CN113591620A - Early warning method, device and system based on integrated mobile acquisition equipment - Google Patents
Early warning method, device and system based on integrated mobile acquisition equipment Download PDFInfo
- Publication number
- CN113591620A CN113591620A CN202110800147.9A CN202110800147A CN113591620A CN 113591620 A CN113591620 A CN 113591620A CN 202110800147 A CN202110800147 A CN 202110800147A CN 113591620 A CN113591620 A CN 113591620A
- Authority
- CN
- China
- Prior art keywords
- license plate
- face
- target
- plate information
- queue
- 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
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000012545 processing Methods 0.000 claims description 45
- 230000008030 elimination Effects 0.000 claims description 11
- 238000000605 extraction Methods 0.000 claims description 11
- 238000010606 normalization Methods 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 7
- 238000003379 elimination reaction Methods 0.000 claims description 7
- 230000003287 optical effect Effects 0.000 claims description 3
- 238000012216 screening Methods 0.000 abstract description 5
- 238000007726 management method Methods 0.000 description 10
- 238000012549 training Methods 0.000 description 8
- 238000001514 detection method Methods 0.000 description 6
- 238000013528 artificial neural network Methods 0.000 description 4
- 238000011161 development Methods 0.000 description 4
- 230000018109 developmental process Effects 0.000 description 4
- 230000001815 facial effect Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- 230000005540 biological transmission Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 230000011218 segmentation Effects 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013523 data management Methods 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 239000000835 fiber Substances 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
Images
Classifications
-
- 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
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/61—Control of cameras or camera modules based on recognised objects
- H04N23/611—Control of cameras or camera modules based on recognised objects where the recognised objects include parts of the human body
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/66—Remote control of cameras or camera parts, e.g. by remote control devices
- H04N23/661—Transmitting camera control signals through networks, e.g. control via the Internet
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/695—Control of camera direction for changing a field of view, e.g. pan, tilt or based on tracking of objects
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/80—Camera processing pipelines; Components thereof
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Data Mining & Analysis (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
- Alarm Systems (AREA)
Abstract
The invention relates to an intelligent police technology, and provides an early warning method, an early warning device, a readable storage medium and a system based on integrated mobile acquisition equipment, wherein the method comprises the following steps: collecting a target image; judging whether the target image contains a human face or/and a license plate; if the human face exists, primarily removing the duplication, determining the position of the human face to extract the target human face characteristics, comparing the target human face characteristics with the human face characteristics in the human face queue, if the human face queue has the same human face characteristics, removing the target human face characteristics, and if not, adding the target human face characteristics into the human face queue; if the license plate is included, detecting a license plate area to obtain target license plate information, comparing the target license plate information with the license plate information in the license plate queue, if the license plate queue has the same license plate information, rejecting the target license plate information, and if not, adding the target license plate information into the license plate queue; and the face features and the license plate information are taken out from the queue and are respectively matched with the specified face features and the specified license plate information, and if the face features and the license plate information are matched with the specified face features and the specified license plate information, the early warning is performed, so that the accurate early warning is realized, and the screening range is wider due to the fact that the face and the license plate are collected simultaneously.
Description
Technical Field
The invention relates to the technical field of intelligent police affairs, in particular to an early warning method, an early warning device, a readable storage medium and a system based on integrated mobile acquisition equipment.
Background
Along with the rapid development of the internet, the mode that the internet combines with police service becomes a common mode of police service law enforcement work gradually, wisdom police car system also arises at the right moment, wisdom police car system is a removal police car and gathers management platform, contain collection terminal, data management platform etc. usually, usually through snapshotting pedestrian's facial image, then discern facial image, early warning is carried out with the recognition result, but such method collection alone has the limitation, gather the mistake and leak easily, and image processing process is also comparatively simple, and data is accurate inadequately.
Disclosure of Invention
The invention mainly aims to provide an early warning method, an early warning device, a readable storage medium and an early warning system based on an integrated mobile acquisition device, which are wider in screening range and more accurate in data.
Based on the above purpose, the invention provides an early warning method based on an integrated mobile acquisition device, which is implemented by the integrated mobile acquisition device, wherein the integrated mobile acquisition device comprises an acquisition camera, and the method comprises the following steps:
collecting a target image through the collecting camera;
judging whether the target image contains a human face or/and a license plate;
if the target image contains a face, performing preliminary duplication removal on the face in the target image, determining the face position in the target image after the preliminary duplication removal to extract target face features, comparing the target face features with face features in a preset face queue to determine whether the face queue has the same face features as the target face features, if so, removing the target face features, and if not, adding the target face features into the face queue;
if the target image contains a license plate, detecting the region of the license plate in the target image, acquiring target license plate information, comparing the target license plate information with license plate information in a preset license plate queue to determine whether the license plate queue has the same license plate information as the target license plate information, if so, rejecting the target license plate information, and if not, adding the target license plate information into the license plate queue;
sequentially taking out the face features from the face queue, matching the face features with specified face features in a preset database of a police system, sequentially taking out the license plate information from the license plate queue, and matching the license plate information with the specified license plate information in the specified database of the police system;
and when the face features are matched with the designated face features or/and the license plate information is matched with the designated license plate information, early warning is carried out.
Further, the step of performing preliminary deduplication on the face in the target image includes:
selecting the target images shot by the same camera to obtain a tracking id of each target in the target images, wherein each target carries a corresponding tracking id, and the target comprises the face;
judging whether the tracking id has a plurality of same tracking ids;
if so, removing the targets corresponding to the same tracking ids, so that only one target corresponding to one tracking id is reserved in the same tracking ids.
Further, the step of obtaining the target license plate information includes:
carrying out image binarization processing on the target image;
and performing single extraction on license plate characters in the target image subjected to image binarization processing to obtain a plurality of license plate characters, wherein the license plate characters are used as the target license plate information.
Further, the license plate information is a segmented character gray image, and the step of matching the license plate information with the designated license plate information in the designated database includes:
carrying out normalization processing on the character gray level image;
carrying out feature extraction on the character gray level image after the normalization processing to obtain license plate features;
and matching the license plate characteristics with the designated license plate information in the designated database, and taking the result with the highest matching degree as a matching result, wherein the designated license plate information comprises a designated character template.
Further, before the step of acquiring the target image by the acquisition camera, the method includes:
and connecting the acquisition camera through an OTG (optical transport group) so as to transmit the target image.
Further, before the steps of sequentially taking out the face features from the face queue, matching the face features with the designated face features in a preset database of a police service system, sequentially taking out the license plate information from the license plate queue, and matching the license plate information with the designated license plate information in the designated database of the police service system, the method includes:
and the police system is connected through a 4G or 5G network.
The invention also provides an early warning device based on the integrated mobile acquisition equipment, which is realized by the integrated mobile acquisition equipment, wherein the integrated mobile acquisition equipment comprises an acquisition camera, and the device comprises:
the image acquisition unit is used for acquiring a target image through the acquisition camera;
the image judging unit is used for judging whether the target image contains a human face or/and a license plate;
a face processing unit, configured to perform preliminary deduplication on a face in the target image when the target image includes the face, determine a face position in the target image after the preliminary deduplication, extract a target face feature, compare the target face feature with a face feature in a preset face queue, and determine whether the face queue has a face feature that is the same as the target face feature, if yes, remove the target face feature, and if not, add the target face feature into the face queue;
the license plate processing unit is used for detecting the area of a license plate in the target image when the target image contains the license plate, acquiring target license plate information, comparing the target license plate information with license plate information in a preset license plate queue to determine whether the license plate queue has the same license plate information as the target license plate information, if so, rejecting the target license plate information, and if not, adding the target license plate information into the license plate queue;
the matching queue unit is used for sequentially taking out the face features from the face queue, matching the face features with specified face features in a preset database of a police system, sequentially taking out the license plate information from the license plate queue, and matching the license plate information with the license plate information specified in the specified database of the police system;
and the early warning information unit is used for carrying out early warning when the face features are matched with the specified face features or/and the license plate information is matched with the specified license plate information.
Further, the face processing unit includes:
a tracking information obtaining subunit, configured to select the target image captured by the same camera to obtain a tracking id of each target in the target image, where each target carries a corresponding tracking id, and the target includes the face;
a tracking information judging subunit, configured to judge whether there are multiple identical tracking ids in the tracking ids;
and the tracking information removing subunit is configured to remove, if yes, the targets corresponding to the multiple same tracking ids, so that only one target corresponding to one tracking id is reserved in the multiple same tracking ids.
The invention also provides a computer readable storage medium, on which a computer program is stored, which when executed by a processor implements the steps of the above-mentioned early warning method based on an integrated mobile acquisition device.
The invention also provides an early warning system based on the integrated mobile acquisition equipment, which comprises the integrated mobile acquisition equipment and an intelligent terminal, wherein the integrated mobile acquisition equipment is connected with the intelligent terminal through an OTG (over the air) network, and the intelligent terminal is connected with a police service system through a 4G or 5G network;
the integrated mobile acquisition equipment comprises an acquisition camera and a processing module used for processing a face and a license plate, wherein a target image is acquired through the acquisition camera, the target image is subjected to duplication elimination through the processing module, target face characteristics and target license plate information are acquired, the target face characteristics are put into a face queue, the target license plate information is put into a license plate queue, and the face characteristics and the license plate information are sequentially taken out from the face queue and the license plate queue and are transmitted to the intelligent terminal;
the intelligent terminal obtains the face features in a preset database through a police system, obtains license plate information in a specified database, matches the face features with the specified face features in the preset database of the police system, matches the license plate information with the specified license plate information in the specified database of the police system, and gives out early warning when the face features are matched with the specified face features or/and the license plate information is matched with the specified license plate information.
The invention has the beneficial effects that: through with people's face and license plate discernment, compare the algorithm and transplant to embedded development board, combine to install and remove the small-size collection camera of taking a candid photograph at integral type mobile acquisition equipment top, realize accurate early warning, avoid the blind area of fixed position candid photograph, and owing to gather people's face and license plate simultaneously, the screening scope is wider to remove the repetition many times so that data is more accurate.
Drawings
Fig. 1 is a schematic diagram illustrating steps of an early warning method based on an integrated mobile acquisition device according to an embodiment of the present invention;
fig. 2 is a schematic block diagram of a structure of an early warning device based on an integrated mobile acquisition device according to an embodiment of the present invention;
fig. 3 is a schematic block diagram of a structure of a storage medium in an embodiment of the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
Referring to fig. 1, the early warning method based on the integrated mobile acquisition device in this embodiment is implemented by the integrated mobile acquisition device, where the integrated mobile acquisition device includes an acquisition camera, and the method includes:
step S1: collecting a target image through the collecting camera;
step S2: judging whether the target image contains a human face or/and a license plate;
step S3: if the target image contains a face, performing preliminary duplication removal on the face in the target image, determining the face position in the target image after the preliminary duplication removal to extract target face features, comparing the target face features with face features in a preset face queue to determine whether the face queue has the same face features as the target face features, if so, removing the target face features, and if not, adding the target face features into the face queue;
step S4: if the target image contains a license plate, detecting the region of the license plate in the target image, acquiring target license plate information, comparing the target license plate information with license plate information in a preset license plate queue to determine whether the license plate queue has the same license plate information as the target license plate information, if so, rejecting the target license plate information, and if not, adding the target license plate information into the license plate queue;
step S5: sequentially taking out the face features from the face queue, matching the face features with specified face features in a preset database of a police system, sequentially taking out the license plate information from the license plate queue, and matching the license plate information with the specified license plate information in the specified database of the police system;
step S6: and when the face features are matched with the designated face features or/and the license plate information is matched with the designated license plate information, early warning is carried out.
The early warning method based on the integrated mobile acquisition equipment can be realized through program software of the intelligent terminal. In this embodiment, The integral type removes collection equipment can be for including gathering The camera, multiple different sensor, treater and mobile device's equipment, also can be for collecting such as collection, remove, multiple functions in The wisdom police car of an organic whole such as early warning, be provided with On this integral type removes collection equipment and gather The camera, preferably, set up in The top of integral type removes collection equipment, for example, it sets up in The roof of wisdom police car to gather The camera, then connect integral type removal collection equipment through OTG (On-The-Go), so that data transmission, for example, transmit The image, because current people's face, picture collection commonly uses bayonet socket camera snapshot, individual soldier's equipment collection, The position is relatively fixed, do not have The mobility snapshot, remove collection equipment through The integral type, can finely compensate traditional collection blind area.
As the step S2, the target image may be acquired by the multiple cameras of the OTG integrated mobile acquisition device, and then whether the target image includes a face and/or a license plate is determined, specifically, when detecting whether the target image includes a face, the detection may be performed by using a standard face template, for example, one or more standard face templates may be designed first, then the matching degree between the image acquired by the test and the standard template is calculated, and whether the face exists is determined by using a threshold; or the corresponding rules are generated by extracting the features in the target image, the corresponding face rules can be generated because the face has certain structural distribution features, and whether the target image contains the face can be judged by referring to the face rules; or, identifying whether a face exists in the target image through a preset face classifier, wherein the face classifier is obtained by training and learning through a training set containing face samples and non-face samples based on an artificial neural network; or detecting according to the rule that the distribution of the face complexion in the color space is relatively concentrated so as to determine whether the target image contains the face; or, a face subspace is preset, the face subspace is a set of a plurality of faces, and then the target image is mapped to the face subspace, so that whether the face exists in the target image can be judged based on the distance between the target image and the projection of the target image in the face subspace. It should be noted that the present invention does not limit the detection method, as long as it can detect whether the target image includes a human face.
When detecting whether a license plate exists in a target image, a machine learning algorithm can be adopted to detect the license plate according to information such as texture features, color features and shape features, and specifically, the detection can be completed through a license plate classifier, wherein the license plate classifier is obtained by training and learning through a training set containing license plate samples and non-license plate samples based on an artificial neural network.
As shown in step S3, if the image includes a face, the face in the target image is subjected to preliminary deduplication, so that face processing is subsequently performed on the target image, and then the face position in the target image after preliminary deduplication is determined to extract the target face feature.
When the face features are extracted, the size, position, distance and other attributes of facial image, then calculate their geometric characteristic quantity, form a feature vector describing the face with these characteristic quantity, also is the face features.
And then comparing the target face features with the face features in a preset face queue to perform second duplication removal, so as to reduce the workload of later period, specifically, storing the introduced face features in the face queue or storing the face features within a specified time, firstly calculating the similarity between the target face features and the face features in the face queue, and judging whether the similarity is consistent with the similarity when the similarity exceeds a preset threshold value, so as to determine whether the face queue has the face features same as the target face features, if so, proving that the target face features are introduced, removing the target face features, and if not, adding the target face features into the face queue.
As described in the step S4, if the target image includes a license plate, detecting a region of the license plate in the target image, then determining the region of the license plate in the target image by dynamic target tracking, obtaining the target license plate information, comparing the target license plate information with license plate information in a preset license plate queue to determine whether the license plate queue has the same license plate information as the target license plate information, if so, removing the target license plate information, and if not, adding the target license plate information into the license plate queue. Specifically, the step of obtaining the target license plate information includes:
step S41: carrying out image binarization processing on the target image;
step S42: and performing single extraction on license plate characters in the target image subjected to image binarization processing to obtain a plurality of license plate characters, wherein the license plate characters are used as the target license plate information.
In this embodiment, an image binarization process is performed on a target image, and then based on a binarization result, information such as structural features of characters, similarity between characters, and intervals between characters is utilized to extract individual characters respectively, so as to obtain single characters of a plurality of license plates, and the single characters are combined to serve as the target license plate information. In another embodiment, the marginalization processing may be performed on the target image, and then the character segmentation may be performed based on the marginalization result.
As described in the above steps S5-S6, the face features are sequentially extracted from the face queue, and the face features are matched with the specified face features in the preset database of the police service system, where the police service system is an existing police service integrated system in use, and a plurality of pieces of specified person information, such as the specified face features, are stored in the police service integrated system, and the specified person is found by comparing the face features with the specified person information. The police system can be connected through a 4G or 5G network to achieve interconnection and intercommunication with the police system.
And simultaneously, the license plate information can be sequentially taken out from the license plate queue, and the license plate information is matched with the license plate information appointed in the police system appointed database, wherein the appointed license plate information can be the license plate information of appointed vehicles such as illegal vehicles. And when the face features are matched with the designated face features or/and the license plate information is matched with the designated license plate information, triggering an early warning program to start early warning, such as sending out an alarm or sending out reminding information.
In one embodiment, the license plate information is a license plate feature, and the step S5 includes:
step S51: carrying out normalization processing on the character gray level image;
step S52: carrying out feature extraction on the character gray level image after the normalization processing to obtain license plate features;
step S53: and matching the license plate characteristics with the designated license plate information in the designated database, and taking the result with the highest matching degree as a matching result, wherein the designated license plate information comprises a designated character template.
In this embodiment, the license plate information is a segmented character gray image, the character gray image is normalized, the character gray image after the normalization is subjected to feature extraction to obtain license plate features, and the license plate features are matched with a designated character module, where the character module is the designated license plate information in the designated database, and a result with the highest matching degree is used as a matching result.
In an embodiment, in the step S3, the step of performing preliminary deduplication on the face in the target image includes:
step S31: selecting the target images shot by the same camera to obtain a tracking id of each target in the target images, wherein each target carries a corresponding tracking id, and the target comprises the face;
step S32: judging whether the tracking id has a plurality of same tracking ids;
step S33: if so, removing the targets corresponding to the same tracking ids, so that only one target corresponding to one tracking id is reserved in the same tracking ids.
In this embodiment, the targets are objects in a target image, such as people, vehicles, trees, and the like, and may also be set as human faces, each target carries a corresponding tracking id, and the tracking ids of the same target in the same camera are identical, the tracking id is a unique id of each target, and may be generated by a random algorithm or a preset rule, in order to reduce workload of subsequent steps, first performing first deduplication by the tracking id, first selecting a target image shot by the same camera to obtain the tracking id of each target in the target image, and then determining whether the tracking ids have a plurality of identical tracking ids; if so, the tracking id is removed, that is, the target corresponding to the tracking id is removed at the same time, so that only one target corresponding to one tracking is reserved in a plurality of identical tracking ids.
In another embodiment, the foregoing duplication elimination step may be implemented by a collecting camera, where the collecting camera is provided with a duplication elimination program, and when a target image is collected, the collecting camera directly enters the duplication elimination program to perform preliminary duplication elimination.
According to the early warning method based on the integrated mobile acquisition equipment, the human face and license plate recognition and comparison algorithm is transplanted into the embedded development board, and the accurate early warning is realized by combining the small snapshot camera which is installed at the top of the integrated mobile acquisition equipment and used for performing mobile snapshot, so that a dead zone of fixed point snapshot is avoided, and the human face and license plate are simultaneously acquired, so that the screening range is wider, and the data is more accurate due to repeated duplication removal. In addition, the OTG is communicated to transmit the acquired images, and meanwhile, the OTG is matched with a mobile 4G or 5G video network to achieve data interconnection and intercommunication with a police service system, so that the effects of remote command and mobile police service are achieved.
The invention also provides an integrated mobile acquisition equipment system which is used for realizing the early warning method based on the integrated mobile acquisition equipment, and the system comprises the integrated mobile acquisition equipment and an intelligent terminal, wherein the integrated mobile acquisition equipment is connected with the intelligent terminal through an OTG (over the air) network, and the intelligent terminal is connected with a police service system through a 4G or 5G network.
Specifically, the integrated mobile acquisition equipment comprises an acquisition camera and a processing module for processing a face and a license plate, wherein the acquisition camera acquires the target image, the processing module performs duplication elimination processing on the target image, acquires target face features and target license plate information, puts the target face features into a face queue, puts the target license plate information into a license plate queue, and sequentially takes out the face features and the license plate information from the face queue and the license plate queue to transmit to the intelligent terminal; the intelligent terminal obtains the face features in a preset database through a police system, obtains license plate information in a specified database, matches the face features with the specified face features in the preset database of the police system, matches the license plate information with the specified license plate information in the specified database of the police system, and gives out early warning when the face features are matched with the specified face features or/and the license plate information is matched with the specified license plate information.
Further, above-mentioned integral type removes collection equipment system still includes the visual statistics module of big data, gathers storehouse, cloth accuse storehouse, early warning storehouse, instant messaging module, gathers retrieval module and sets up the module, and wherein the visual statistics module of big data is used for making statistics of and visual show the relevant data of integral type removal collection equipment, specifically includes alert feelings statistics, early warning information statistics, gathers statistics, license plate statistics of attendance, map show etc.. The system comprises a collection library, a control and deployment library and an early warning library, wherein the collection library, the control and deployment library and the early warning library are used for storing different data, an instant communication module is used for carrying out real-time communication, and a collection and retrieval module is used for retrieving human images, license plates and the like, for example, the human face characteristics are obtained, human face matching is carried out at the background through a human face recognition algorithm, the recognition result is displayed, the preset license plate and the license plate collected by the police car are subjected to condition searching, and the collected historical data are displayed. The setting module is used for carrying out back-end setting and maintenance on program software updating, algorithm updating, users, departments, authorities, maps and the like, and specifically comprises APP management, police car management, back-end user management, front-end user management, authority management, department management, menu management, map management, algorithm management and the like.
Referring to fig. 2, in this embodiment, an early warning apparatus based on an integrated mobile acquisition device is provided, where the apparatus corresponds to the above early warning method based on an integrated mobile acquisition device, where the integrated mobile acquisition device includes an acquisition camera, and the apparatus includes:
the image acquisition unit 1 is used for acquiring a target image through the acquisition camera;
the image judging unit 2 is used for judging whether the target image contains a human face or/and a license plate;
a face processing unit 3, configured to perform preliminary deduplication on a face in the target image when the target image includes the face, determine a face position in the target image after the preliminary deduplication, to extract a target face feature, compare the target face feature with a face feature in a preset face queue, to determine whether the face queue has a face feature that is the same as the target face feature, if yes, remove the target face feature, and if not, add the target face feature into the face queue;
the license plate processing unit 4 is used for detecting the area of a license plate in the target image when the target image contains the license plate, acquiring target license plate information, comparing the target license plate information with license plate information in a preset license plate queue to determine whether the license plate queue has the same license plate information as the target license plate information, if so, rejecting the target license plate information, and if not, adding the target license plate information into the license plate queue;
the matching queue unit 5 is used for sequentially taking out the face features from the face queue, matching the face features with specified face features in a preset database of a police system, sequentially taking out the license plate information from the license plate queue, and matching the license plate information with the specified license plate information in the specified database of the police system;
and the early warning information unit 6 is used for carrying out early warning when the face features are matched with the specified face features or/and the license plate information is matched with the specified license plate information.
In this embodiment, The integral type removes collection equipment can be for collecting such as collection, remove, multiple functions in The wisdom police car of an organic whole such as early warning, be provided with The collection camera On The integral type removes collection equipment, preferably, set up in The top of integral type removes collection equipment, for example, gather The camera and set up in The roof of wisdom police car, then connect integral type removes collection equipment through OTG (On-The-Go) for data transmission, for example, transmit images, because current people's face, picture collection commonly use bayonet socket camera snapshot, individual soldier's equipment is gathered, The position is relatively fixed, do not have The mobility snapshot, remove collection equipment through The integral type, can compensate traditional collection blind area very well.
As the image judging unit 2, the target image acquired by the multiple cameras of the integrated mobile acquisition device can be acquired through the OTG, and then whether the target image includes a face or/and a license plate is judged, specifically, when detecting whether the target image includes a face, the detection can be realized through a standard face template, for example, one or more standard face templates can be designed first, then the matching degree between the image acquired by the test and the standard template is calculated, and whether the face exists is judged through a threshold; or the corresponding rules are generated by extracting the features in the target image, the corresponding face rules can be generated because the face has certain structural distribution features, and whether the target image contains the face can be judged by referring to the face rules; or, identifying whether a face exists in the target image through a preset face classifier, wherein the face classifier is obtained by training and learning through a training set containing face samples and non-face samples based on an artificial neural network; or detecting according to the rule that the distribution of the face complexion in the color space is relatively concentrated so as to determine whether the target image contains the face; or, a face subspace is preset, the face subspace is a set of a plurality of faces, and then the target image is mapped to the face subspace, so that whether the face exists in the target image can be judged based on the distance between the target image and the projection of the target image in the face subspace. It should be noted that the present invention does not limit the detection method, as long as it can detect whether the target image includes a human face.
When detecting whether a license plate exists in a target image, a machine learning algorithm can be adopted to detect the license plate according to information such as texture features, color features and shape features, and specifically, the detection can be completed through a license plate classifier, wherein the license plate classifier is obtained by training and learning through a training set containing license plate samples and non-license plate samples based on an artificial neural network.
As mentioned above, if the image includes a face, the face processing unit 3 performs preliminary deduplication on the face in the target image, so as to subsequently continue to perform face processing on the target image, and then determines the face position in the target image after the preliminary deduplication, so as to extract the target face feature.
When the face features are extracted, the size, position, distance and other attributes of facial image, then calculate their geometric characteristic quantity, form a feature vector describing the face with these characteristic quantity, also is the face features.
And then comparing the target face features with the face features in a preset face queue to perform second duplication removal, so as to reduce the workload of later period, specifically, storing the introduced face features in the face queue or storing the face features within a specified time, firstly calculating the similarity between the target face features and the face features in the face queue, and judging whether the similarity is consistent with the similarity when the similarity exceeds a preset threshold value, so as to determine whether the face queue has the face features same as the target face features, if so, proving that the target face features are introduced, removing the target face features, and if not, adding the target face features into the face queue.
As described in the license plate processing unit 4, if the target image includes a license plate, the region of the license plate in the target image is detected, then the region of the license plate in the target image can be determined through dynamic target tracking, then the target license plate information is acquired, the target license plate information is compared with the license plate information in a preset license plate queue to determine whether the license plate queue has the same license plate information as the target license plate information, if yes, the target license plate information is removed, and if not, the target license plate information is added into the license plate queue. Specifically, the license plate processing unit 4 includes:
a binarization processing subunit, configured to perform image binarization processing on the target image;
and the character extraction subunit is used for performing single extraction on license plate characters in the target image subjected to image binarization processing to obtain a plurality of license plate characters, and taking the license plate characters as the target license plate information.
In this embodiment, an image binarization process is performed on a target image, and then based on a binarization result, information such as structural features of characters, similarity between characters, and intervals between characters is utilized to extract individual characters respectively, so as to obtain single characters of a plurality of license plates, and the single characters are combined to serve as the target license plate information. In another embodiment, the marginalization processing may be performed on the target image, and then the character segmentation may be performed based on the marginalization result.
As described in the matching queue unit 5 and the warning information unit 6, the face features are sequentially extracted from the face queue, and the face features are matched with the specified face features in the preset database of the police service system, where the police service system is an existing police service integrated system in use, and a plurality of pieces of specified person information, such as the specified face features, are stored in the police service integrated system, and the specified person is found by comparing the face features and the specified person information. The police system can be connected through a 4G or 5G network to achieve interconnection and intercommunication with the police system.
And meanwhile, the license plate information can be sequentially taken out from the license plate queue, and the license plate information is matched with the license plate information appointed in the police system appointed database, wherein the license plate information of appointed vehicles such as criminal vehicles or illegal vehicles can be obtained. And when the face features are matched with the designated face features or/and the license plate information is matched with the designated license plate information, triggering an early warning program to start early warning, such as sending out an alarm or sending out reminding information.
In one embodiment, the license plate information is a license plate feature, and the matching queue unit 5 includes:
the normalization processing subunit is used for carrying out normalization processing on the character gray level image;
the license plate feature subunit is used for performing feature extraction on the character gray level image after the normalization processing to obtain license plate features;
and the license plate matching subunit is used for matching the license plate characteristics with the designated license plate information in the designated database, and taking the result with the highest matching degree as the matching result, wherein the designated license plate information comprises a designated character template.
In this embodiment, the license plate information is a segmented character gray image, the character gray image is normalized, the character gray image after the normalization is subjected to feature extraction to obtain license plate features, and the license plate features are matched with a designated character module, where the character module is the designated license plate information in the designated database, and a result with the highest matching degree is used as a matching result.
In an embodiment, the face processing unit includes:
a tracking information obtaining subunit, configured to select the target image captured by the same camera to obtain a tracking id of each target in the target image, where each target carries a corresponding tracking id, and the target includes the face;
a tracking information judging subunit, configured to judge whether there are multiple identical tracking ids in the tracking ids;
and the tracking information removing subunit is configured to remove, if yes, the targets corresponding to the multiple same tracking ids, so that only one target corresponding to one tracking id is reserved in the multiple same tracking ids.
In this embodiment, the targets are objects in a target image, such as people, vehicles, trees, and the like, and may also be set as human faces, each target carries a corresponding tracking id, and the tracking ids of the same target in the same camera are identical, the tracking id is a unique id of each target, and may be generated by a random algorithm or a preset rule, in order to reduce workload of subsequent steps, first performing first deduplication by the tracking id, first selecting a target image shot by the same camera to obtain the tracking id of each target in the target image, and then determining whether the tracking ids have a plurality of identical tracking ids; if so, the tracking id is removed, that is, the target corresponding to the tracking id is removed at the same time, so that only one target corresponding to one tracking is reserved in a plurality of identical tracking ids.
In another embodiment, the foregoing duplication elimination step may be implemented by a collecting camera, where the collecting camera is provided with a duplication elimination program, and when a target image is collected, the collecting camera directly enters the duplication elimination program to perform initial duplication elimination.
According to the early warning device based on the integrated mobile acquisition equipment, provided by the invention, the human face and license plate recognition and comparison algorithm is transplanted into the embedded development board, and the accurate early warning is realized by combining the small snapshot camera which is installed at the top of the integrated mobile acquisition equipment and used for carrying out mobile snapshot, so that a dead zone of fixed positioning snapshot is avoided, and the human face and license plate are simultaneously acquired, so that the screening range is wider, and the data is more accurate by carrying out repeated duplication removal. In addition, the OTG is communicated to transmit the acquired images, and meanwhile, the OTG is matched with a mobile 4G or 5G video network to achieve data interconnection and intercommunication with a police service system, so that the effects of remote command and mobile police service are achieved.
Referring to fig. 3, the present application further provides a computer-readable storage medium 10, in which a computer program 20 is stored, and when the computer program runs on a computer, the computer is caused to execute the early warning method based on the all-in-one mobile acquisition device described in the above embodiment.
In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, cause the processes or functions described in accordance with the embodiments of the application to occur, in whole or in part. The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, from one website site, computer, server, or data center to another website site, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can store or a data storage device, such as a server, a data center, etc., that is integrated with one or more available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., Solid State Disk (SSD)), among others.
The above description is only a preferred embodiment of the present invention, and not intended to limit the scope of the present invention, and all modifications of equivalent structures and equivalent processes, which are made by using the contents of the present specification and the accompanying drawings, or directly or indirectly applied to other related technical fields, are included in the scope of the present invention.
Claims (10)
1. The early warning method based on the integrated mobile acquisition equipment is characterized by being realized by the integrated mobile acquisition equipment, wherein the integrated mobile acquisition equipment comprises an acquisition camera, and the method comprises the following steps:
collecting a target image through the collecting camera;
judging whether the target image contains a human face or/and a license plate;
if the target image contains a face, performing preliminary duplication removal on the face in the target image, determining the face position in the target image after the preliminary duplication removal to extract target face features, comparing the target face features with face features in a preset face queue to determine whether the face queue has the same face features as the target face features, if so, removing the target face features, and if not, adding the target face features into the face queue;
if the target image contains a license plate, detecting the region of the license plate in the target image, acquiring target license plate information, comparing the target license plate information with license plate information in a preset license plate queue to determine whether the license plate queue has the same license plate information as the target license plate information, if so, rejecting the target license plate information, and if not, adding the target license plate information into the license plate queue;
sequentially taking out the face features from the face queue, matching the face features with specified face features in a preset database of a police system, sequentially taking out the license plate information from the license plate queue, and matching the license plate information with the specified license plate information in the specified database of the police system;
and when the face features are matched with the designated face features or/and the license plate information is matched with the designated license plate information, early warning is carried out.
2. The early warning method based on the integrated mobile acquisition equipment as claimed in claim 1, wherein the step of performing preliminary de-duplication on the human face in the target image comprises:
selecting the target image shot by the same collecting camera to obtain the tracking id of each target in the target image, wherein each target carries the corresponding tracking id, and the target comprises the face;
judging whether the tracking id has a plurality of same tracking ids;
if so, removing the targets corresponding to the same tracking ids, so that only one target corresponding to one tracking id is reserved in the same tracking ids.
3. The early warning method based on the integrated mobile acquisition equipment as claimed in claim 1, wherein the step of obtaining the target license plate information comprises:
carrying out image binarization processing on the target image;
and performing single extraction on license plate characters in the target image subjected to image binarization processing to obtain a plurality of license plate characters, wherein the license plate characters are used as the target license plate information.
4. The integrated mobile acquisition device-based early warning method according to claim 3, wherein the license plate information is a segmented character gray image, and the step of matching the license plate information with the designated license plate information in a designated database comprises:
carrying out normalization processing on the character gray level image;
carrying out feature extraction on the character gray level image after the normalization processing to obtain license plate features;
and matching the license plate characteristics with the designated license plate information in the designated database, and taking the result with the highest matching degree as a matching result, wherein the designated license plate information comprises a designated character template.
5. The early warning method based on the integrated mobile acquisition equipment as claimed in claim 1, wherein the step of acquiring the target image by the acquisition camera is preceded by:
and connecting the acquisition camera through an OTG (optical transport group) so as to transmit the target image.
6. The early warning method based on the integrated mobile acquisition device as claimed in claim 4, wherein the steps of sequentially taking out the face features from the face queue, matching the face features with designated face features in a preset database of a police system, sequentially taking out the license plate information from the license plate queue, and matching the license plate information with designated license plate information in a designated database of the police system are preceded by:
and the police system is connected through a 4G or 5G network.
7. The utility model provides an early warning device based on integral type removes collection equipment which characterized in that, removes collection equipment through the integral type and realizes, integral type removes collection equipment including gathering the camera, the device includes:
the image acquisition unit is used for acquiring a target image through the acquisition camera;
the image judging unit is used for judging whether the target image contains a human face or/and a license plate;
a face processing unit, configured to perform preliminary deduplication on a face in the target image when the target image includes the face, determine a face position in the target image after the preliminary deduplication, extract a target face feature, compare the target face feature with a face feature in a preset face queue, and determine whether the face queue has a face feature that is the same as the target face feature, if yes, remove the target face feature, and if not, add the target face feature into the face queue;
the license plate processing unit is used for detecting the area of a license plate in the target image when the target image contains the license plate, acquiring target license plate information, comparing the target license plate information with license plate information in a preset license plate queue to determine whether the license plate queue has the same license plate information as the target license plate information, if so, rejecting the target license plate information, and if not, adding the target license plate information into the license plate queue;
the matching queue unit is used for sequentially taking out the face features from the face queue, matching the face features with specified face features in a preset database of a police system, sequentially taking out the license plate information from the license plate queue, and matching the license plate information with the license plate information specified in the specified database of the police system;
and the early warning information unit is used for carrying out early warning when the face features are matched with the specified face features or/and the license plate information is matched with the specified license plate information.
8. The early warning device based on the integrated mobile acquisition equipment as claimed in claim 7, wherein the face processing unit comprises:
a tracking information obtaining subunit, configured to select the target image captured by the same camera to obtain a tracking id of each target in the target image, where each target carries a corresponding tracking id, and the target includes the face;
a tracking information judging subunit, configured to judge whether there are multiple identical tracking ids in the tracking ids;
and the tracking information removing subunit is configured to remove, if yes, the targets corresponding to the multiple same tracking ids, so that only one target corresponding to the tracking is reserved in the multiple same tracking ids.
9. A readable storage medium, on which a computer program is stored, wherein the computer program, when executed by a processor, implements the steps of the all-in-one mobile acquisition device-based early warning method according to any one of claims 1 to 6.
10. The early warning system based on the integrated mobile acquisition equipment is characterized by comprising the integrated mobile acquisition equipment and an intelligent terminal, wherein the integrated mobile acquisition equipment is connected with the intelligent terminal through an OTG (over the air) network, and the intelligent terminal is connected with a police service system through a 4G or 5G network;
the integrated mobile acquisition equipment comprises an acquisition camera and a processing module used for processing a face and a license plate, wherein a target image is acquired through the acquisition camera, the target image is subjected to duplication elimination through the processing module, target face characteristics and target license plate information are acquired, the target face characteristics are put into a face queue, the target license plate information is put into a license plate queue, and the face characteristics and the license plate information are sequentially taken out from the face queue and the license plate queue and are transmitted to the intelligent terminal;
the intelligent terminal obtains the face features in a preset database through a police system, obtains license plate information in a specified database, matches the face features with the specified face features in the preset database of the police system, matches the license plate information with the specified license plate information in the specified database of the police system, and gives out early warning when the face features are matched with the specified face features or/and the license plate information is matched with the specified license plate information.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110800147.9A CN113591620A (en) | 2021-07-15 | 2021-07-15 | Early warning method, device and system based on integrated mobile acquisition equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202110800147.9A CN113591620A (en) | 2021-07-15 | 2021-07-15 | Early warning method, device and system based on integrated mobile acquisition equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN113591620A true CN113591620A (en) | 2021-11-02 |
Family
ID=78247669
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202110800147.9A Pending CN113591620A (en) | 2021-07-15 | 2021-07-15 | Early warning method, device and system based on integrated mobile acquisition equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN113591620A (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115565227A (en) * | 2022-10-08 | 2023-01-03 | 国家计算机网络与信息安全管理中心 | Intelligent networking automobile data compliance assessment method and device |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109858470A (en) * | 2019-03-06 | 2019-06-07 | 百度在线网络技术(北京)有限公司 | Method and apparatus for output information |
CN110799972A (en) * | 2017-07-21 | 2020-02-14 | 北京市商汤科技开发有限公司 | Dynamic human face image storage method and device, electronic equipment, medium and program |
CN110826390A (en) * | 2019-09-09 | 2020-02-21 | 博云视觉(北京)科技有限公司 | Video data processing method based on face vector characteristics |
CN110909619A (en) * | 2019-10-29 | 2020-03-24 | 汉腾汽车有限公司 | Recognition method based on intelligent police vehicle-mounted camera front-facing image processing |
-
2021
- 2021-07-15 CN CN202110800147.9A patent/CN113591620A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110799972A (en) * | 2017-07-21 | 2020-02-14 | 北京市商汤科技开发有限公司 | Dynamic human face image storage method and device, electronic equipment, medium and program |
CN109858470A (en) * | 2019-03-06 | 2019-06-07 | 百度在线网络技术(北京)有限公司 | Method and apparatus for output information |
CN110826390A (en) * | 2019-09-09 | 2020-02-21 | 博云视觉(北京)科技有限公司 | Video data processing method based on face vector characteristics |
CN110909619A (en) * | 2019-10-29 | 2020-03-24 | 汉腾汽车有限公司 | Recognition method based on intelligent police vehicle-mounted camera front-facing image processing |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN115565227A (en) * | 2022-10-08 | 2023-01-03 | 国家计算机网络与信息安全管理中心 | Intelligent networking automobile data compliance assessment method and device |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US8750573B2 (en) | Hand gesture detection | |
US8792722B2 (en) | Hand gesture detection | |
CN109299683B (en) | Security protection evaluation system based on face recognition and behavior big data | |
US9141184B2 (en) | Person detection system | |
CN109784274B (en) | Method for identifying trailing and related product | |
CN106844492B (en) | A kind of method of recognition of face, client, server and system | |
WO2019127273A1 (en) | Multi-person face detection method, apparatus, server, system, and storage medium | |
CN109766755B (en) | Face recognition method and related product | |
KR101781358B1 (en) | Personal Identification System And Method By Face Recognition In Digital Image | |
CN101300588A (en) | Determining a particular person from a collection | |
CN111723773B (en) | Method and device for detecting carryover, electronic equipment and readable storage medium | |
CN106851229B (en) | Security and protection intelligent decision method and system based on image recognition | |
CN113269091A (en) | Personnel trajectory analysis method, equipment and medium for intelligent park | |
CN109492509A (en) | Personal identification method, device, computer-readable medium and system | |
CN111723656B (en) | Smog detection method and device based on YOLO v3 and self-optimization | |
CN113627321A (en) | Image identification method and device based on artificial intelligence and computer equipment | |
CN111539257A (en) | Personnel re-identification method, device and storage medium | |
CN111126411A (en) | Abnormal behavior identification method and device | |
CN113591620A (en) | Early warning method, device and system based on integrated mobile acquisition equipment | |
JP6463095B2 (en) | Image search apparatus and method | |
CN207020820U (en) | Red-running Monitor System | |
CN113470013A (en) | Method and device for detecting moved article | |
CN112689120A (en) | Monitoring method and device | |
CN114863364B (en) | Security detection method and system based on intelligent video monitoring | |
CN214151721U (en) | Recognition all-in-one machine system based on face algorithm |
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 |