CN105760844A - Video stream data processing method, apparatus and system - Google Patents
Video stream data processing method, apparatus and system Download PDFInfo
- Publication number
- CN105760844A CN105760844A CN201610113373.9A CN201610113373A CN105760844A CN 105760844 A CN105760844 A CN 105760844A CN 201610113373 A CN201610113373 A CN 201610113373A CN 105760844 A CN105760844 A CN 105760844A
- Authority
- CN
- China
- Prior art keywords
- video stream
- stream data
- data
- comprised
- feature
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/41—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items
- G06V20/42—Higher-level, semantic clustering, classification or understanding of video scenes, e.g. detection, labelling or Markovian modelling of sport events or news items of sport video content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
- G06V20/63—Scene text, e.g. street names
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V30/00—Character recognition; Recognising digital ink; Document-oriented image-based pattern recognition
- G06V30/10—Character recognition
- G06V30/14—Image acquisition
- G06V30/148—Segmentation of character regions
- G06V30/153—Segmentation of character regions using recognition of characters or words
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
- H04N7/181—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a plurality of remote sources
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/60—Type of objects
- G06V20/62—Text, e.g. of license plates, overlay texts or captions on TV images
- G06V20/625—License plates
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computational Linguistics (AREA)
- Software Systems (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a video stream data processing method, apparatus and system. The method comprises the steps of receiving a plurality of video stream data transmitted by at least one camera device, each of the video stream data including acquired image data of a set object by the camera device; according to the image data included in each of the video stream data, extracting features of the set object that are included in each of the video stream data and used for representing properties the set object has; and classifying the received plurality of video stream data according to the features. When the acquired image data of each set object by the camera device are received, the features of each set object are extracted and the received video stream data are classified according to the extracted features, the conversion of the video stream data to structural data is effectively achieved, a basis is also laid for subsequently searching a great amount of video stream data for video stream data satisfying a target object, and the efficiency of acquiring video stream data of the target object is effectively improved.
Description
Technical field
The present invention relates to field of computer technology, particularly relate to the processing method of a kind of video stream data, equipment and system.
Background technology
Picture pick-up device is as a kind of video input apparatus, it is possible to is attached by video frequency collection card and computer equipment, has the functions such as video camera, video transmission and still image seizure.Picture pick-up device mainly comprises camera lens, photosensory assembly circuit and controls assembly, wherein, camera lens is for being acquired image, photosensory assembly and control assembly are for processing the image collected, and convert the image collected to digital signal, the image collected can be stored on video frequency collection card by picture pick-up device, it is also possible to the digital signal being converted to is sent to computer equipment.Computer equipment is after receiving digital signal, it is possible to by realizing playing video image to the conversion of digital signal.Therefore, picture pick-up device is widely used in monitoring field.
Specifically, dispose picture pick-up device in the target location to monitor, and data transmission link will be set up between itself and computer equipment.So, when the image of target location is acquired by picture pick-up device, the image of collection can be displayed by computer equipment, to realize the monitoring to target location.Such as, at a certain traffic intersection, picture pick-up device is installed, it is possible to achieve the monitoring to the traffic of this traffic intersection.
But, in actual applications, computer equipment can gather the time sequencing of image according to picture pick-up device and successively the view data received be stored.Under some special screnes, if need the data searching the particular event occurred in target location, need from computer equipment, find the history image data that this target location is corresponding, and the history image data found are analyzed, to obtain the data of the particular event occurred in target location.In this course, if the data volume of the view data searched is relatively larger, causes that the data volume that subsequent image data is analyzed is also relatively large, and then make the efficiency comparison obtaining required data low.
Summary of the invention
In view of this, embodiments provide the processing method of a kind of video stream data, equipment and system, for solve existing by searching data that the image data acquisition that collects of picture pick-up device needs time, the problem that the efficiency comparison of the relatively larger data obtaining needs caused of the quantity of the view data owing to requiring to look up is low.
The invention provides the processing method of a kind of video stream data, including:
Receive multiple video stream datas that at least one picture pick-up device sends, wherein, each described video stream data comprises the view data setting object that described picture pick-up device gathers;
According to the view data comprised in video stream data each described, extracting the feature of the described setting object comprised in each described video stream data, wherein, the feature of described setting object is for characterizing the attribute that described setting object possesses;
According to described feature, the plurality of video stream data received is classified.
The invention provides the process equipment of a kind of video stream data, including:
Receive unit, for receiving multiple video stream datas that at least one picture pick-up device sends, wherein, each described video stream data comprises the view data setting object that described picture pick-up device gathers;
Extraction unit, for according to the view data comprised in video stream data each described, extracting the feature of the described setting object comprised in each described video stream data, wherein, the feature of described setting object is for characterizing the attribute that described setting object possesses;
Taxon, for according to described feature, classifying to the plurality of video stream data received.
The invention provides the process system of a kind of video stream data, including:
At least one picture pick-up device, for gathering the view data setting object, obtain multiple video stream data, and the multiple video stream datas obtained are sent to data handling equipment, wherein, each described video stream data comprises the view data setting object that described picture pick-up device gathers;
Data handling equipment, for receiving multiple video stream datas that at least one picture pick-up device described sends, according to the view data comprised in video stream data each described, extract the feature of the described setting object comprised in each described video stream data, wherein, the feature of described setting object is for characterizing the attribute that described setting object possesses;According to described feature, the plurality of video stream data received is classified.
The present invention has the beneficial effect that:
The embodiment of the present invention receives multiple video stream datas that at least one picture pick-up device sends, and wherein, comprises the view data setting object that described picture pick-up device gathers in each described video stream data;According to the view data comprised in video stream data each described, extracting the feature of the described setting object comprised in each described video stream data, wherein, the feature of described setting object is for characterizing the attribute that described setting object possesses;According to described feature, the plurality of video stream data received is classified.After receiving the view data setting object that picture pick-up device gathers, by extracting the feature of described setting object, and according to the feature extracted, the video stream data received is classified, effectively achieve the video stream data conversion to structural data, also lay the foundation for the follow-up video stream data meeting destination object that finds from substantial amounts of video stream data simultaneously, effectively promote the efficiency of the video stream data obtaining destination object.
Accompanying drawing explanation
In order to be illustrated more clearly that the technical scheme in the embodiment of the present invention, below the accompanying drawing used required during embodiment is described is briefly introduced, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skill in the art, under the premise not paying creative work, it is also possible to obtain other accompanying drawing according to these accompanying drawings.
The schematic flow sheet of the processing method of a kind of video stream data that Fig. 1 provides for the embodiment of the present invention;
A kind of algorithm flow schematic diagram identifying the number-plate number that Fig. 2 provides for the embodiment of the present invention;
The method flow schematic diagram of a kind of type of vehicle identifying described vehicle that Fig. 3 provides for the embodiment of the present invention;
A kind of method flow schematic diagram identifying described vehicle size that Fig. 4 provides for the embodiment of the present invention;
A kind of method flow schematic diagram identifying face characteristic that Fig. 5 provides for the embodiment of the present invention;
The structural representation processing equipment of a kind of video stream data that Fig. 6 provides for the embodiment of the present invention;
The structural representation of the process system of a kind of video stream data that Fig. 7 provides for the embodiment of the present invention.
Detailed description of the invention
In order to realize the purpose of the present invention, the embodiment of the present invention provides the processing method of a kind of video stream data, equipment and system, receive multiple video stream datas that at least one picture pick-up device sends, each described video stream data comprises the view data setting object that described picture pick-up device gathers;According to the view data comprised in video stream data each described, extracting the feature of the described setting object comprised in each described video stream data, wherein, the feature of described setting object is for characterizing the attribute that described setting object possesses;According to described feature, the plurality of video stream data received is classified.
After receiving the view data setting object that picture pick-up device gathers, by extracting the feature of described setting object, and according to the feature extracted, the video stream data received is classified, effectively achieve the video stream data conversion to structural data, also lay the foundation for the follow-up video stream data meeting destination object that finds from substantial amounts of video stream data simultaneously, effectively promote the efficiency of the video stream data obtaining destination object.
Below in conjunction with Figure of description, each embodiment of the present invention is described in further detail, it is clear that described embodiment is only a part of embodiment of the present invention, rather than whole embodiments.Based on the embodiment in the present invention, all other embodiments that those of ordinary skill in the art obtain under not making creative work premise, broadly fall into the scope of protection of the invention.
The schematic flow sheet of the processing method of a kind of video stream data that Fig. 1 provides for the embodiment of the present invention, described method is as described below.
Step 101: receive multiple video stream datas that at least one picture pick-up device sends.
Wherein, each described video stream data comprises the view data setting object that described picture pick-up device gathers.
In a step 101, picture pick-up device can carry out image acquisition to setting object, after the view data obtaining described setting object, it is possible to sending the video stream data comprising the view data of described setting object to data handling equipment, described data handling equipment can receive described video stream data.
It should be noted that, here picture pick-up device refers to the equipment that can image be acquired, can be photographic head, it is also possible to be video camera, be not specifically limited, the number of described picture pick-up device can be one, can also being multiple, be not specifically limited, each described picture pick-up device can send a video stream data to data handling equipment, multiple video stream data can also be sent to data handling equipment, also be not specifically limited here.
Described data handling equipment refers to the equipment that can receive described video stream data and described video stream data is processed, it is possible to be computer equipment, it is also possible to is other smart machines, is not specifically limited.
In actual applications, it is possible to described picture pick-up device is deployed in the setting position needing to carry out image data acquiring, so, the setting object that described setting position can be comprised by described picture pick-up device carries out image acquisition, and obtains the view data of described setting object.Wherein, the number of described setting object can be one, it is also possible to is multiple, is not specifically limited.
At least one picture pick-up device described is after the view data obtaining described setting object, can send multiple video stream datas of the view data comprising described setting object to described data handling equipment, described data handling equipment can receive the plurality of video stream data that at least one picture pick-up device described sends.
Alternatively, after receiving multiple video stream datas that at least one picture pick-up device sends, described method also includes:
Determine the positional information at each described picture pick-up device place;And set up the mapping relations between each positional information with the video stream data sent by the described picture pick-up device on the position corresponding at described positional information.
Described data handling equipment is after receiving multiple video stream datas that at least one picture pick-up device described sends, it is also possible to determine the positional information of each picture pick-up device sending the plurality of video stream data.Wherein, the positional information of at least one picture pick-up device described can be determined by the positional information of at least one picture pick-up device described in the record when disposing at least one picture pick-up device described, the positional information of at least one picture pick-up device described can also be determined by other means, be not specifically limited here.
After the positional information determining at least one picture pick-up device described, the mapping relations between each positional information with the video stream data sent by the described picture pick-up device on the position corresponding at described positional information can be set up, so, after setting up described mapping relations, it is possible to determine, according to described mapping relations, the positional information that each described video stream data is corresponding.
Step 102: according to the view data comprised in video stream data each described, extract the feature of the described setting object comprised in each described video stream data.
Wherein, the feature of described setting object is for characterizing the attribute that described setting object possesses.
In a step 102, described data handling equipment is after receiving described video stream data, it is possible to the view data comprised in each video stream data received is analyzed, to extract the feature of the described setting object comprised in described view data.
It should be noted that in embodiments of the present invention, described setting object can be vehicle, it can also be face, can also is that other set object, but, in actual applications, after getting the view data that picture pick-up device gathers, generally vehicle and face to comprising in view data are analyzed, and obtain the information relevant to vehicle and the information relevant with face, therefore, in embodiments of the present invention, it is possible to illustrate for the described object that sets as vehicle and face.
Described data handling equipment is when the view data comprised according to video stream data each described is analyzed, it is possible to extract the feature of the vehicle comprised in described video stream data, it is also possible to extract the face characteristic comprised in described video stream data.
It should be noted that, in the invention process, the feature of described vehicle can comprise the number-plate number of described vehicle, the type of vehicle of described vehicle and the color of described vehicle, described face characteristic can comprise other features such as facial characteristics, is not specifically limited here.
Separately below for extracting in described video stream data the feature of the vehicle comprised and extracting the face characteristic comprised in described video stream data and illustrate.
First: extract the feature of the vehicle comprised in the plurality of video stream data.
When vehicle is carried out image acquisition by described picture pick-up device, the view data comprising vehicle can be collected, so, described data handling equipment is after receiving the plurality of video stream data, according to the view data of the described vehicle comprised in the plurality of video stream data, the feature of described vehicle can be extracted.
Specifically, according to the view data comprised in video stream data each described, extract the feature of the described setting object comprised in each described video stream data, including:
The view data comprised in video stream data each described is analyzed, extracts the image data comprising vehicle;
Utilize and preset the color of image data described in recognizer identification, and when comprising the number-plate number in determining described image data, utilize the number-plate number comprised in image data described in preset algorithm identification, it is determined that the number-plate number comprised in described image data.
Described data handling equipment can extract the image data comprising described vehicle from the view data of described vehicle, and determines the color of described image data, and wherein, the color of described image data can as the color of described vehicle.
Such as, if it is determined that the color of described image data is red, then, the color of described vehicle is redness.
It should be noted that, in actual applications, each vehicle is including at least a kind of color, therefore, in embodiments of the present invention, when determining the color of described image data, it is possible to obtain at least two color, if it is determined that the color of the described vehicle obtained has at least two kinds, so, can using the one of which color color as described vehicle, it is also possible to using two kinds of colors of at least a part of which as the color of described vehicle, be not specifically limited.
Such as, the color of described image data comprises white and two kinds of colors of black, it is possible to using the black color as vehicle, it is also possible to using the white color as vehicle, it is also possible to using color as vehicle of white and black.
In actual applications, each vehicle has a number-plate number, each number-plate number is different, that is, the number-plate number of vehicle is unique mark of vehicle, therefore, when extracting the image data comprising vehicle, if described image data comprises the number-plate number, it is possible to the described number-plate number is identified.
Specifically, when comprising the number-plate number in determining described image data, it is possible to use described image data is identified by preset algorithm, to determine the described number-plate number.
It should be noted that described preset algorithm can be character recognition algorithm, it is also possible to be that other may identify which and are not specifically limited the algorithm of the described number-plate number.
A kind of algorithm flow schematic diagram identifying the number-plate number that Fig. 2 provides for the embodiment of the present invention.
From figure 2 it can be seen that when extracting the number-plate number of described vehicle, it is possible to it is divided into five steps:
The first step: obtain the image data comprising the number-plate number extracted
Second step: the image data comprising the number-plate number extracted is carried out coarse positioning, positions the described number-plate number;
3rd step: the described number-plate number is accurately positioned;
4th step: the number-plate number that obtains of location is carried out Character segmentation, and utilize the character that character recognition algorithm identification splits;
5th step: output recognition result.
At this point it is possible to determine the number-plate number comprised in described image data.
Alternatively, after the number-plate number determining described vehicle, can also further determining that the kind of described car plate, wherein, the kind of described car plate includes but not limited to: normal domestic car plate, monolayer yellow card, double-deck yellow card, new-type People's Armed Police's car plate, army's board, police car plate, agricultural car plate, port board etc..
Specifically, according to the view data comprised in video stream data each described, extract the feature of the described setting object comprised in each described video stream data, including:
The view data comprised in video stream data each described is analyzed, extracts the data type schema comprising vehicle;
Described data type schema is mated with the auto model data comprised in preset vehicle model database;
According to matching result, identifying the type of vehicle of described vehicle, wherein, described type of vehicle comprises the model of vehicle.
When described data handling equipment comprises the data type schema of described vehicle in determining described video stream data, it is possible to extract the data type schema of described vehicle, and determine the type of vehicle of described vehicle according to the data type schema of described vehicle.
In embodiments of the present invention, described type of vehicle can include but not limited to: the brand (such as: popular etc.) of described vehicle, model (such as: Passat etc.), year money (such as: 2013 sections etc.), classification (such as: minibus, car etc.), etc..
After the data type schema extracting described vehicle, it is possible to the data type schema of described vehicle is mated with the auto model data comprised in described preset vehicle model library, and then identifies the type of vehicle of described vehicle.
It should be noted that, described preset vehicle model library has prestored the auto model data of existing various different vehicle type, so, described data handling equipment is after the model data extracting described vehicle, the data type schema of described vehicle extracted can be mated with the auto model data comprised in preset vehicle model library, and according to matching result identification the type of vehicle of vehicle.
The method flow schematic diagram of a kind of type of vehicle identifying described vehicle that Fig. 3 provides for the embodiment of the present invention.
As can be seen from Figure 3, car face Sample Storehouse can be pre-build, wherein, described car face Sample Storehouse comprises the auto model data of various different vehicle type, so, after the model data extracting described vehicle, the local feature vectors of described vehicle can be determined according to the model data of described vehicle, and the local feature vectors of the local feature vectors of described vehicle with all vehicles comprised in described car face Sample Storehouse is mated, the vehicle maximum with the Similarity of Car Model of described vehicle is found according to matching result, if described similarity is more than setting threshold value, then can using the vehicle maximum for the Similarity of Car Model with described vehicle that the finds vehicle as described vehicle.
It should be noted that in actual applications, it needs to be determined that beyond the type of vehicle of described vehicle, in addition it is also necessary to determine that described vehicle belongs to cart or dolly.
In embodiments of the present invention, it is possible to determine that described vehicle belongs to cart or dolly according to the plurality of video stream data.
Specifically, it is possible to determine that described vehicle belongs to cart or dolly according to the model data of described vehicle.
It should be noted that in embodiments of the present invention, to identifying that described vehicle belongs to cart and still falls within the recognizer of dolly and be not specifically limited.
A kind of method flow schematic diagram identifying described vehicle size that Fig. 4 provides for the embodiment of the present invention.
As can be seen from Figure 4, vehicle sample database can be pre-build, wherein, described vehicle sample database comprises the model data of cart and the model data of dolly, so, after the model data obtaining described vehicle, it is possible to extract the characteristic vector of the vehicle of described vehicle according to the model data of described vehicle, and the characteristic vector of described vehicle is mated with the characteristic vector of the vehicle in vehicle sample database, judge that according to matching result the vehicle of described vehicle is cart or dolly.
Alternatively, after the feature extracting described vehicle, described method also includes:
Feature according to the described vehicle comprised in each the described video stream data extracted, generates the vehicle pictures comprising described feature.
It should be noted that, method according to above-mentioned record, described data handling equipment may determine that three features of described vehicle, the i.e. described number-plate number, the color of described type of vehicle and described vehicle, so, described data handling equipment is when generating vehicle pictures, one of which vehicle characteristics can be selected to generate the vehicle pictures of one of them feature described comprising selection, plurality of feature can also be selected to generate the vehicle pictures of the plurality of feature comprising selection, it is not specifically limited, wherein, the method that can pass through to calculate generates described vehicle pictures, the method that can also pass through analog simulation generates described vehicle pictures, here it is not specifically limited.
Second: extract the face characteristic comprised in the plurality of video stream data.
When face is carried out image acquisition by described picture pick-up device, the view data comprising face can be collected, so, described data handling equipment is after receiving described video stream data, according to the view data of the described face comprised in described video stream data, the face characteristic comprised in the plurality of video stream data can be extracted.
Specifically, according to the view data comprised in video stream data each described, extract the feature of the described setting object comprised in each described video stream data, including:
The view data comprised in video stream data each described is analyzed, extracts the view data comprising face;
The described view data comprising face is mated with the face characteristic data comprised in face database;
According to matching result, identify the face characteristic comprised in described video stream data.
When described data handling equipment comprises the view data of face in determining described video stream data, extract the view data comprising described face, and described view data is mated with the face characteristic data comprised in face database, and the face characteristic comprised in video stream data according to matching result identification.
In actual applications, the characteristic of all with within the scope of setting regions for the view data comprising described face extracted faces can be mated, now, described face database can prestore all of face characteristic within the scope of described setting regions, so, the characteristic of the face that the described view data comprising face prestores with described face database can be mated, the face characteristic comprised in video stream data according to matching result identification, wherein, described setting regions scope can be determined according to practical situation, here it is not especially limited.
A kind of method flow schematic diagram identifying face characteristic that Fig. 5 provides for the embodiment of the present invention.
As shown in Figure 5, face identification method can be divided into: face collection service (RFSS), feature extraction service (FRS), Characteristic Contrast service (FCS), face retrieval service (FSS), real time contrast service (RFCS), wherein:
Face collection services: obtain video flowing, extracts the view data comprising face, and report state is to task scheduling server (VMS), and processes configuring request instruction;
Feature extraction services: the incoming view data comprising face is done Face datection and feature extraction and is stored in data base, returns face characteristic (faceId) result, and report state is to task scheduling server, and processes configuring request instruction;
Characteristic Contrast services: receiving incoming face characteristic, obtain eigenvalue according to face characteristic from data base and do eigenvalue comparison, return 2 eigenvalue similarities, report state is to task scheduling server and processes configuring request instruction;
Face retrieval services: receive incoming face characteristic, obtains eigenvalue according to face characteristic from data base and does face retrieval, returns retrieval result sequence, and report state is to task scheduling server and processes configuring request instruction;
Real-time comparison service: receive incoming face characteristic, obtains eigenvalue according to face characteristic from data base and does face contrast, return comparing result, and report state is to task scheduling server and processes configuring request instruction.
Step 103: according to described feature, the plurality of video stream data received is classified.
In step 103, described data handling equipment is after the feature extracting described setting object, it is possible to according to described feature, the described video stream data received is classified.
Alternatively, during the feature setting object comprised in extracting each described video stream data, the mapping relations between the feature of each described video stream data and described setting object are set up respectively.
Described data handling equipment is when extracting the feature of described setting object, it is possible to set up the mapping relations between each described video stream data and described feature.
Based in above-mentioned steps 102 record content, the feature of described setting object can be the feature of described vehicle, it is also possible to be described face characteristic.
It should be noted that, the number of the feature of the described setting object that described data handling equipment is extracted is probably one, it is also possible to multiple, then, in the mapping relations set up between each described video stream data and described feature, it is possible to be respectively directed to above-mentioned two situations and set up described mapping relations.
When determining that the number of the described feature setting object extracted is as 1, it is possible to set up the mapping relations between described feature and each described video stream data.
Such as, " color of vehicle is redness " it is characterized as described in, then, it is possible to set up the mapping relations between the described video stream data comprising red vehicle and " color of vehicle is red ".
When determining that the Characteristic Number of the described setting object extracted is more than 1, for each feature, perform following operation respectively:
Select at least one of which feature, using the video stream data of the described feature that comprises selection as same class, and set up the mapping relations between the described feature of selection and the video stream data of the described feature that comprises selection.
That is, when determining the number of feature of the described setting object extracted be more than or equal to 2, the mapping relations between one of them feature and the described video stream data comprising one feature can be set up, it is also possible to set up multiple (be more than or equal to 2) mapping relations between feature and the described video stream data comprising the plurality of feature.
Such as, it is characterized as " color of vehicle is red " and " type of vehicle of vehicle is car " described in determining, so, the mapping relations between the described video stream data comprising red vehicle and " color of vehicle is red " can be set up, the mapping relations between the described video stream data comprising car and " type of vehicle of vehicle is car " can also be set up, it is also possible to set up the mapping relations between the described video stream data comprising red car and " color of vehicle is red " and " type of vehicle of vehicle is car ".
Alternatively, described method also includes:
After the plurality of video stream data received is classified, store the plurality of video stream data according to classification results.
Described data handling equipment is after classifying according to the feature of the described setting object extracted, according to classification results, the plurality of video stream data can be stored to described data handling equipment, so, when requiring to look up the data of described setting object, it is possible to be quickly found out the view data corresponding with described feature according to the feature of described setting object.
Alternatively, when storing each described video stream data according to classification results, it is also possible to the positional information of each described video stream data is stored.
It should be noted that, in a step 101, described data handling equipment has set up the mapping relations between described positional information with the video stream data sent by the described picture pick-up device on the position corresponding at described positional information, therefore, when storing each described video stream data according to classification results, can according to the mapping relations between described video stream data and described positional information, determine the positional information of each described video stream data, and according to described feature, positional information corresponding to the video stream data comprising described feature and described video stream data is stored.
The embodiment of the present invention receives multiple video stream datas that at least one picture pick-up device sends, and wherein, comprises the view data setting object that described picture pick-up device gathers in each described video stream data;According to the view data comprised in video stream data each described, extracting the feature of the described setting object comprised in each described video stream data, wherein, the feature of described setting object is for characterizing the attribute that described setting object possesses;According to described feature, the plurality of video stream data received is classified.After receiving the view data setting object that picture pick-up device gathers, by extracting the feature of described setting object, and according to the feature extracted, the video stream data received is classified, effectively achieve the video stream data conversion to structural data, also lay the foundation for the follow-up video stream data meeting destination object that finds from substantial amounts of video stream data simultaneously, effectively promote the efficiency of the video stream data obtaining destination object.
The structural representation processing equipment of a kind of video stream data that Fig. 6 provides for the embodiment of the present invention.The process equipment of described video stream data includes: receive unit 61, extraction unit 62, taxon 63, set up unit 64, memory element 65 and generate unit 66, wherein:
Receive unit 61, for receiving multiple video stream datas that at least one picture pick-up device sends, wherein, each described video stream data comprises the view data setting object that described picture pick-up device gathers;
Extraction unit 62, for according to the view data comprised in video stream data each described, extracting the feature of the described setting object comprised in each described video stream data, wherein, the feature of described setting object is for characterizing the attribute that described setting object possesses;
Taxon 63, for according to described feature, classifying to the plurality of video stream data received.
Alternatively, the process equipment of described video stream data also includes: set up unit 64, wherein
Described set up unit 64, for when described extraction unit 62 extracts the feature setting object comprised in each described video stream data, setting up the mapping relations between the feature of each described video stream data and described setting object respectively.
Specifically, the plurality of video stream data received is classified by described taxon 63, including:
When determining that the Characteristic Number of the described setting object extracted is more than 1, for each feature, perform following operation respectively:
Select at least one of which feature, using the video stream data of the described feature that comprises selection as same class, and set up the mapping relations between the described feature of selection and the video stream data of the described feature that comprises selection.
Alternatively, the process equipment of described video stream data also includes: memory element 65, wherein:
Described memory element 65, for, after the plurality of video stream data received is classified by described taxon 63, storing the plurality of video stream data according to classification results.
Alternatively, described set up unit 64, be additionally operable to after described reception unit 61 receives multiple video stream datas that at least one picture pick-up device sends, it is determined that the positional information at each described picture pick-up device place;And set up the mapping relations between each positional information with the video stream data sent by the described picture pick-up device on the position corresponding at described positional information.
Described extraction unit 62, according to the view data comprised in video stream data each described, extracts the feature of the described setting object comprised in each described video stream data, including:
The view data comprised in video stream data each described is analyzed, extracts the image data comprising the number-plate number;
Utilize preset algorithm that the described image data comprising the number-plate number is identified, it is determined that the number-plate number comprised in described image data.
Described extraction unit 62, according to the view data comprised in video stream data each described, extracts the feature of the described setting object comprised in each described video stream data, including:
The view data comprised in video stream data each described is analyzed, extracts the data type schema comprising vehicle;
Described data type schema is mated with the auto model data comprised in preset vehicle model database;
According to matching result, identifying the type of vehicle of described vehicle, wherein, described type of vehicle comprises the model of vehicle.
Described extraction unit 62, according to the view data comprised in video stream data each described, extracts the feature of the described setting object comprised in each described video stream data, including:
The view data comprised in video stream data each described is analyzed, extracts the image data comprising vehicle;
Utilize and preset the color of image data described in recognizer identification.
Alternatively, described video stream data processes equipment and also includes: generate unit 66, wherein:
Generate unit 66, the feature of the described vehicle for comprising in each the described video stream data according to the extraction of described extraction unit 62, generate the vehicle pictures comprising described feature.
Alternatively, described set up unit 64, it is additionally operable to, according to the image data comprised in described vehicle pictures, from the video stream data received, select the video stream data comprising described image data, and set up the mapping relations between described vehicle pictures and the video stream data of selection.
Described extraction unit 62, according to the view data comprised in video stream data each described, extracts the feature of the described setting object comprised in each described video stream data, including:
The view data comprised in video stream data each described is analyzed, extracts the view data comprising face;
The described view data comprising face is mated with the face characteristic data comprised in face database;
According to matching result, identify the face characteristic comprised in described video stream data;
The plurality of video stream data received, according to described feature, is classified by described taxon 63, including:
According to described face characteristic, the plurality of video stream data received is classified
It should be noted that the process equipment of the video stream data of embodiment of the present invention offer can be realized by hardware mode, it is also possible to realized by software mode, do not limit here.
The structural representation of the process system of a kind of video stream data that Fig. 7 provides for the embodiment of the present invention.Described system includes: at least one picture pick-up device 711~71n, data handling equipment 72 and storage device 73, wherein:
At least one picture pick-up device 711~71n, for gathering the view data setting object, obtain multiple video stream data, and the multiple video stream datas obtained are sent to data handling equipment, wherein, each described video stream data comprises the view data setting object that described picture pick-up device gathers;
Data handling equipment 72, for receiving described at least one picture pick-up device 711~71n multiple video stream datas sent, according to the view data comprised in video stream data each described, extract the feature of the described setting object comprised in each described video stream data, wherein, the feature of described setting object is for characterizing the attribute that described setting object possesses;According to described feature, the plurality of video stream data received is classified.
Alternatively, described video stream data processes system and also includes: storage device 73, wherein:
Described storage device 73, for the classification results according to described data handling equipment, stores the plurality of video stream data.
Specifically, described data handling equipment 72 also includes: vehicle characteristics analysis module 721 and Facial Feature Analysis module 722, wherein:
Described vehicle analysis module 721, for the view data of the vehicle comprised at least one picture pick-up device described 711~71n multiple video stream datas sent is analyzed, extracts the feature of described vehicle;
Described Facial Feature Analysis module 722, for the view data of the face comprised at least one picture pick-up device described 711~71n multiple video stream datas sent is analyzed, extracts the face characteristic comprised in the plurality of video stream data.
Alternatively, described vehicle analysis module 721 also includes: number-plate number analysis module 7211, type of vehicle analysis module 7212 and vehicle color analysis module 7213, wherein:
Described number-plate number analysis module 7211, for the view data of the vehicle comprised at least one picture pick-up device described 711~71n multiple video stream datas sent is analyzed, extracts the number-plate number of described vehicle;
Described type of vehicle analysis module 7212, for the view data of the vehicle comprised at least one picture pick-up device described 711~71n multiple video stream datas sent is analyzed, extracts the type of vehicle of described vehicle;
Described vehicle color analysis module 7213, for the view data of the vehicle comprised at least one picture pick-up device described 711~71n multiple video stream datas sent is analyzed, extract the image data comprising described vehicle, and identify the color of described image data.
It will be understood by those skilled in the art that embodiments of the invention can be provided as method, device (equipment) or computer program.Therefore, the present invention can adopt the form of complete hardware embodiment, complete software implementation or the embodiment in conjunction with software and hardware aspect.And, the present invention can adopt the form at one or more upper computer programs implemented of computer-usable storage medium (including but not limited to disk memory, CD-ROM, optical memory etc.) wherein including computer usable program code.
The present invention is that flow chart and/or block diagram with reference to method according to embodiments of the present invention, device (equipment) and computer program describe.It should be understood that can by the combination of the flow process in each flow process in computer program instructions flowchart and/or block diagram and/or square frame and flow chart and/or block diagram and/or square frame.These computer program instructions can be provided to produce a machine to the processor of general purpose computer, special-purpose computer, Embedded Processor or other programmable data processing device so that the instruction performed by the processor of computer or other programmable data processing device is produced for realizing the device of function specified in one flow process of flow chart or multiple flow process and/or one square frame of block diagram or multiple square frame.
These computer program instructions may be alternatively stored in and can guide in the computer-readable memory that computer or other programmable data processing device work in a specific way, the instruction making to be stored in this computer-readable memory produces to include the manufacture of command device, and this command device realizes the function specified in one flow process of flow chart or multiple flow process and/or one square frame of block diagram or multiple square frame.
These computer program instructions also can be loaded in computer or other programmable data processing device, make on computer or other programmable devices, to perform sequence of operations step to produce computer implemented process, thus the instruction performed on computer or other programmable devices provides for realizing the step of function specified in one flow process of flow chart or multiple flow process and/or one square frame of block diagram or multiple square frame.
Although preferred embodiments of the present invention have been described, but those skilled in the art are once know basic creative concept, then these embodiments can be made other change and amendment.So, claims are intended to be construed to include preferred embodiment and fall into all changes and the amendment of the scope of the invention.
Obviously, the present invention can be carried out various change and modification without deviating from the scope of the present invention by those skilled in the art.So, if these amendments of the present invention and modification belong within the scope of the claims in the present invention and equivalent technologies thereof, then the present invention is also intended to comprise these change and modification.
Claims (10)
1. the processing method of a video stream data, it is characterised in that including:
Receive multiple video stream datas that at least one picture pick-up device sends, wherein, each described video stream data comprises the view data setting object that described picture pick-up device gathers;
According to the view data comprised in video stream data each described, extracting the feature of the described setting object comprised in each described video stream data, wherein, the feature of described setting object is for characterizing the attribute that described setting object possesses;
According to described feature, the plurality of video stream data received is classified.
2. processing method as claimed in claim 1, it is characterised in that described method also includes:
During the feature setting object comprised in extracting each described video stream data, set up the mapping relations between the feature of each described video stream data and described setting object respectively.
3. processing method as claimed in claim 1, it is characterised in that after receiving multiple video stream datas that at least one picture pick-up device sends, described method also includes:
Determine the positional information at each described picture pick-up device place;And set up the mapping relations between each positional information with the video stream data sent by the described picture pick-up device on the position corresponding at described positional information.
4. processing method as claimed in claim 1, it is characterised in that according to the view data comprised in video stream data each described, extract the feature of the described setting object comprised in each described video stream data, including:
The view data comprised in video stream data each described is analyzed, extracts the image data comprising vehicle;
Utilize and preset the color of image data described in recognizer identification, and when comprising the number-plate number in determining described image data, utilize the number-plate number comprised in image data described in preset algorithm identification, it is determined that the number-plate number comprised in described image data.
5. processing method as claimed in claim 1, it is characterised in that according to the view data comprised in video stream data each described, extract the feature of the described setting object comprised in each described video stream data, including:
The view data comprised in video stream data each described is analyzed, extracts the data type schema comprising vehicle;
Described data type schema is mated with the auto model data comprised in preset vehicle model database;
According to matching result, identifying the type of vehicle of described vehicle, wherein, described type of vehicle comprises the model of vehicle.
6. processing method as described in claim 4 or 5, it is characterised in that described method also includes:
Feature according to the described vehicle comprised in each the described video stream data extracted, generates the vehicle pictures comprising described feature.
7. processing method as claimed in claim 6, it is characterised in that described method also includes:
According to the image data comprised in described vehicle pictures, from the video stream data received, select the video stream data comprising described image data, and set up the mapping relations between described vehicle pictures and the video stream data of selection.
8. processing method as claimed in claim 1, it is characterised in that according to the view data comprised in video stream data each described, extract the feature of the described setting object comprised in each described video stream data, including:
The view data comprised in video stream data each described is analyzed, extracts the view data comprising face;
The described view data comprising face is mated with the face characteristic data comprised in face database;
According to matching result, identify the face characteristic comprised in described video stream data;
According to described feature, the plurality of video stream data received is classified, including:
According to described face characteristic, the plurality of video stream data received is classified.
9. the process equipment of a video stream data, it is characterised in that including:
Receive unit, for receiving multiple video stream datas that at least one picture pick-up device sends, wherein, each described video stream data comprises the view data setting object that described picture pick-up device gathers;
Extraction unit, for according to the view data comprised in video stream data each described, extracting the feature of the described setting object comprised in each described video stream data, wherein, the feature of described setting object is for characterizing the attribute that described setting object possesses;
Taxon, for according to described feature, classifying to the plurality of video stream data received.
10. the process system of a video stream data, it is characterised in that including:
At least one picture pick-up device, for gathering the view data setting object, obtain multiple video stream data, and the multiple video stream datas obtained are sent to data handling equipment, wherein, each described video stream data comprises the view data setting object that described picture pick-up device gathers;
Data handling equipment, for receiving multiple video stream datas that at least one picture pick-up device described sends, according to the view data comprised in video stream data each described, extract the feature of the described setting object comprised in each described video stream data, wherein, the feature of described setting object is for characterizing the attribute that described setting object possesses;According to described feature, the plurality of video stream data received is classified.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610113373.9A CN105760844A (en) | 2016-02-29 | 2016-02-29 | Video stream data processing method, apparatus and system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610113373.9A CN105760844A (en) | 2016-02-29 | 2016-02-29 | Video stream data processing method, apparatus and system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN105760844A true CN105760844A (en) | 2016-07-13 |
Family
ID=56330478
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610113373.9A Pending CN105760844A (en) | 2016-02-29 | 2016-02-29 | Video stream data processing method, apparatus and system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105760844A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106485241A (en) * | 2016-12-07 | 2017-03-08 | 云赛智联股份有限公司 | A kind of data analysis system |
CN107682712A (en) * | 2017-09-30 | 2018-02-09 | 安徽联智创新软件有限公司 | A kind of net cast network plug-flow management system |
CN108509827A (en) * | 2017-02-27 | 2018-09-07 | 阿里巴巴集团控股有限公司 | The recognition methods of anomalous content and video flow processing system and method in video flowing |
CN108984799A (en) * | 2018-08-21 | 2018-12-11 | 北京深瞐科技有限公司 | A kind of video data handling procedure and device |
CN109003318A (en) * | 2018-08-09 | 2018-12-14 | 北京深瞐科技有限公司 | A kind of method for processing video frequency, device and storage medium |
CN110119456A (en) * | 2019-05-14 | 2019-08-13 | 深圳市商汤科技有限公司 | Retrieve the method and device of image |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103198110A (en) * | 2013-03-28 | 2013-07-10 | 广州中国科学院软件应用技术研究所 | Method and system for rapid video data characteristic retrieval |
CN103235956A (en) * | 2013-03-28 | 2013-08-07 | 天脉聚源(北京)传媒科技有限公司 | Method and device for detecting advertisements |
CN103916626A (en) * | 2013-01-05 | 2014-07-09 | 中兴通讯股份有限公司 | Monitoring video information providing method and device and video monitoring system |
-
2016
- 2016-02-29 CN CN201610113373.9A patent/CN105760844A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103916626A (en) * | 2013-01-05 | 2014-07-09 | 中兴通讯股份有限公司 | Monitoring video information providing method and device and video monitoring system |
CN103198110A (en) * | 2013-03-28 | 2013-07-10 | 广州中国科学院软件应用技术研究所 | Method and system for rapid video data characteristic retrieval |
CN103235956A (en) * | 2013-03-28 | 2013-08-07 | 天脉聚源(北京)传媒科技有限公司 | Method and device for detecting advertisements |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106485241A (en) * | 2016-12-07 | 2017-03-08 | 云赛智联股份有限公司 | A kind of data analysis system |
CN108509827A (en) * | 2017-02-27 | 2018-09-07 | 阿里巴巴集团控股有限公司 | The recognition methods of anomalous content and video flow processing system and method in video flowing |
CN108509827B (en) * | 2017-02-27 | 2022-07-08 | 阿里巴巴集团控股有限公司 | Method for identifying abnormal content in video stream and video stream processing system and method |
CN107682712A (en) * | 2017-09-30 | 2018-02-09 | 安徽联智创新软件有限公司 | A kind of net cast network plug-flow management system |
CN109003318A (en) * | 2018-08-09 | 2018-12-14 | 北京深瞐科技有限公司 | A kind of method for processing video frequency, device and storage medium |
CN108984799A (en) * | 2018-08-21 | 2018-12-11 | 北京深瞐科技有限公司 | A kind of video data handling procedure and device |
CN110119456A (en) * | 2019-05-14 | 2019-08-13 | 深圳市商汤科技有限公司 | Retrieve the method and device of image |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN105760844A (en) | Video stream data processing method, apparatus and system | |
US11321583B2 (en) | Image annotating method and electronic device | |
CN108596277B (en) | Vehicle identity recognition method and device and storage medium | |
KR101289085B1 (en) | Images searching system based on object and method thereof | |
CN105320705B (en) | The search method and device of similar vehicle | |
CN103942811B (en) | Distributed parallel determines the method and system of characteristic target movement locus | |
CN105321350B (en) | Fake-licensed car detection method and device | |
CN110235138A (en) | System and method for appearance search | |
CN107885430B (en) | Audio playing method and device, storage medium and electronic equipment | |
KR101788225B1 (en) | Method and System for Recognition/Tracking Construction Equipment and Workers Using Construction-Site-Customized Image Processing | |
Li et al. | An anti-fraud system for car insurance claim based on visual evidence | |
CN101840422A (en) | Intelligent video retrieval system and method based on target characteristic and alarm behavior | |
CN111695495A (en) | Face recognition method, electronic device and storage medium | |
KR20200069911A (en) | Method and apparatus for identifying object and object location equality between images | |
CN109146923B (en) | Processing method and system for target tracking broken frame | |
CN111353504B (en) | Source camera identification method based on image block diversity selection and residual prediction module | |
CN113516102A (en) | Deep learning parabolic behavior detection method based on video | |
CN103970901A (en) | Geographic information graphic data integration method | |
CN114627394A (en) | Muck vehicle fake plate identification method and system based on unmanned aerial vehicle | |
WO2022222036A1 (en) | Method and apparatus for determining parking space | |
CN112802027A (en) | Target object analysis method, storage medium and electronic device | |
Špaňhel et al. | Vehicle fine-grained recognition based on convolutional neural networks for real-world applications | |
CN113743359B (en) | Vehicle re-identification method, model training method and related devices | |
CN112257666B (en) | Target image content aggregation method, device, equipment and readable storage medium | |
CN211044247U (en) | Video monitoring system for hotels |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20160713 |