CN104239881A - Method and system for automatically finding and registering target in surveillance video - Google Patents
Method and system for automatically finding and registering target in surveillance video Download PDFInfo
- Publication number
- CN104239881A CN104239881A CN201310231645.1A CN201310231645A CN104239881A CN 104239881 A CN104239881 A CN 104239881A CN 201310231645 A CN201310231645 A CN 201310231645A CN 104239881 A CN104239881 A CN 104239881A
- Authority
- CN
- China
- Prior art keywords
- target
- target signature
- described target
- signature
- warning list
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Abstract
The invention relates to the field of intelligent security protection, and discloses a method and system for automatically finding and registering a target in surveillance video. The method system for automatically finding and registering the target in the surveillance video comprises the following steps that the target is detected in a video stream; feature extraction is carried out on the detected target so that the target feature is obtained; whether the target feature has already existed in a target list or not is judged, if yes, the frequency of appearance of the target feature is increased by one, and if not, the target feature is added into the target list, and the frequency of the appearance of the target feature is set as one; whether the frequency of appearance of the target feature exceeds a preset threshold value or not is judged, and if yes, the target feature is arranged in an alarm list in a registration mode. On the basis of a comparison system, the key target which appears repeatedly is automatically found and automatically arranged in the alarm list in a registration mode to serve as an object which is analyzed and processed subsequently in a focused mode, and the common comparison system is more intelligent and practical.
Description
Technical field
The present invention relates to intelligent security guard field, particularly in a kind of monitor video, target finds and automatic registration method automatically.
Background technology
Real-time comparison is a kind of application mode of intelligent video analysis technology, target is detected from video flowing, and compare to the warning list in system with registered in advance, the threshold value reaching setting then sends warning, its typical apply is bayonet fashion, namely the passing target image of system real-time grasp shoot be set in important channel and report to the police with the real-time comparison of warning list of registration, as being applied in the face Gate System etc. of the positions such as the vehicle card port system on traffic main artery and railway station, bus station, important place gateway.
Warning list: the list to be compared in real time in Compare System, registered images in the target image detected and warning list carries out comparison one by one, once reach the alarm threshold value of setting, carry out reporting to the police (due to the singularity of car plate, the Word message with clear and definite implication can be identified as and then compare).Corresponding is white list with it, and the target detected and target to be compared are compared, and has any one (sometimes require and only have one) to reach comparison threshold value and then passes through, otherwise alarm.
Warning list is registered: by the process of image registration to be compared to real-time Compare System.
Real-time comparison application is one of the typical apply mode in intelligent security guard field, in traditional real-time Compare System, warning list is add to system by the mode of registration (importing or incremental update containing from other system batch) mostly, and the flow of personnel of important place is very rapidly, only according to traditional warning list logon mode, requirement that is intelligent and real-time can not be met.
Summary of the invention
The method that the object of the present invention is to provide target in a kind of monitor video automatically to find and register and system thereof, make Compare System can meet requirement that is intelligent and real-time.
For solving the problems of the technologies described above, embodiments of the present invention disclose target in a kind of monitor video and automatically find and register method, comprise the following steps:
Target is detected from video flowing;
Feature extraction is carried out to the target detected, obtains target signature;
Judge whether target signature has been present in object listing, if so, then the occurrence number of target signature is added 1; If not, then target signature is joined in object listing, and the occurrence number of target signature is set to 1;
Judge whether the occurrence number of target signature exceedes the threshold value preset, if exceed, then target signature is registered in warning list.
Embodiments of the present invention also disclose target in a kind of monitor video and automatically find and Accreditation System, comprising:
Detecting unit, for detecting target from video flowing;
Target's feature-extraction unit, carries out feature extraction for the target detected detecting unit, obtains target signature;
First judging unit, for judging whether target signature has been present in object listing;
Counting unit, for when the first judging unit judges that target signature is present in object listing, adds 1 by the occurrence number of target signature; When the first judging unit judges that target signature is not present in object listing, target signature is joined in object listing, and the occurrence number of target signature is set to 1;
Second judging unit, for judging whether the occurrence number of the target signature that counting unit exports exceedes the threshold value preset;
Registering unit, for when the second judging unit judges that the occurrence number of target signature exceedes the threshold value preset, is registered to target signature in warning list.
Compared with prior art, the key distinction and effect thereof are embodiment of the present invention:
On the basis of Compare System, the common-denominator target repeated found automatically and is automatically registered in warning list, as the object of follow-up selective analysis and process, making general Compare System more intelligent and practical.
Further, after destination object is registered to warning list data storehouse, deletes the information of destination object in object listing, the storage space in object listing can be saved further, reduce the workload of follow-up comparison.
Further, destination object is registered to warning list data storehouse can select real-time registration, and namely frequency of occurrence is once reach the upper limit, is just registered to warning list in real time; Also non real-time registration can be selected, namely the upper limit of frequency of occurrence in period of time T is set in, the number of times that in accumulative this period, all destination objects repeat, when reached during cycle T, the destination object of all eligible (multiplicity exceedes the upper limit) is registered to warning list data storehouse, and system implementations is convenient, flexible.
Accompanying drawing explanation
Fig. 1 is that in first embodiment of the invention, in a kind of monitor video, target finds the schematic flow sheet with register method automatically;
Fig. 2 is that in second embodiment of the invention, in a kind of monitor video, target finds the schematic flow sheet with register method automatically;
Fig. 3 is that in third embodiment of the invention, in a kind of monitor video, target finds the schematic flow sheet with register method automatically;
Fig. 4 is that in four embodiment of the invention, in a kind of monitor video, target finds the schematic flow sheet with register method automatically;
Fig. 5 is that in four embodiment of the invention, in a kind of monitor video, target finds the schematic flow sheet with register method automatically;
Fig. 6 is that in fifth embodiment of the invention, in a kind of monitor video, target finds the schematic flow sheet with register method automatically;
Fig. 7 is that in sixth embodiment of the invention, in a kind of monitor video, target finds the structural representation with Accreditation System automatically.
Embodiment
In the following description, many ins and outs are proposed in order to make reader understand the application better.But, persons of ordinary skill in the art may appreciate that even without these ins and outs with based on the many variations of following embodiment and amendment, also can realize each claim of the application technical scheme required for protection.
For making the object, technical solutions and advantages of the present invention clearly, below in conjunction with accompanying drawing, embodiments of the present invention are described in further detail.
First embodiment of the invention relates to target in a kind of monitor video and automatically finds and register method.Fig. 1 is that in this monitor video, target finds the schematic flow sheet with register method automatically.
Specifically, as shown in Figure 1, in this monitor video, target finds to comprise the following steps with register method automatically:
In a step 101, from video flowing, target is detected.
System carries out video acquisition, by sensor, vision signal is converted to the digital signal can carrying out analyzing.System carries out target detection to digital video signal, so that the further process of system.
In order to understand the present invention better, as the preferred implementation of one of the present invention, hereafter will be applied as a kind of implementation of the present invention with the real-time comparison of face, the complete procedure that detail system realizes, but be not limited to the present invention.
The image acquisition of system, target detection adopts face snap machine to realize, face snap machine is the IP video camera with face snap function, specific implementation is real-time detection and tracking face from the video code flow gathered, and a top-quality picture is selected to same person upload, there is the advantage that frontal chromatography ensures real-time and reduces volume of transmitted data, camera exposure parameter can be carried out according to target automatically to regulate simultaneously, thus better target image can be obtained, concrete, Face datection can adopt the Boosting method of Haar feature to realize, face tracking can adopt the method for light stream to realize, face is selected to comprise and is carried out comprehensive evaluation to the sharpness of facial image and attitude two indices.
After this enter step 102, feature extraction is carried out to the target detected, obtains target signature.
This step comprises necessary step such as the pre-service of feature extraction necessary target image, illumination and geometrical normalization, feature extraction and conversion etc.
Such as, the characteristic extraction step of face is realized in the following manner, first in facial image, eyes coordinate is located, then eyes coordinate is snapped to coordinate (48, 48) and (112, 48), obtain the facial image through geometrical normalization, the method of entropy diagram picture is adopted to carry out unitary of illumination to it again, then the LBP characteristic sum Gabor characteristic (method of feature selecting uses the method for Boosting to select the feature being conducive to most realizing face classification) through selecting is extracted, finally the method for Metric Learning is adopted to carry out further dimensionality reduction and the measure space conversion of feature further to the feature extracted, make the most applicable similarity degree being used for calculating two facial images of feature.
The concrete enforcement technology of the present invention to target's feature-extraction and similarity comparison is not specifically limited, reason mainly contains two: one feature extraction and comparison technology belong to the content of existing real-time comparison, its two relevant feature extraction and alignment algorithm mutation very many.Such as: what adopt herein is that Gabor+LBP compound characteristics characterizes face, adopt Metric Learning to obtain metric matrix and improve Ma Shi distance calculating human face similarity degree.
Preferably, the target of detection comprises vehicle and face, but not as limit, in some other embodiment of the present invention, can also comprise the discovery of other targets such as human body, gait, animal and automatically register.
In addition, be appreciated that the target signature of carrying out feature extraction gained to the target detected can be one, such as license plate number also can be multiple combination, the combination of such as, multiple features in face, combination of car plate and vehicle color etc.
After this enter step 103, judge whether target signature has been present in object listing.
If so, then step 104 is entered; If not, then step 105 is entered.
Judge that the mode of target whether in object listing or warning list has two kinds: a kind of as face, be by extract target signature and calculate similarity and and the threshold value set compare; Another kind of as car plate, if face is changed to car plate, then can be described accurately by license plate recognition technology (being equivalent to feature), comparison then can be converted to table look-up and determine whether in object listing or warning list.
At step 104, the occurrence number of target signature is added 1.
After this step 106 is entered.
In step 105, target signature is joined in object listing, and the occurrence number of target signature is set to 1.
After this step 106 is entered.
Judge whether described target signature has been present in object listing, if so, then the occurrence number of described target signature is added 1; If not, then described target signature is joined in described object listing, and the occurrence number of described target signature is set to 1.
In step 106, judge whether the occurrence number of target signature exceedes the threshold value preset.
If exceed, then enter step 107; If be no more than, then again get back to step 101.
It should be noted that, number of times threshold value can preset in systems in which, and general employing is advisable for 3-10 time.
In step 107, target signature is registered in warning list.
After this process ends.
Target signature is registered to warning list data storehouse can select real-time registration, and namely frequency of occurrence is once reach the upper limit, is just registered to warning list in real time.Also non real-time registration can be selected, the i.e. upper limit of frequency of occurrence in setting-up time cycle T, the number of times that in accumulative this period, all destination objects repeat, when reached during cycle T, be registered to warning list data storehouse to the destination object of all eligible (multiplicity exceedes the upper limit), system implementations is convenient, flexible.
Step 101,102 is the step identical with real-time Compare System general in prior art, thereafter be the step with general real-time Compare System difference to some extent, on the basis of real-time Compare System, namely add automatic discovery and the registration of warning list of common-denominator target.
On the basis of real-time Compare System, the common-denominator target repeated found automatically and is automatically registered in warning list, as the object of follow-up selective analysis and process, making general real-time Compare System more intelligent and practical.
Second embodiment of the invention relates to target in a kind of monitor video and automatically finds and register method.Fig. 2 is that in this monitor video, target finds the schematic flow sheet with register method automatically.
Second embodiment improves on the basis of the first embodiment, and specifically, as shown in Figure 2, main improvements are:
First, before target signature is registered in warning list by step 107, also comprise:
Step 108: judge whether target signature has been present in warning list.
If not, then step 107 is performed; If so, then process ends.
Step 108 confirms whether registered this target in warning list.If with the alignment similarity of whole registration pictures of warning list all lower than setting threshold value, namely there is not this target signature, then perform step 107, this target signature is registered in warning list.
Secondly, after target signature is registered in warning list by step 107, also comprise:
Step 109: delete the record of target signature in object listing.
After destination object is registered to warning list data storehouse, deletes the information of destination object in object listing, object listing is upgraded, the storage space in object listing can be saved further, reduce the workload of follow-up comparison.
To the renewal of object listing, can real-time update be selected, when namely target signature is registered to warning list data storehouse, just delete the information of this target signature in object listing in time; Also can select non real-time registration, such as, timer mechanism can be used to upgrade according to certain time interval.
The concrete steps using timer mechanism to upgrade object listing are as follows:
Step 1, starts timer.
Step 2, timer reaches the renewal interval of setting.
Step 3, system is removed from object listing exists the target signature that duration reaches setting value.
As a kind of implementation of the present invention, target is treated that the triggering intervals of the timer that list upgrades can be set as 1 minute, the checkout time of target signature sets in systems in which as required, as being set as 30 minutes or other, but not as limit.
Third embodiment of the invention relates to target in a kind of monitor video and automatically finds and register method.Fig. 3 is that in this monitor video, target finds the schematic flow sheet with register method automatically.
3rd embodiment improves on the basis of the first embodiment, and specifically, as shown in Figure 3, main improvements are:
First, before step 103 judges whether target signature has been present in object listing, also comprise:
Step 108: judge whether target signature has been present in warning list.
If not, then step 103 is performed; If so, then process ends.
Step 108 confirms whether registered this target in warning list.If with the alignment similarity of whole registration pictures of warning list all lower than setting threshold value, namely there is not this target signature, then perform step 103, judge whether target signature has been present in object listing.
Secondly, after target signature is registered in warning list by step 107, also comprise:
Step 109: delete the record of target signature in object listing.
After destination object is registered to warning list data storehouse, deletes the information of destination object in object listing, object listing is upgraded, the storage space in object listing can be saved further, reduce the workload of follow-up comparison.
Four embodiment of the invention relates to target in a kind of monitor video and automatically finds and register method.Fig. 4 is that in this monitor video, target finds the schematic flow sheet with register method automatically.
Specifically, as shown in Figure 4, comprise the following steps:
Step 1, system needs to carry out video acquisition, by sensor, vision signal is converted to the digital signal can carrying out analyzing.
Step 2, system carries out target detection to digital video signal, so that the further process of system.
Step 3, system carries out feature extraction to target, and this step comprises necessary step such as the pre-service of feature extraction necessary target image, illumination and geometrical normalization, feature extraction and conversion etc.
Step 1 is the step identical with general real-time Compare System to 3, is thereafter the step with general real-time Compare System difference to some extent, on the basis of real-time Compare System, namely adds key person automatically find and branch's flow process of list registration of reporting to the police.
Real-time comparison alarm flow branch:
Step 4, the feature obtain extraction and warning list carry out comparison one by one, and judge that alignment similarity reaches the threshold value of setting.
Step 5, alignment similarity reaches the threshold value of setting, and system sends real-time comparison and reports to the police.
Key person finds automatically and list of reporting to the police registers branch:
Step 4, will extract the feature that obtains and repeat target and carry out comparison one by one, when exist alignment similarity reach setting repeat confirm threshold value, then think that this target repeats.
It should be noted that, by the destination object extracted with repeat target and carry out similarity comparison, " adopting which kind of characteristic sum matching algorithm " is not specifically limited.
Step 5, confirms that target repeats, then adds up to the counter that repeats of this target, and judge whether accumulated counts reaches the number of times of setting.If the result of carrying out comparison one by one with object listing exists multiple alignment similarity and reaches threshold value, then only the highest to similarity target is carried out counter and is added up.
Step 6, the counter that there is target in object listing adds up and reaches the number of times of setting, then itself and warning list are carried out comparison one by one, is confirmed whether to register this target.
Step 7, if with the alignment similarity of whole registration pictures of warning list all lower than setting check threshold value, then this target is not registered.
Step 8, will through confirming that unregistered target registered is to warning list.
Step 5 to 8, particularly, as shown in Figure 5.
Be registered to warning list to comprise further: if with the alignment similarity of whole registration pictures of warning list all lower than setting check threshold value, then this target is not registered; Will through confirming that unregistered target registered is to warning list.
Above-mentioned " destination object exceeding frequency of occurrence is registered to warning list in real time ", separately optional taking non real-timely is registered to warning list, such as: set a time period T, the destination object occurring exceeding multiplicity in T time is carried out being registered to warning list.
It should be noted that the warning list real-time comparison flow process shown in Fig. 4 and automatic registration alarming name single process can divide open shop at a processor or two processors.
As one preferred embodiment, adopt two flow processs to divide open shop in the present invention, mainly consider that the processing capability in real time of existing single processor system is limited, two flow processs complete the performance that can affect processor simultaneously in a processor.
Describe as can be seen from above, core of the present invention is automatic discovery and the registration automatically of common-denominator target, may be summarized to be and is realized by the method for two cover warning lists and three comparisons, specific as follows:
Two cover warning lists: 1. for the warning list of comparison warning; 2. the object listing (object listing namely occurred in system statistics time section) for upgrading warning list.
Three comparisons: 1. comparison is in real time reported to the police; 2. be confirmed whether to repeat by comparison; 3. be confirmed whether to register by comparison.
Fifth embodiment of the invention relates to target in a kind of monitor video and automatically finds and register method.Fig. 6 is that in this monitor video, target finds the schematic flow sheet with register method automatically.
Specifically, as shown in Figure 6, can find out, this embodiment is compared with the 4th embodiment, and the key distinction is that the registration order of warning list, database is inconsistent.
In like manner, in the present embodiment, destination object is registered to warning list data storehouse can select real-time registration, and namely frequency of occurrence is once reach the upper limit, is just registered to warning list in real time; Also non real-time registration can be selected, the i.e. upper limit of frequency of occurrence in setting-up time cycle T, the number of times that in accumulative this period, all destination objects repeat, when reached during cycle T, warning list data storehouse is registered to all eligible (multiplicity exceedes the destination object of the upper limit).
Use timer mechanism to upgrade according to certain time interval to the renewal repeating target list to be compared, concrete steps are as follows:
Step 1, system starts timer.
Step 2, timer reaches the renewal interval of setting.
Step 3, there is from repeating target list removing to be compared the target that duration reaches setting value in system.
In addition, for the renewal repeating object listing, real-time update can be selected, be different from above-mentioned " upgrading according to certain time interval ", namely, in the respective embodiments described above, just delete this destination object in time when destination object is registered to warning list data storehouse and repeating the information in list to be compared.
For a better understanding of the present invention; hereafter will be applied as a kind of complete procedure realizing detail system and realize of the present invention with the real-time comparison of face; but be not limited to the present invention; adopt personage's target detection of the other biological such as human body, gait feature and automatic registration should belong to protection category of the present invention equally, the same discovery for other targets such as vehicle (utilizing the features such as car plate, vehicle, color), animals (utilizing the feature such as hair color, body) and automatic registration also should belong to protection category of the present invention.
As a kind of implementation of the present invention, the image acquisition of system, target detection two steps adopt face snap machine to realize, face snap machine is the IP video camera with face snap function, specific implementation is real-time detection and tracking face from the video code flow gathered, and a top-quality picture is selected to same person upload, there is the advantage that frontal chromatography ensures real-time and reduces volume of transmitted data, camera exposure parameter can be carried out according to target automatically to regulate simultaneously, thus better target image can be obtained, concrete, Face datection adopts the Boosting method of Haar feature to realize, face tracking adopts the method for light stream to realize, face is selected to comprise and is carried out comprehensive evaluation to the sharpness of facial image and attitude two indices.
As a kind of implementation of the present invention, the characteristic extraction step of face is realized in the following manner, first in facial image, eyes coordinate is located, then eyes coordinate is snapped to coordinate (48, 48) and (112, 48), obtain the facial image through geometrical normalization, the method of entropy diagram picture is adopted to carry out unitary of illumination to it again, then the LBP characteristic sum Gabor characteristic (method of feature selecting uses the method for Boosting to select the feature being conducive to most realizing face classification) through selecting is extracted, finally the method for Metric Learning is adopted to carry out further dimensionality reduction and the measure space conversion of feature further to the feature extracted, make the most applicable similarity degree being used for calculating two facial images of feature.
As a kind of implementation of the present invention, what the Similarity Measure step in three real-time comparisons adopted is mahalanobis distance, but carry out measure space conversion owing to have employed Metric Learning method in characteristic extraction step, so this distance effectively can weigh the similarity degree of two facial images, and the simple superiority of calculating can be obtained.
As a kind of implementation of the present invention, the threshold value that real-time comparison is reported to the police can set in systems in which, is usually set to 0.70.
As a kind of implementation of the present invention, repeat confirm and registered warning list confirm threshold value employing be identical value, this value can also set in systems in which, but usually sets slightly higher than real-time comparison alarm threshold value, as 0.80.
As a kind of implementation of the present invention, the frequency threshold value repeating counter can set in systems in which, and general employing is advisable for 3-10 time.
As a kind of implementation of the present invention, 1 minute is set as to the trigger event interval of repeating the timer that target list to be compared is safeguarded, the checkout time repeating target sets in systems in which as required, as being set as 30 minutes or other.
The present invention has a good application prospect; potential user's (inquiring repeatedly appears in people usually that buy large part) is found as the service location such as 4 S auto shop, house property medium can be applied to; be applied to the sensitive sites such as bank, national treasury or the public place such as school, station discovery suspicious figure (suspect can repeatedly occur usually); the application mode more than enumerated only understands the present invention for helping; be not limited to the present invention, other application according to the invention also should belong to protection category of the present invention.
Each method embodiment of the present invention all can realize in modes such as software, hardware, firmwares.No matter the present invention realizes with software, hardware or firmware mode, instruction code can be stored in the addressable storer of computing machine of any type (such as permanent or revisable, volatibility or non-volatile, solid-state or non-solid, fixing or removable medium etc.).Equally, storer can be such as programmable logic array (Programmable Array Logic, be called for short " PAL "), random access memory (Random Access Memory, be called for short " RAM "), programmable read only memory (Programmable Read Only Memory, be called for short " PROM "), ROM (read-only memory) (Read-Only Memory, be called for short " ROM "), Electrically Erasable Read Only Memory (Electrically Erasable Programmable ROM, be called for short " EEPROM "), disk, CD, digital versatile disc (Digital Versatile Disc, be called for short " DVD ") etc.
Sixth embodiment of the invention relates to target in a kind of monitor video and automatically finds and Accreditation System, and as shown in Figure 7, in this monitor video, target finds to comprise with Accreditation System automatically:
Detecting unit, for detecting target from video flowing.
Target's feature-extraction unit, carries out feature extraction for the target detected detecting unit, obtains target signature.
First judging unit, for judging whether target signature has been present in object listing.
Counting unit, for when the first judging unit judges that target signature is present in object listing, adds 1 by the occurrence number of target signature.When the first judging unit judges that target signature is not present in object listing, target signature is joined in object listing, and the occurrence number of target signature is set to 1.
Second judging unit, for judging whether the occurrence number of the target signature that counting unit exports exceedes the threshold value preset.
Registering unit, for when the second judging unit judges that the occurrence number of target signature exceedes the threshold value preset, is registered to target signature in warning list.
First embodiment is the method embodiment corresponding with present embodiment, and present embodiment can be worked in coordination with the first embodiment and be implemented.The relevant technical details mentioned in first embodiment is still effective in the present embodiment, in order to reduce repetition, repeats no more here.Correspondingly, the relevant technical details mentioned in present embodiment also can be applicable in the first embodiment.
Seventh embodiment of the invention relates to target in a kind of monitor video and automatically finds and Accreditation System.
7th embodiment improves on the basis of the 6th embodiment, and main improvements are also to comprise:
3rd judging unit, before being registered in warning list in registering unit by target signature, judges whether target signature has been present in warning list.
Updating block, after being registered in warning list in registering unit by target signature, deletes the record of target signature in object listing.
Timing unit, for setting the operating interval of registering unit and updating block.
Second and third, four, five embodiments are method embodiments corresponding with present embodiment, present embodiment can with second and third, four, five embodiments work in coordination and implement.Second and third, four, the relevant technical details mentioned in five embodiments is still effective in the present embodiment, in order to reduce repetition, repeat no more here.Correspondingly, the relevant technical details mentioned in present embodiment also can be applicable to second and third, four, in five embodiments.
It should be noted that, the each unit mentioned in the present invention's each equipment embodiment is all logical block, physically, a logical block can be a physical location, also can be a part for a physical location, can also realize with the combination of multiple physical location, the Physical realization of these logical blocks itself is not most important, and the combination of the function that these logical blocks realize is only the key solving technical matters proposed by the invention.In addition, in order to outstanding innovative part of the present invention, the unit not too close with solving technical matters relation proposed by the invention is not introduced by the above-mentioned each equipment embodiment of the present invention, and this does not show that the said equipment embodiment does not exist other unit.
It should be noted that, in the claim and instructions of this patent, the such as relational terms of first and second grades and so on is only used for an entity or operation to separate with another entity or operational zone, and not necessarily requires or imply the relation that there is any this reality between these entities or operation or sequentially.And, term " comprises ", " comprising " or its any other variant are intended to contain comprising of nonexcludability, thus make to comprise the process of a series of key element, method, article or equipment and not only comprise those key elements, but also comprise other key elements clearly do not listed, or also comprise by the intrinsic key element of this process, method, article or equipment.When not more restrictions, the key element " being comprised " limited by statement, and be not precluded within process, method, article or the equipment comprising described key element and also there is other identical element.
Although by referring to some of the preferred embodiment of the invention, to invention has been diagram and describing, but those of ordinary skill in the art should be understood that and can do various change to it in the form and details, and without departing from the spirit and scope of the present invention.
Claims (10)
1. in monitor video, target finds and a register method automatically, it is characterized in that, comprises the following steps:
Target is detected from video flowing;
Feature extraction is carried out to the target detected, obtains target signature;
Judge whether described target signature has been present in object listing, if so, then the occurrence number of described target signature is added 1; If not, then described target signature is joined in described object listing, and the occurrence number of described target signature is set to 1;
Judge whether the occurrence number of described target signature exceedes the threshold value preset, if exceed, then described target signature is registered in warning list.
2. in monitor video according to claim 1, target finds and register method automatically, it is characterized in that, described, before described target signature is registered to the step in warning list, further comprising the steps of:
Judge whether described target signature has been present in warning list, if do not exist, then then described target signature is registered to the step in warning list described in performing.
3. in monitor video according to claim 1, target finds and register method automatically, it is characterized in that, described judge the step whether described target signature has been present in object listing before, further comprising the steps of:
Judge whether described target signature has been present in warning list, if do not exist, then perform and describedly judge whether described target has been present in the step in object listing.
4. in monitor video according to any one of claim 1 to 3, target finds and register method automatically, it is characterized in that, described, after described target signature is registered to the step in warning list, further comprising the steps of:
Delete the record of described target signature in described object listing.
5. in monitor video according to claim 4, target finds and register method automatically, it is characterized in that, describedly judge whether the occurrence number of described target signature exceedes the threshold value preset, if exceed, then described target signature is registered in warning list, delete the step of the record of described target signature in described object listing, perform once every the time interval preset.
6. in monitor video according to claim 5, target finds and register method automatically, it is characterized in that, described target comprises: vehicle and face.
7. in monitor video, target finds and an Accreditation System automatically, it is characterized in that, comprising:
Detecting unit, for detecting target from video flowing;
Target's feature-extraction unit, carries out feature extraction for the target detected described detecting unit, obtains target signature;
First judging unit, for judging whether described target signature has been present in object listing;
Counting unit, for when described first judging unit judges that described target signature is present in object listing, adds 1 by the occurrence number of described target signature; When described first judging unit judges that described target signature is not present in object listing, described target signature is joined in described object listing, and the occurrence number of described target signature is set to 1;
Second judging unit, for judging whether the occurrence number of the described target signature that described counting unit exports exceedes the threshold value preset;
Registering unit, for when described second judging unit judges that the occurrence number of described target signature exceedes the threshold value preset, is registered to described target signature in warning list.
8. in monitor video according to claim 7, target finds and Accreditation System automatically, it is characterized in that, also comprises:
3rd judging unit, before being registered in warning list in described registering unit by described target signature, judges whether described target signature has been present in warning list.
9. in the monitor video according to claim 7 or 8, target finds and Accreditation System automatically, it is characterized in that, also comprises:
Updating block, after being registered in warning list in described registering unit by described target signature, deletes the record of described target signature in described object listing.
10. in monitor video according to claim 9, target finds and Accreditation System automatically, it is characterized in that, also comprises:
Timing unit, for setting the operating interval of described registering unit and updating block.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310231645.1A CN104239881B (en) | 2013-06-08 | 2013-06-08 | Method and system for automatically discovering and registering target in monitoring video |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201310231645.1A CN104239881B (en) | 2013-06-08 | 2013-06-08 | Method and system for automatically discovering and registering target in monitoring video |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104239881A true CN104239881A (en) | 2014-12-24 |
| CN104239881B CN104239881B (en) | 2020-04-24 |
Family
ID=52227906
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201310231645.1A Active CN104239881B (en) | 2013-06-08 | 2013-06-08 | Method and system for automatically discovering and registering target in monitoring video |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104239881B (en) |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106295313A (en) * | 2016-08-03 | 2017-01-04 | 北京地平线机器人技术研发有限公司 | Object identity management method, device and electronic equipment |
| CN106485217A (en) * | 2016-09-30 | 2017-03-08 | 四川世纪云道科技有限公司 | A kind of method and system of identification visit scenic spot stream of people's saturation degree |
| CN106650603A (en) * | 2016-10-18 | 2017-05-10 | 北京小米移动软件有限公司 | Vehicle surrounding monitoring method, apparatus and the vehicle |
| CN106709424A (en) * | 2016-11-19 | 2017-05-24 | 北京中科天云科技有限公司 | Optimized surveillance video storage system and equipment |
| CN106971142A (en) * | 2017-02-07 | 2017-07-21 | 深圳云天励飞技术有限公司 | A kind of image processing method and device |
| CN108540758A (en) * | 2017-03-01 | 2018-09-14 | 中国电信股份有限公司 | The processing method and system of image data |
| CN109214258A (en) * | 2017-07-05 | 2019-01-15 | 杭州海康威视系统技术有限公司 | Lose the detection method and device that the personnel that drive drive in violation of rules and regulations |
| CN109271554A (en) * | 2018-09-07 | 2019-01-25 | 江西省云眼大视界科技有限公司 | A kind of intelligent video identifying system and its application |
| CN109376672A (en) * | 2018-10-31 | 2019-02-22 | 银河水滴科技（北京）有限公司 | Target object localization method and device based on living things feature recognition |
| CN109671241A (en) * | 2017-10-16 | 2019-04-23 | 中国电信股份有限公司 | Alarm method and system |
| CN110516528A (en) * | 2019-07-08 | 2019-11-29 | 杭州电子科技大学 | A kind of moving-target detection and tracking method based under movement background |
| CN110516656A (en) * | 2019-09-04 | 2019-11-29 | 上海眼控科技股份有限公司 | Video monitoring method, device, computer equipment and readable storage medium storing program for executing |
| CN110675639A (en) * | 2019-12-03 | 2020-01-10 | 武汉中科通达高新技术股份有限公司 | Method for analyzing true cards of fake-licensed vehicle based on bayonet vehicle passing data |
| CN110705332A (en) * | 2018-07-09 | 2020-01-17 | 浙江宇视科技有限公司 | Face tracking method and device |
| CN111161206A (en) * | 2018-11-07 | 2020-05-15 | 杭州海康威视数字技术股份有限公司 | Image capturing method, monitoring camera and monitoring system |
| CN111625794A (en) * | 2019-02-28 | 2020-09-04 | 广东美的制冷设备有限公司 | Input method, operation control module, household appliance, system and storage medium |
| CN113420183A (en) * | 2021-06-21 | 2021-09-21 | 成都神州数码索贝科技有限公司 | Cross-video target retrieval and collision method and system in video detection system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060227237A1 (en) * | 2005-03-31 | 2006-10-12 | International Business Machines Corporation | Video surveillance system and method with combined video and audio recognition |
| CN101894428A (en) * | 2010-05-12 | 2010-11-24 | 北京海鑫智圣技术有限公司 | ATM (Automated Teller Machine) intelligent monitoring system |
| CN102158379A (en) * | 2011-05-17 | 2011-08-17 | 福建求实电子有限公司 | Identification network managing system and method for residents of indemnificatory residences |
| US20110280442A1 (en) * | 2010-05-13 | 2011-11-17 | Hon Hai Precision Industry Co., Ltd. | Object monitoring system and method |
| CN102426786A (en) * | 2011-11-15 | 2012-04-25 | 无锡港湾网络科技有限公司 | Intelligent video analyzing system and method for automatically identifying fake plate vehicle |
| CN102610102A (en) * | 2012-02-27 | 2012-07-25 | 安科智慧城市技术（中国）有限公司 | Suspect vehicle inspection and control method and system |
-
2013
- 2013-06-08 CN CN201310231645.1A patent/CN104239881B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060227237A1 (en) * | 2005-03-31 | 2006-10-12 | International Business Machines Corporation | Video surveillance system and method with combined video and audio recognition |
| CN101894428A (en) * | 2010-05-12 | 2010-11-24 | 北京海鑫智圣技术有限公司 | ATM (Automated Teller Machine) intelligent monitoring system |
| US20110280442A1 (en) * | 2010-05-13 | 2011-11-17 | Hon Hai Precision Industry Co., Ltd. | Object monitoring system and method |
| CN102158379A (en) * | 2011-05-17 | 2011-08-17 | 福建求实电子有限公司 | Identification network managing system and method for residents of indemnificatory residences |
| CN102426786A (en) * | 2011-11-15 | 2012-04-25 | 无锡港湾网络科技有限公司 | Intelligent video analyzing system and method for automatically identifying fake plate vehicle |
| CN102610102A (en) * | 2012-02-27 | 2012-07-25 | 安科智慧城市技术（中国）有限公司 | Suspect vehicle inspection and control method and system |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106295313A (en) * | 2016-08-03 | 2017-01-04 | 北京地平线机器人技术研发有限公司 | Object identity management method, device and electronic equipment |
| CN106485217A (en) * | 2016-09-30 | 2017-03-08 | 四川世纪云道科技有限公司 | A kind of method and system of identification visit scenic spot stream of people's saturation degree |
| CN106650603A (en) * | 2016-10-18 | 2017-05-10 | 北京小米移动软件有限公司 | Vehicle surrounding monitoring method, apparatus and the vehicle |
| CN106709424A (en) * | 2016-11-19 | 2017-05-24 | 北京中科天云科技有限公司 | Optimized surveillance video storage system and equipment |
| CN106971142A (en) * | 2017-02-07 | 2017-07-21 | 深圳云天励飞技术有限公司 | A kind of image processing method and device |
| CN108540758A (en) * | 2017-03-01 | 2018-09-14 | 中国电信股份有限公司 | The processing method and system of image data |
| CN109214258A (en) * | 2017-07-05 | 2019-01-15 | 杭州海康威视系统技术有限公司 | Lose the detection method and device that the personnel that drive drive in violation of rules and regulations |
| CN109671241A (en) * | 2017-10-16 | 2019-04-23 | 中国电信股份有限公司 | Alarm method and system |
| CN110705332B (en) * | 2018-07-09 | 2022-10-04 | 浙江宇视科技有限公司 | Face tracking method and device |
| CN110705332A (en) * | 2018-07-09 | 2020-01-17 | 浙江宇视科技有限公司 | Face tracking method and device |
| CN109271554A (en) * | 2018-09-07 | 2019-01-25 | 江西省云眼大视界科技有限公司 | A kind of intelligent video identifying system and its application |
| CN109376672A (en) * | 2018-10-31 | 2019-02-22 | 银河水滴科技（北京）有限公司 | Target object localization method and device based on living things feature recognition |
| CN111161206A (en) * | 2018-11-07 | 2020-05-15 | 杭州海康威视数字技术股份有限公司 | Image capturing method, monitoring camera and monitoring system |
| CN111625794A (en) * | 2019-02-28 | 2020-09-04 | 广东美的制冷设备有限公司 | Input method, operation control module, household appliance, system and storage medium |
| CN111625794B (en) * | 2019-02-28 | 2024-03-05 | 广东美的制冷设备有限公司 | Recording method, operation control module, household appliance, system and storage medium |
| CN110516528A (en) * | 2019-07-08 | 2019-11-29 | 杭州电子科技大学 | A kind of moving-target detection and tracking method based under movement background |
| CN110516656A (en) * | 2019-09-04 | 2019-11-29 | 上海眼控科技股份有限公司 | Video monitoring method, device, computer equipment and readable storage medium storing program for executing |
| CN110675639B (en) * | 2019-12-03 | 2020-05-19 | 武汉中科通达高新技术股份有限公司 | Method for analyzing true cards of fake-licensed vehicle based on bayonet vehicle passing data |
| CN110675639A (en) * | 2019-12-03 | 2020-01-10 | 武汉中科通达高新技术股份有限公司 | Method for analyzing true cards of fake-licensed vehicle based on bayonet vehicle passing data |
| CN113420183A (en) * | 2021-06-21 | 2021-09-21 | 成都神州数码索贝科技有限公司 | Cross-video target retrieval and collision method and system in video detection system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104239881B (en) | 2020-04-24 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104239881A (en) | Method and system for automatically finding and registering target in surveillance video | |
| CN109166261B (en) | Image processing method, device and equipment based on image recognition and storage medium | |
| CN110428091B (en) | Risk identification method based on data analysis and related equipment | |
| US20220092881A1 (en) | Method and apparatus for behavior analysis, electronic apparatus, storage medium, and computer program | |
| WO2018223955A1 (en) | Target monitoring method, target monitoring device, camera and computer readable medium | |
| US20170300786A1 (en) | Methods and systems for accurately recognizing vehicle license plates | |
| CN204155449U (en) | A kind of Car license recognition comparison high-definition camera and supervisory system | |
| CN110263633B (en) | Method, system and storage medium for detecting and early warning of toxic person based on space-time correlation | |
| US9704201B2 (en) | Method and system for detecting uninsured motor vehicles | |
| KR20200071799A (en) | object recognition and counting method using deep learning artificial intelligence technology | |
| CN102164270A (en) | Intelligent video monitoring method and system capable of exploring abnormal events | |
| CN111291682A (en) | Method and device for determining target object, storage medium and electronic device | |
| CN102902960A (en) | Leave-behind object detection method based on Gaussian modelling and target contour | |
| WO2014193220A2 (en) | System and method for multiple license plates identification | |
| CN110717358B (en) | Visitor number counting method and device, electronic equipment and storage medium | |
| CN109712291B (en) | Opening method and device of electronic gate and server | |
| CN111090777A (en) | Video data management method, management equipment and computer storage medium | |
| WO2023071188A1 (en) | Abnormal-behavior detection method and apparatus, and electronic device and storage medium | |
| CN109783663B (en) | Archiving method and device | |
| CN113920585A (en) | Behavior recognition method and device, equipment and storage medium | |
| Feris et al. | Case study: IBM smart surveillance system | |
| Kwon et al. | Scene modeling-based anomaly detection for intelligent transport system | |
| CN111368617B (en) | Vehicle access data processing method and device | |
| CN114283361A (en) | Method and apparatus for determining status information, storage medium, and electronic apparatus | |
| CN108831158A (en) | It disobeys and stops monitoring method, device and electric terminal |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |