CN111639578A - Method, device, equipment and storage medium for intelligently identifying illegal parabola - Google Patents
Method, device, equipment and storage medium for intelligently identifying illegal parabola Download PDFInfo
- Publication number
- CN111639578A CN111639578A CN202010451236.2A CN202010451236A CN111639578A CN 111639578 A CN111639578 A CN 111639578A CN 202010451236 A CN202010451236 A CN 202010451236A CN 111639578 A CN111639578 A CN 111639578A
- Authority
- CN
- China
- Prior art keywords
- target
- picture
- preset
- group
- video information
- 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
Images
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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- 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
- G06V20/49—Segmenting video sequences, i.e. computational techniques such as parsing or cutting the sequence, low-level clustering or determining units such as shots or scenes
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Software Systems (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Computation (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- General Engineering & Computer Science (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Image Analysis (AREA)
Abstract
The invention relates to a method, a device, equipment and a storage medium for intelligently identifying an illegal parabola, wherein the method comprises the following steps: acquiring video information in a preset time period of a monitoring site; determining a first target graph group according to a first preset picture selection rule and video information, determining a second target graph group according to a second preset picture selection rule and the video information, adjusting the sizes of all pictures in the first target graph group and the second target graph group to be 256 × 256, and converting the formats of all pictures into a single-channel gray-scale graph; determining a track picture according to the first target picture group, the second target picture group and a preset algorithm; and uploading the track picture to a pre-trained parabolic track recognition model to obtain a violation parabolic judgment result. The method and the device for identifying the illegal parabola is simple and efficient in process, less in computing resource requirement and lower in identification cost.
Description
Technical Field
The invention relates to the technical field of intelligent identification, in particular to a method, a device, equipment and a storage medium for intelligently identifying illegal parabolas.
Background
When the staff in the center is transported in the express delivery carries out the express delivery letter sorting, always have some staff to carry out violence letter sorting for some reason, like throw away, trample or the foot kicks the express delivery piece. These violent sorting activities may cause different degrees of damage to the courier, affecting the reputation of the courier company. Therefore, when being necessary to monitor staff in the express delivery transportation center to sort the express delivery, whether violent sorting behaviors exist or not, so that the violent sorting behaviors are processed in time, and the occurrence of the violent sorting behaviors is reduced or even avoided.
In the related technology, deep learning algorithm recognition is carried out on each frame of picture in the monitoring video of the express delivery transit center, so that all illegal parabolic behaviors in the monitoring video can be recognized. However, an express company may have hundreds of express transit centers, and each express transit center has thousands of cameras, so that if deep learning algorithm recognition is performed on each frame of picture in a surveillance video of an express transit center, the recognition process is complicated, a large amount of GPU resources are consumed, and the cost is high.
Disclosure of Invention
In view of this, a method, an apparatus, a device and a storage medium for intelligently identifying an illegal parabola are provided to solve the problems of complicated identification process and high identification cost in the related art.
The invention adopts the following technical scheme:
in a first aspect, the present invention provides a method for intelligently identifying an offending parabola, the method comprising:
acquiring video information in a preset time period of a monitoring site;
determining a first target graph group according to a first preset graph selection rule and the video information, adjusting the sizes of all the graphs in the first target graph group to be 256 × 256, and converting the formats of the graphs into a single-channel gray graph; the first preset picture selection rule is that a plurality of first target pictures are uniformly selected from all pictures corresponding to the video information at intervals of a first preset frame number; all the first target pictures form the first target picture group;
determining a second target graph group according to a second preset graph selection rule and the video information, adjusting the sizes of all the graphs in the second target graph group to be 256 × 256, and converting the formats of the graphs into a single-channel gray graph; the second preset picture selection rule is that a plurality of second target pictures are uniformly selected from all pictures corresponding to the video information at intervals of a second preset frame number; all the second target pictures form the second target picture group;
determining a track picture according to the first target graph group, the second target graph group and a preset algorithm;
uploading the trajectory picture to a pre-trained parabolic trajectory recognition model to obtain a violation parabolic determination result; the parabolic track recognition model is used for judging whether parabolic behaviors exist in the track picture according to the track of the article in the track picture.
Further, the determining a first target group according to a first preset picture selection rule and the video information specifically includes:
determining all of said first target pictures P (t) in said video informationn) Obtaining the first target graph group; wherein, P (t)n) Indicating the t-th in the video informationnFrame picture, tnThe calculation formula of (a) is as follows:
tn＝2*n-1
wherein n is a positive integer;
the method for determining the second target image group according to the second preset image selection rule and the video information specifically comprises the following steps:
determining all of the second target pictures P (T) in the video informationN) Obtaining the second target graph group; wherein, P (T)N) Indicating the Tth in the video informationNFrame picture, TNThe calculation formula of (a) is as follows:
TN＝4*N-3
wherein N is a positive integer.
Further, the determining a track picture according to the first target graph group, the second target graph group and a preset algorithm specifically includes:
when t isn＝TNWhen not equal to 1, comparing and determining P (t)n) And P (t)n-1) A first target pixel point with a changed middle pixel value is judged, whether the first pixel value change range of the first target pixel point is within a preset pixel value change range or not is judged, if yes, the corresponding first target pixel point is marked as 1, otherwise, the corresponding first target pixel point is marked as 0, and a first matrix a is obtainedn(ii) a Wherein n is a positive integer;
comparing and determining P (T)N) And P (T)N-1) A second target pixel point with a changed middle pixel value is judged, whether the second pixel value change range of the second target pixel point is within the preset pixel value change range or not is judged, if yes, the corresponding second target pixel point is marked as 1, otherwise, the corresponding second target pixel point is marked as 0, and a second matrix b is obtainedN(ii) a Wherein N is a positive integer;
with said anSubtracting the corresponding bNTo obtain the target matrix Cnn；
Accumulating the object matrix CnnAnd obtaining the track picture.
Further, the parabolic track recognition model is a YOLOv3 model.
In a second aspect, the present invention provides an apparatus for intelligently identifying an offending parabola, the apparatus comprising:
the video acquisition module is used for acquiring video information in a preset time period of a monitoring site;
a first target image group determining module, configured to determine a first target image group according to a first preset image selection rule and the video information, adjust sizes of all images in the first target image group to 256 × 256, and convert formats of all images into a single-channel grayscale image; the first preset picture selection rule is that a plurality of first target pictures are uniformly selected from all pictures corresponding to the video information at intervals of a first preset frame number; all the first target pictures form the first target picture group;
a second target image group determining module, configured to determine a second target image group according to a second preset image selection rule and the video information, adjust sizes of all images in the second target image group to 256 × 256, and convert formats of all images into a single-channel grayscale image; the second preset picture selection rule is that a plurality of second target pictures are uniformly selected from all pictures corresponding to the video information at intervals of a second preset frame number; all the second target pictures form the second target picture group;
the track picture determining module is used for determining a track picture according to the first target picture group, the second target picture group and a preset algorithm;
the illegal parabolic determining module is used for uploading the track picture to a pre-trained parabolic track recognition model to obtain an illegal parabolic determination result; the parabolic track recognition model is used for judging whether parabolic behaviors exist in the track picture according to the track of the article in the track picture.
Further, the module for determining the first target group is specifically configured to:
determining all of said first target pictures P (t) in said video informationn) Obtaining the first target graph group; wherein, P (t)n) Indicating the t-th in the video informationnFrame picture, tnThe calculation formula of (a) is as follows:
tn＝2*n-1
wherein n is a positive integer;
the determine second target group module is specifically configured to:
determining all of the second target pictures P (T) in the video informationN) Obtaining the second target graph group; wherein, P (T)N) Indicating the Tth in the video informationNFrame picture, TNThe calculation formula of (a) is as follows:
TN＝4*N-3
wherein N is a positive integer.
Further, the track picture determining module is specifically configured to:
when t isn＝TNWhen not equal to 1, comparing and determining P (t)n) And P (t)n-1) A first target pixel point with a changed middle pixel value is judged, whether the first pixel value change range of the first target pixel point is within a preset pixel value change range or not is judged, if yes, the corresponding first target pixel point is marked as 1, otherwise, the corresponding first target pixel point is marked as 0, and a first matrix a is obtainedn(ii) a Wherein n is a positive integer;
comparing and determining P (T)N) And P (T)N-1) A second target pixel point with a changed middle pixel value, and judging whether the second pixel value change range of the second target pixel point is in the rangeWithin the preset pixel value variation range, if yes, marking the corresponding second target pixel point as 1, otherwise, marking the corresponding second target pixel point as 0 to obtain a second matrix bN(ii) a Wherein N is a positive integer;
with said anSubtracting the corresponding bNTo obtain the target matrix Cnn；
Accumulating the object matrix CnnAnd obtaining the track picture.
Further, the violation parabolic module is specifically configured to:
and uploading the track picture to a pre-trained YOLOv3 model to obtain a violation parabolic determination result.
In a third aspect, the present invention provides an apparatus comprising: a processor, and a memory coupled to the processor; the memory is used for storing a computer program for at least performing the above-mentioned method for intelligently identifying an offending parabola; the processor is used for calling and executing the computer program in the memory.
In a fourth aspect, the present invention provides a storage medium storing a computer program, which when executed by a processor, implements the steps of the above method for intelligently identifying an offending parabola.
By adopting the technical scheme, the method comprises the steps of firstly obtaining video information in a preset time period of a monitoring site; then, determining a first target graph group according to a first preset graph selection rule and the video information, adjusting the sizes of all the graphs in the first target graph group to 256 × 256, and converting the formats of the graphs into a single-channel gray graph; determining a second target graph group according to a second preset graph selection rule and the video information, adjusting the sizes of all the graphs in the second target graph group to be 256 × 256, and converting the formats of the graphs into a single-channel gray graph; determining a track picture according to the first target graph group, the second target graph group and a preset algorithm; and finally, uploading the track picture to a pre-trained parabolic track recognition model to obtain a violation parabolic judgment result. Because the video information in the preset time period of the monitoring field is converted into the picture containing the motion tracks of all the objects in the preset time period in the monitoring field, algorithm recognition is only needed to be carried out on one picture, all the illegal parabolic behaviors in the preset time period can be judged, the recognition process is simple and efficient, the operation resources are saved, and the recognition cost is reduced.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
Fig. 1 is a schematic flowchart of a method for intelligently identifying an offending parabola according to an embodiment of the present invention.
Fig. 2 is a schematic structural diagram of an apparatus for intelligently identifying an illegal parabola according to an embodiment of the present invention.
Fig. 3 is a schematic structural diagram of an apparatus according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the technical solutions of the present invention will be described in detail below. It is to be understood that the described embodiments are merely exemplary of the invention, and not restrictive of the full scope of the invention. All other embodiments, which can be derived by a person skilled in the art from the examples given herein without any inventive step, are within the scope of the present invention.
Fig. 1 is a schematic flowchart of a method for intelligently identifying an offending parabola according to an embodiment of the present invention. As shown in fig. 1, the method includes:
s101, video information in a preset time period of a monitoring field is obtained.
Specifically, a Capture Video function in an Open Source Computer Vision library (OpenCV) may be used to obtain a Video in a preset Video streaming protocol (RTSP), so as to obtain Video information in a preset Time period of a monitoring site. The duration of the preset time period may be determined according to actual requirements, for example, a video stream with a duration of 10S may be selected from the target video library as the target video stream, that is, the video information described in this application.
S102, determining a first target graph group according to a first preset graph selection rule and the video information, adjusting the size of all the graphs in the first target graph group to 256 × 256, and converting the format of the graphs into a single-channel gray graph; the first preset picture selection rule is that a plurality of first target pictures are uniformly selected from all pictures corresponding to the video information at intervals of a first preset frame number; and all the first target pictures form the first target picture group.
In detail, since a plurality of first target pictures are uniformly selected from all pictures corresponding to the video information at intervals of a first preset frame number, a first object motion track determined according to all the first target pictures is approximately the same as an object motion track in the video information, and therefore, the first object motion track can be used for representing the object motion track in the video information; and a plurality of first target pictures are uniformly selected from all pictures corresponding to the video information at intervals of a first preset frame number, so that the number of pictures to be analyzed by a computer system is reduced, namely, the workload of the computer system is reduced.
S103, determining a second target graph group according to a second preset graph selection rule and the video information, adjusting the size of all the graphs in the second target graph group to 256 × 256, and converting the format of the graphs into a single-channel gray graph; the second preset picture selection rule is that a plurality of second target pictures are uniformly selected from all pictures corresponding to the video information at intervals of a second preset frame number; and all the second target pictures form the second target picture group.
Specifically, the second preset frame number is greater than the first preset frame number, and the determination method of the second target graph group is similar to the determination method of the first target graph group, but the number of interval frames is different, so that the motion trajectory of the object in the video information can also be determined according to the second target graph group, the motion trajectory is called a second object motion trajectory, and the second object motion trajectory is also approximately the same as the object motion trajectory in the video information, and can also be used for representing the object motion trajectory in the video information.
And S104, determining a track picture according to the first target graph group, the second target graph group and a preset algorithm.
In detail, the first object motion track and the second object motion track both include an object motion track with a slow motion speed, and the object motion speed corresponding to the illegal parabolic motion is fast, so that in order to judge the illegal parabolic motion more accurately and efficiently, the object motion track only including the fast motion speed can be determined. And determining that the track picture does not contain the motion track of the object with the slower motion speed according to the first target image group, the second target image group and a preset algorithm.
S105, uploading the trajectory picture to a pre-trained parabolic trajectory recognition model to obtain a violation parabolic determination result; the parabolic track recognition model is used for judging whether parabolic behaviors exist in the track picture according to the track of the article in the track picture.
By adopting the technical scheme, the method comprises the steps of firstly obtaining video information in a preset time period of a monitoring site; then, determining a first target graph group according to a first preset graph selection rule and the video information, adjusting the sizes of all the graphs in the first target graph group to 256 × 256, and converting the formats of the graphs into a single-channel gray graph; determining a second target graph group according to a second preset graph selection rule and the video information, adjusting the sizes of all the graphs in the second target graph group to be 256 × 256, and converting the formats of the graphs into a single-channel gray graph; determining a track picture according to the first target graph group, the second target graph group and a preset algorithm; and finally, uploading the track picture to a pre-trained parabolic track recognition model to obtain a violation parabolic judgment result. Because the video information in the preset time period of the monitoring field is converted into the picture containing the motion tracks of all the objects in the preset time period in the monitoring field, algorithm recognition is only needed to be carried out on one picture, all the illegal parabolic behaviors in the preset time period can be judged, the recognition process is simple and efficient, the operation resources are saved, and the recognition cost is reduced.
Further, the determining a first target group according to a first preset picture selection rule and the video information specifically includes:
determining all of said first target pictures P (t) in said video informationn) Obtaining the first target graph group; wherein, P (t)n) Indicating the t-th in the video informationnFrame picture, tnThe calculation formula of (a) is as follows:
tn＝2*n-1
wherein n is a positive integer;
the method for determining the second target image group according to the second preset image selection rule and the video information specifically comprises the following steps:
determining all of the second target pictures P (T) in the video informationN) Obtaining the second target graph group; wherein, P (T)N) Indicating the Tth in the video informationNFrame picture, TNThe calculation formula of (a) is as follows:
TN＝4*N-3
wherein N is a positive integer.
Further, determining a track picture according to the first target graph group, the second target graph group and a preset algorithm, and specifically comprising:
when t isn＝TNWhen not equal to 1, comparing and determining P (t)n) And P (t)n-1) A first target pixel point with a changed middle pixel value is judged, whether the first pixel value change range of the first target pixel point is within a preset pixel value change range or not is judged, if yes, the corresponding first target pixel point is marked as 1, otherwise, the corresponding first target pixel point is marked as 0, and a first matrix a is obtainedn(ii) a Wherein n is a positive integer;
comparing and determining P (T)N) And P (T)N-1) Middle imageA second target pixel point with a changed pixel value is judged, whether the second pixel value change range of the second target pixel point is within the preset pixel value change range or not is judged, if yes, the corresponding second target pixel point is marked as 1, otherwise, the corresponding second target pixel point is marked as 0, and a second matrix b is obtainedN(ii) a Wherein N is a positive integer;
with said anSubtracting the corresponding bNTo obtain the target matrix Cnn(ii) a nn is a positive integer;
accumulating the object matrix CnnAnd obtaining the track picture.
Specifically, the first preset frame number is 1 frame, and the second preset frame number is 3 frames, which are easily obtained according to the above contents. In a specific example, the video information includes 9 frames of pictures, the pictures selected according to the first predetermined picture selection rule include P (1), P (3), P (5), P (7), and P (9), and the pictures selected according to the second predetermined picture selection rule include P (1), P (5), and P (9), so that the first target group includes P (1), P (3), P (5), P (7), and P (9), and the second target group includes P (1), P (5), and P (9).
tn＝TNThe case of not equal to 1 includes: t is tn＝TN5 and tn＝TN9. When t isn＝TNWhen the pixel value of the first target pixel point is 5, N is 3, and N is 2, comparing and determining the first target pixel point with the pixel value of the P (5) and the P (3) and judging whether the first pixel value variation range of the first target pixel point is in the preset pixel value variation range, if so, marking the corresponding first target pixel point as 1, otherwise, marking the corresponding first target pixel point as 0, and obtaining a first matrix a3(ii) a Comparing and determining a second target pixel point with a changed pixel value in P (5) and P (1), and judging whether the second pixel value change range of the second target pixel point is within the preset pixel value change range, if so, marking the corresponding second target pixel point as 1, otherwise, marking the corresponding second target pixel point as 0, and obtaining a second matrix b3With said a3Subtracting the corresponding b3To obtain the target matrix C1。
In the same way, when tn＝TNWhen the pixel value is equal to 9, N is equal to 5, N is equal to 3, and the first target pixel point with the pixel value changing in P (9) and P (7) is compared and determined, so that the first matrix a is obtained5Comparing and determining a second target pixel point with a changed pixel value in P (9) and P (5) to obtain a second matrix b3By a5Minus the corresponding b3To obtain the target matrix C2. Cumulative object matrix C1And C2And obtaining the track picture.
Further, the parabolic track recognition model is a YOLOv3 model.
Fig. 2 is a schematic structural diagram of an apparatus for intelligently identifying an illegal parabola according to an embodiment of the present invention. As shown in fig. 2, the apparatus includes: the video acquisition module 21, the first target map group determining module 22, the second target map group determining module 23, the track picture determining module 24 and the violation parabolic module 25.
The video acquisition module 21 is configured to acquire video information in a preset time period of a monitoring site; a first target group determining module 22, configured to determine a first target group according to a first preset picture selection rule and the video information, adjust sizes of all pictures in the first target group to 256 × 256, and convert formats of all pictures into a single-channel grayscale; the first preset picture selection rule is that a plurality of first target pictures are uniformly selected from all pictures corresponding to the video information at intervals of a first preset frame number; all the first target pictures form the first target picture group; a second target group determining module 23, configured to determine a second target group according to a second preset picture selection rule and the video information, adjust the sizes of all pictures in the second target group to 256 × 256, and convert the formats of all pictures into a single-channel grayscale image; the second preset picture selection rule is that a plurality of second target pictures are uniformly selected from all pictures corresponding to the video information at intervals of a second preset frame number; all the second target pictures form the second target picture group; a track picture determining module 24, configured to determine a track picture according to the first target group, the second target group, and a preset algorithm; the illegal parabolic determining module 25 is used for uploading the track picture to a pre-trained parabolic track recognition model to obtain an illegal parabolic determination result; the parabolic track recognition model is used for judging whether parabolic behaviors exist in the track picture according to the track of the article in the track picture.
Further, the determine first target group module 22 is specifically configured to:
determining all of said first target pictures P (t) in said video informationn) Obtaining the first target graph group; wherein, P (t)n) Indicating the t-th in the video informationnFrame picture, tnThe calculation formula of (a) is as follows:
tn＝2*n-1
wherein n is a positive integer;
the determine second target group module 23 is specifically configured to:
determining all of the second target pictures P (T) in the video informationN) Obtaining the second target graph group; wherein, P (T)N) Indicating the Tth in the video informationNFrame picture, TNThe calculation formula of (a) is as follows:
TN＝4*N-3
wherein N is a positive integer.
Further, the track picture determining module 24 is specifically configured to:
when t isn＝TNWhen not equal to 1, comparing and determining P (t)n) And P (t)n-1) A first target pixel point with a changed middle pixel value is judged, whether the first pixel value change range of the first target pixel point is within a preset pixel value change range or not is judged, if yes, the corresponding first target pixel point is marked as 1, otherwise, the corresponding first target pixel point is marked as 0, and a first matrix a is obtainedn(ii) a Wherein n is a positive integer;
comparing and determining P (T)N) And P (T)N-1) A second target pixel point with a changed middle pixel value, and judging whether the second pixel value change range of the second target pixel point is within the preset pixel value change range, if so, corresponding second target imageMarking the pixel point as 1, otherwise, marking the corresponding second target pixel point as 0 to obtain a second matrix bN(ii) a Wherein N is a positive integer;
with said anSubtracting the corresponding bNTo obtain the target matrix Cnn；
Accumulating the object matrix CnnAnd obtaining the track picture.
Further, the module for determining an offending parabola 25 is specifically configured to:
and uploading the track picture to a pre-trained YOLOv3 model to obtain a violation parabolic determination result.
The device for intelligently identifying the illegal object is used for executing the method for intelligently identifying the illegal object, has corresponding executing process and intentional effect, and is not repeated herein.
Fig. 3 is a schematic structural diagram of an apparatus according to an embodiment of the present invention. As shown in fig. 3, the apparatus includes: a processor 310, and a memory 320 coupled to the processor 310.
The memory 320 is configured to store a computer program for performing at least the method for intelligently identifying an offending parabola as described herein, the method comprising:
acquiring video information in a preset time period of a monitoring site;
determining a first target graph group according to a first preset graph selection rule and the video information, adjusting the sizes of all the graphs in the first target graph group to be 256 × 256, and converting the formats of the graphs into a single-channel gray graph; the first preset picture selection rule is that a plurality of first target pictures are uniformly selected from all pictures corresponding to the video information at intervals of a first preset frame number; all the first target pictures form the first target picture group;
determining a second target graph group according to a second preset graph selection rule and the video information, adjusting the sizes of all the graphs in the second target graph group to be 256 × 256, and converting the formats of the graphs into a single-channel gray graph; the second preset picture selection rule is that a plurality of second target pictures are uniformly selected from all pictures corresponding to the video information at intervals of a second preset frame number; all the second target pictures form the second target picture group;
determining a track picture according to the first target graph group, the second target graph group and a preset algorithm;
uploading the trajectory picture to a pre-trained parabolic trajectory recognition model to obtain a violation parabolic determination result; the parabolic track recognition model is used for judging whether parabolic behaviors exist in the track picture according to the track of the article in the track picture.
The processor 310 is used to invoke and execute the computer program in the memory 320.
The application also provides a storage medium, which stores a computer program, and when the computer program is executed by a processor, the steps in the method for intelligently identifying the violation parabola are realized.
It is understood that the same or similar parts in the above embodiments may be mutually referred to, and the same or similar parts in other embodiments may be referred to for the content which is not described in detail in some embodiments.
It should be noted that the terms "first," "second," and the like in the description of the present invention are used for descriptive purposes only and are not to be construed as indicating or implying relative importance. Further, in the description of the present invention, the meaning of "a plurality" means at least two unless otherwise specified.
Any process or method descriptions in flow diagrams or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps of the process, and the scope of the preferred embodiments of the present invention includes additional implementations in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the embodiments of the present invention.
It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and when the program is executed, the program includes one or a combination of the steps of the method embodiments.
In addition, functional units in the embodiments of the present invention may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium.
The storage medium mentioned above may be a read-only memory, a magnetic or optical disk, etc.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made to the above embodiments by those of ordinary skill in the art within the scope of the present invention.
Claims (10)
1. A method for intelligently identifying an offending parabola, comprising:
acquiring video information in a preset time period of a monitoring site;
determining a first target graph group according to a first preset graph selection rule and the video information, adjusting the sizes of all the graphs in the first target graph group to be 256 × 256, and converting the formats of the graphs into a single-channel gray graph; the first preset picture selection rule is that a plurality of first target pictures are uniformly selected from all pictures corresponding to the video information at intervals of a first preset frame number; all the first target pictures form the first target picture group;
determining a second target graph group according to a second preset graph selection rule and the video information, adjusting the sizes of all the graphs in the second target graph group to be 256 × 256, and converting the formats of the graphs into a single-channel gray graph; the second preset picture selection rule is that a plurality of second target pictures are uniformly selected from all pictures corresponding to the video information at intervals of a second preset frame number; all the second target pictures form the second target picture group;
determining a track picture according to the first target graph group, the second target graph group and a preset algorithm;
uploading the trajectory picture to a pre-trained parabolic trajectory recognition model to obtain a violation parabolic determination result; the parabolic track recognition model is used for judging whether parabolic behaviors exist in the track picture according to the track of the article in the track picture.
2. The method for intelligently identifying an illegal parabola according to claim 1, wherein the determining a first target graph group according to a first preset graph selection rule and the video information specifically comprises:
determining all of said first target pictures P (t) in said video informationn) Obtaining the first target graph group; wherein, P (t)n) Indicating the t-th in the video informationnFrame picture, tnThe calculation formula of (a) is as follows:
tn＝2*n-1
wherein n is a positive integer;
the method for determining the second target image group according to the second preset image selection rule and the video information specifically comprises the following steps:
determining all of the second target pictures P (T) in the video informationN) Obtaining the second target graph group; wherein, P (T)N) Indicating the Tth in the video informationNFrame picture, TNThe calculation formula of (a) is as follows:
TN＝4*N-3
wherein N is a positive integer.
3. The method for intelligently identifying an illegal parabola according to claim 2, wherein the determining a trajectory picture according to the first target map group, the second target map group and a preset algorithm specifically comprises:
when t isn＝TNWhen not equal to 1, comparing and determining P (t)n) And P (t)n-1) A first target pixel point with a changed middle pixel value is judged, whether the first pixel value change range of the first target pixel point is within a preset pixel value change range or not is judged, if yes, the corresponding first target pixel point is marked as 1, otherwise, the corresponding first target pixel point is marked as 0, and a first matrix a is obtainedn(ii) a Wherein n is a positive integer;
comparing and determining P (T)N) And P (T)N-1) A second target pixel point with a changed middle pixel value, and judging whether the second pixel value change range of the second target pixel point is within the preset pixel value change range, if so, the corresponding second target pixel point is usedMarking the mark pixel point as 1, otherwise, marking the corresponding second target pixel point as 0 to obtain a second matrix bN(ii) a Wherein N is a positive integer;
with said anSubtracting the corresponding bNTo obtain the target matrix Cnn(ii) a nn is a positive integer;
accumulating the object matrix CnnAnd obtaining the track picture.
4. The method for intelligently identifying an offending parabola as recited in claim 1, wherein the parabola trajectory identification model is a YOLOv3 model.
5. An apparatus for intelligently identifying an offending parabola, comprising:
the video acquisition module is used for acquiring video information in a preset time period of a monitoring site;
a first target image group determining module, configured to determine a first target image group according to a first preset image selection rule and the video information, adjust sizes of all images in the first target image group to 256 × 256, and convert formats of all images into a single-channel grayscale image; the first preset picture selection rule is that a plurality of first target pictures are uniformly selected from all pictures corresponding to the video information at intervals of a first preset frame number; all the first target pictures form the first target picture group;
a second target image group determining module, configured to determine a second target image group according to a second preset image selection rule and the video information, adjust sizes of all images in the second target image group to 256 × 256, and convert formats of all images into a single-channel grayscale image; the second preset picture selection rule is that a plurality of second target pictures are uniformly selected from all pictures corresponding to the video information at intervals of a second preset frame number; all the second target pictures form the second target picture group;
the track picture determining module is used for determining a track picture according to the first target picture group, the second target picture group and a preset algorithm;
the illegal parabolic determining module is used for uploading the track picture to a pre-trained parabolic track recognition model to obtain an illegal parabolic determination result; the parabolic track recognition model is used for judging whether parabolic behaviors exist in the track picture according to the track of the article in the track picture.
6. The apparatus for intelligently identifying an offending parabola as recited in claim 5, wherein the determine first target group module is specifically configured to:
determining all of said first target pictures P (t) in said video informationn) Obtaining the first target graph group; wherein, P (t)n) Indicating the t-th in the video informationnFrame picture, tnThe calculation formula of (a) is as follows:
tn＝2*n-1
wherein n is a positive integer;
the determine second target group module is specifically configured to:
determining all of the second target pictures P (T) in the video informationN) Obtaining the second target graph group; wherein, P (T)N) Indicating the Tth in the video informationNFrame picture, TNThe calculation formula of (a) is as follows:
TN＝4*N-3
wherein N is a positive integer.
7. The apparatus for intelligently identifying an offending parabola as recited in claim 6, wherein the determine trajectory picture module is specifically configured to:
when t isn＝TNWhen not equal to 1, comparing and determining P (t)n) And P (t)n-1) A first target pixel point with a changed middle pixel value is judged, whether the first pixel value change range of the first target pixel point is within a preset pixel value change range or not is judged, if yes, the corresponding first target pixel point is marked as 1, otherwise, the corresponding first target pixel point is marked as 0, and a first matrix a is obtainedn(ii) a Wherein n is a positive integer;
comparing and determining P (T)N) And P (T)N-1) A second target pixel point with a changed middle pixel value is judged, whether the second pixel value change range of the second target pixel point is within the preset pixel value change range or not is judged, if yes, the corresponding second target pixel point is marked as 1, otherwise, the corresponding second target pixel point is marked as 0, and a second matrix b is obtainedN(ii) a Wherein N is a positive integer;
with said anSubtracting the corresponding bNTo obtain the target matrix Cnn；
Accumulating the object matrix CnnAnd obtaining the track picture.
8. The apparatus for intelligently identifying an offending parabola as recited in claim 5, wherein the module for determining an offending parabola is specifically configured to:
and uploading the track picture to a pre-trained YOLOv3 model to obtain a violation parabolic determination result.
9. An apparatus, comprising:
a processor, and a memory coupled to the processor;
the memory for storing a computer program for performing at least the method of intelligently identifying an offending parabola of any one of claims 1-4;
the processor is used for calling and executing the computer program in the memory.
10. A storage medium, characterized in that the storage medium stores a computer program which, when executed by a processor, implements the steps of the method for intelligently identifying an offending parabola according to any one of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010451236.2A CN111639578B (en) | 2020-05-25 | 2020-05-25 | Method, device, equipment and storage medium for intelligently identifying illegal parabolic objects |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010451236.2A CN111639578B (en) | 2020-05-25 | 2020-05-25 | Method, device, equipment and storage medium for intelligently identifying illegal parabolic objects |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111639578A true CN111639578A (en) | 2020-09-08 |
| CN111639578B CN111639578B (en) | 2023-09-19 |
Family
ID=72329239
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010451236.2A Active CN111639578B (en) | 2020-05-25 | 2020-05-25 | Method, device, equipment and storage medium for intelligently identifying illegal parabolic objects |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111639578B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112329627A (en) * | 2020-11-05 | 2021-02-05 | 重庆览辉信息技术有限公司 | High-altitude throwing object distinguishing method |
| CN113516102A (en) * | 2021-08-06 | 2021-10-19 | 上海中通吉网络技术有限公司 | Deep learning parabolic behavior detection method based on video |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080200287A1 (en) * | 2007-01-10 | 2008-08-21 | Pillar Vision Corporation | Trajectory detection and feedfack system for tennis |
| CN107894252A (en) * | 2017-11-14 | 2018-04-10 | 江苏科沃纺织有限公司 | It is a kind of to monitor the buried telescopic monitoring system for being sprayed filling device running status in real time |
| WO2018095082A1 (en) * | 2016-11-28 | 2018-05-31 | 江苏东大金智信息***有限公司 | Rapid detection method for moving target in video monitoring |
| WO2018228218A1 (en) * | 2017-06-16 | 2018-12-20 | 腾讯科技（深圳）有限公司 | Identification method, computing device, and storage medium |
| CN110378935A (en) * | 2019-07-22 | 2019-10-25 | 四创科技有限公司 | Parabolic recognition methods based on image, semantic information |
| CN110782433A (en) * | 2019-10-15 | 2020-02-11 | 浙江大华技术股份有限公司 | Dynamic information violent parabolic detection method and device based on time sequence and storage medium |
| CN111079663A (en) * | 2019-12-19 | 2020-04-28 | 深圳云天励飞技术有限公司 | High-altitude parabolic monitoring method and device, electronic equipment and storage medium |
| CN111091098A (en) * | 2019-12-20 | 2020-05-01 | 浙江大华技术股份有限公司 | Training method and detection method of detection model and related device |
-
2020
- 2020-05-25 CN CN202010451236.2A patent/CN111639578B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20080200287A1 (en) * | 2007-01-10 | 2008-08-21 | Pillar Vision Corporation | Trajectory detection and feedfack system for tennis |
| WO2018095082A1 (en) * | 2016-11-28 | 2018-05-31 | 江苏东大金智信息***有限公司 | Rapid detection method for moving target in video monitoring |
| WO2018228218A1 (en) * | 2017-06-16 | 2018-12-20 | 腾讯科技（深圳）有限公司 | Identification method, computing device, and storage medium |
| CN107894252A (en) * | 2017-11-14 | 2018-04-10 | 江苏科沃纺织有限公司 | It is a kind of to monitor the buried telescopic monitoring system for being sprayed filling device running status in real time |
| CN110378935A (en) * | 2019-07-22 | 2019-10-25 | 四创科技有限公司 | Parabolic recognition methods based on image, semantic information |
| CN110782433A (en) * | 2019-10-15 | 2020-02-11 | 浙江大华技术股份有限公司 | Dynamic information violent parabolic detection method and device based on time sequence and storage medium |
| CN111079663A (en) * | 2019-12-19 | 2020-04-28 | 深圳云天励飞技术有限公司 | High-altitude parabolic monitoring method and device, electronic equipment and storage medium |
| CN111091098A (en) * | 2019-12-20 | 2020-05-01 | 浙江大华技术股份有限公司 | Training method and detection method of detection model and related device |
Non-Patent Citations (2)
| Title |
|---|
| 何亮明;胡茂林;刘海涛;: "周界视频监控中抛物检测算法" * |
| 田合雷;丁胜;于长伟;周立;: "监控视频中的移动目标侦测算法研究" * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112329627A (en) * | 2020-11-05 | 2021-02-05 | 重庆览辉信息技术有限公司 | High-altitude throwing object distinguishing method |
| CN112329627B (en) * | 2020-11-05 | 2024-02-09 | 重庆览辉信息技术有限公司 | High-altitude throwing object distinguishing method |
| CN113516102A (en) * | 2021-08-06 | 2021-10-19 | 上海中通吉网络技术有限公司 | Deep learning parabolic behavior detection method based on video |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111639578B (en) | 2023-09-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10628961B2 (en) | Object tracking for neural network systems | |
| US20190304102A1 (en) | Memory efficient blob based object classification in video analytics | |
| US10489634B2 (en) | Image processing | |
| US9710716B2 (en) | Computer vision pipeline and methods for detection of specified moving objects | |
| US10152645B2 (en) | Method and apparatus for updating a background model used for background subtraction of an image | |
| US9767570B2 (en) | Systems and methods for computer vision background estimation using foreground-aware statistical models | |
| CN110633610B (en) | Student state detection method based on YOLO | |
| US8472669B2 (en) | Object localization using tracked object trajectories | |
| US9830736B2 (en) | Segmenting objects in multimedia data | |
| US7822275B2 (en) | Method for detecting water regions in video | |
| CN108875531B (en) | Face detection method, device and system and computer storage medium | |
| CN111639578A (en) | Method, device, equipment and storage medium for intelligently identifying illegal parabola | |
| KR20170038144A (en) | Attention detection apparatus and attention detection method | |
| CN111046746A (en) | License plate detection method and device | |
| US20140003660A1 (en) | Hand detection method and apparatus | |
| CN112137591A (en) | Target object position detection method, device, equipment and medium based on video stream | |
| CN112528908A (en) | Living body detection method, living body detection device, electronic apparatus, and storage medium | |
| JP2020160804A (en) | Information processing device, program, and information processing method | |
| CN103824307A (en) | Method and device for determining invalid moving-object pixels | |
| Lee et al. | Multiple moving object segmentation using motion orientation histogram in adaptively partitioned blocks for high-resolution video surveillance systems | |
| CN116760968A (en) | Video playing effect detection method and device and computer readable storage medium | |
| US20240221426A1 (en) | Behavior detection method, electronic device, and computer readable storage medium | |
| CN111199179B (en) | Target object tracking method, terminal equipment and medium | |
| Blachut et al. | High-definition event frame generation using SoC FPGA devices | |
| CN111860070A (en) | Method and device for identifying changed object |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |