device, system and storage medium based on recognition of face |
| CN109582824A (en) * | 2018-12-29 | 2019-04-05 | 深圳英飞拓科技股份有限公司 | A kind of region security management system and method based on video structural |
| CN109740573B (en) * | 2019-01-24 | 2021-10-01 | 北京旷视科技有限公司 | Video analysis method, device, equipment and server |
| CN110084103A (en) * | 2019-03-15 | 2019-08-02 | 深圳英飞拓科技股份有限公司 | A kind of same pedestrian's analysis method and system based on face recognition technology |
-
2019
- 2019-08-09 CN CN201910732754.9A patent/CN110598551B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN110598551A (en) | 2019-12-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106709424B (en) | Optimized monitoring video storage system | |
| JP6854881B2 (en) | Face image matching system and face image search system | |
| CN110598551B (en) | Method, device, equipment and medium for improving pedestrian identity recognition efficiency | |
| CN111131902B (en) | Method for determining target object information and video playing equipment | |
| US8570376B1 (en) | Method and system for efficient sampling of videos using spatiotemporal constraints for statistical behavior analysis | |
| US8314854B2 (en) | Apparatus and method for image recognition of facial areas in photographic images from a digital camera | |
| CN109344271A (en) | Video portrait records handling method and its system | |
| JP6139364B2 (en) | Person identification device, person identification method and program | |
| US20130266181A1 (en) | Object tracking and best shot detection system | |
| US20080186381A1 (en) | Customer service counter/checkpoint registration system with video/image capturing, indexing, retrieving and black list matching function | |
| US20130329059A1 (en) | Person detection system | |
| CN105488478A (en) | Face recognition system and method | |
| CN111222373B (en) | Personnel behavior analysis method and device and electronic equipment | |
| KR20060100341A (en) | Image monitoring system for object identification | |
| CN111814510B (en) | Method and device for detecting legacy host | |
| CN109635146A (en) | A kind of target query method and system based on characteristics of image | |
| CN111090777B (en) | Video data management method, management equipment and computer storage medium | |
| KR101084914B1 (en) | Indexing management system of vehicle-number and man-image | |
| CN114926757A (en) | Method and system for retrieving lost article, electronic equipment and computer storage medium | |
| CN110717428A (en) | Identity recognition method, device, system, medium and equipment fusing multiple features | |
| KR20090035798A (en) | Store and search system and the method of the license plate images | |
| CN112132057A (en) | Multi-dimensional identity recognition method and system | |
| US11995915B2 (en) | Systems and methods for collecting video clip evidence from a plurality of video streams of a video surveillance system | |
| CN110876090B (en) | Video abstract playback method and device, electronic equipment and readable storage medium | |
| CN112232113B (en) | Person identification method, person identification device, storage medium, and electronic apparatus |
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 |