CN112163544B - Method and system for judging random placement of non-motor vehicles - Google Patents
Method and system for judging random placement of non-motor vehicles Download PDFInfo
- Publication number
- CN112163544B CN112163544B CN202011087082.XA CN202011087082A CN112163544B CN 112163544 B CN112163544 B CN 112163544B CN 202011087082 A CN202011087082 A CN 202011087082A CN 112163544 B CN112163544 B CN 112163544B
- Authority
- CN
- China
- Prior art keywords
- motor vehicle
- image
- training
- label
- placement
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000012545 processing Methods 0.000 claims abstract description 34
- 238000000605 extraction Methods 0.000 claims abstract description 30
- 238000013145 classification model Methods 0.000 claims abstract description 27
- 238000001514 detection method Methods 0.000 claims abstract description 25
- 230000015654 memory Effects 0.000 claims description 39
- 238000012549 training Methods 0.000 claims description 37
- 230000001788 irregular Effects 0.000 claims description 11
- 125000004122 cyclic group Chemical group 0.000 claims description 7
- 230000006870 function Effects 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 6
- 230000004913 activation Effects 0.000 claims description 2
- 238000004891 communication Methods 0.000 description 8
- 238000010586 diagram Methods 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 230000001133 acceleration Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000000802 evaporation-induced self-assembly Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002194 synthesizing effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
- G06V20/54—Surveillance or monitoring of activities, e.g. for recognising suspicious objects of traffic, e.g. cars on the road, trains or boats
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- 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
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Abstract
The invention discloses a method and a system for judging the random arrangement of a non-motor vehicle, wherein the method comprises the following steps: acquiring a video stream of a scene to be detected, which is shot by image acquisition equipment in real time, wherein the video stream comprises: a multi-frame image; carrying out scribing configuration of a preset detection area on a frame extraction image in a scene video stream to be detected; extracting a minimum external rectangular frame of the frame extraction image scribing configuration and performing mask processing to obtain a local image after mask processing; classifying the partial images processed by the mask by using a pre-trained non-motor vehicle random placement classification model to generate a random placement label and confidence of the non-motor vehicle; judging whether the non-motor vehicle to be detected has a random arrangement phenomenon or not according to the random arrangement label and the confidence level of the non-motor vehicle. The method provided by the invention improves the accuracy of judging the random arrangement of the non-motor vehicle, saves resources and is flexible and easy to use.
Description
Technical Field
The invention relates to the technical field of classification, in particular to a method and a system for judging the random arrangement of non-motor vehicles.
Background
In recent years, with the acceleration of the urban process, the improvement of the living standard of people, the development of sharing economy and the promotion of green travel in more and more cities, the bicycle travel is greatly promoted, and meanwhile, the non-motor vehicle lane special for parking the non-motor vehicle is planned and marked with a striking lane line. At present, sharing single vehicles is spread over all large cities and becomes an indispensable travel tool in people's life, the work load of a city administration is increased due to the fact that the bicycles are placed in disorder, at present, a video recognition technical scheme is adopted, vehicle information cannot be effectively recognized for a non-motor vehicle, and although the technical scheme of license plate video recognition and face recognition is adopted, license plates of the non-motor vehicle are not unified in specification, and license plate installation positions and modes are not unified, so that license plate recognition difficulty is high; on the other hand, the non-motor vehicle type has no unified specification, the recognition rate is low, the efficiency of non-motor vehicle video recognition is affected, and the problem of low recognition rate of the non-motor vehicle random arrangement phenomenon exists.
Disclosure of Invention
Therefore, the judging method and the judging system for the non-motor vehicle random arrangement overcome the defect of low recognition rate of the non-motor vehicle random arrangement phenomenon in the prior art.
In order to achieve the above purpose, the present invention provides the following technical solutions:
in a first aspect, an embodiment of the present invention provides a method for determining a non-motor vehicle is placed in disorder, including:
acquiring a video stream of a scene to be detected, which is shot by image acquisition equipment in real time, wherein the video stream comprises: a multi-frame image;
carrying out scribing configuration of a preset detection area on a frame extraction image in a scene video stream to be detected;
extracting a minimum external rectangular frame of the frame extraction image scribing configuration and performing mask processing to obtain a local image after mask processing;
classifying the partial images processed by the mask by using a pre-trained non-motor vehicle random placement classification model to generate a random placement label and confidence of the non-motor vehicle;
judging whether the non-motor vehicle to be detected has a random arrangement phenomenon or not according to the random arrangement label and the confidence level of the non-motor vehicle.
In an embodiment, the placing the tag includes: orderly placement, irregular placement and no non-motor vehicles.
In one embodiment, a training process for a non-motor vehicle shuffle classification model includes:
acquiring an image of an actual traffic scene, carrying out data enhancement on the image of the actual traffic scene, carrying out scribing configuration on the frame extraction image with the data enhancement, and marking a minimum circumscribed rectangular frame and a vehicle label of a non-motor vehicle target detection area of the traffic scene to synthesize a training data set;
extracting a minimum circumscribed rectangular frame and a vehicle label of a target detection area of the non-motor vehicle in a training data set, inputting the minimum circumscribed rectangular frame as a regression parameter true value into a network model, calculating a regression loss function, and training the network model;
inputting information of the minimum circumscribed rectangular frame and the vehicle label into an input network model, calculating a loss value, adjusting a learning rate, and performing cyclic training until the loss value is smaller than a first preset threshold value or the cycle number is larger than a preset value, stopping training, and obtaining a trained non-motor vehicle disordered classification model.
In one embodiment, the non-motor vehicle is classified by using the Efficient Net B3 model.
In an embodiment, further comprising: and outputting early warning information of irregular placement when the irregular placement and the confidence coefficient are larger than a second preset threshold value.
In an embodiment, the early warning information further includes: places where non-motor vehicles are placed in disorder and time for placing in disorder.
In a second aspect, an embodiment of the present invention provides a system for determining a non-motor vehicle is out of order, including:
the image acquisition module is used for acquiring a video stream of a scene to be detected, which is shot by the image acquisition equipment in real time, wherein the video stream comprises: a multi-frame image;
the scribing configuration module is used for carrying out scribing configuration of a preset detection area on the frame extraction image in the scene video stream to be detected;
the image processing module is used for extracting the minimum external rectangular frame of the frame-drawing image scribing configuration and performing mask processing to obtain a local image after the mask processing;
the model classification module is used for classifying the partial images processed by the mask by using a pre-trained non-motor vehicle random placement classification model to generate a non-motor vehicle random placement label and confidence;
the model judging module is used for judging whether the non-motor vehicle to be detected has a random arrangement phenomenon or not according to the random arrangement label and the confidence level of the non-motor vehicle.
In a third aspect, an embodiment of the present invention provides a terminal, including: the system comprises at least one processor and a memory communicatively connected with the at least one processor, wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to cause the at least one processor to execute the non-motor vehicle misarrangement judging method according to the first aspect of the embodiment of the invention.
In a fourth aspect, an embodiment of the present invention provides a computer readable storage medium, where computer instructions are stored, where the computer instructions are configured to cause the computer to execute the method for determining a non-motor vehicle misarrangement according to the first aspect of the embodiment of the present invention.
The technical scheme of the invention has the following advantages:
according to the judging method and system for the random arrangement of the non-motor vehicle, provided by the invention, the video stream of the scene to be detected, which is shot in real time by the image acquisition equipment, is obtained, and the video stream comprises the following components: a multi-frame image; carrying out scribing configuration of a preset detection area on a frame extraction image in a scene video stream to be detected; extracting a minimum external rectangular frame of the frame extraction image scribing configuration and performing mask processing to obtain a local image after mask processing; classifying the partial images processed by the mask by using a pre-trained non-motor vehicle random placement classification model to generate a random placement label and confidence of the non-motor vehicle; judging whether the non-motor vehicle to be detected has a random arrangement phenomenon or not according to the random arrangement label and the confidence level of the non-motor vehicle. According to the invention, through training of the non-motor vehicle random placement classification model, scribing configuration of a preset detection area is carried out according to frame extraction images in different scenes to be detected, a minimum external rectangular frame of the frame extraction image scribing configuration is extracted, mask processing is carried out, a local image after mask processing is obtained, the local image is input into the non-motor vehicle random placement classification model, and whether the non-motor vehicle to be detected has random placement phenomenon is judged according to random placement labels and confidence of the non-motor vehicle. The method provided by the invention improves the accuracy of judging the random arrangement of the non-motor vehicle, saves resources and is flexible and easy to use.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the description of the embodiments or the prior art will be briefly described, and it is obvious that the drawings in the description below are some embodiments of the present invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.
FIG. 1 is a flowchart of a specific example of a method for determining a non-motor vehicle is disordered according to an embodiment of the invention;
FIG. 2 is a block diagram of a system for determining a non-motor vehicle is randomly placed according to an embodiment of the present invention;
fig. 3 is a composition diagram of a specific example of a terminal according to an embodiment of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made apparent and fully in view of the accompanying drawings, in which some, but not all embodiments of the invention are shown. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
In addition, the technical features of the different embodiments of the present invention described below may be combined with each other as long as they do not collide with each other.
Example 1
The method for judging the random arrangement of the non-motor vehicle provided by the embodiment of the invention, as shown in fig. 1, comprises the following steps:
step S1: acquiring a video stream of a scene to be detected, which is shot by image acquisition equipment in real time, wherein the video stream comprises: and (5) multi-frame images.
In the embodiment of the invention, the video stream of the scene to be detected is obtained in real time by installing the corresponding camera acquisition device, the video stream is composed of a plurality of frames of images, and the camera acquisition device is selected according to actual requirements in practical application by way of example only and not limitation; according to the requirements of scene detection to be detected, the installation positions and the number of the corresponding camera shooting acquisition devices are selected, and corresponding changes are made in practice according to actual requirements.
Step S2: and carrying out scribing configuration of a preset detection area on the frame extraction image in the video stream of the scene to be detected.
In the embodiment of the invention, the frame extraction image is a key frame in the extracted video stream, and the key frame is: an image extracted every twenty minutes is only taken as an example, but not limited to, and in practical application, a corresponding time interval is selected according to practical requirements to extract frames; the preset detection area is a scene to be detected of the non-motor vehicle, and the corresponding scene to be detected is selected according to actual requirements in actual application.
In the embodiment of the invention, the method for scribing configuration according to different scenes to be detected comprises the following steps: drawing polygons of the area to be detected, and marking the polygons by adopting manual marking; for example: marking a bicycle running area along one row of bicycle edges, properly extending the areas at two ends, starting from the upper left corner when the marking frame marks a quadrangle, recording polygon vertex information at the same time according to the clockwise direction, and selecting corresponding polygons and marking methods for marking according to the actual requirements of the area to be detected in practical application. Compared with the method using an algorithm, the embodiment adopts a scribing configuration mode, so that the video memory resource is saved, and meanwhile, the time consumed by the algorithm is saved.
Step S3: and extracting the minimum circumscribed rectangular frame of the frame extraction image scribing configuration, and performing mask processing to obtain a local image after the mask processing.
In the embodiment of the invention, the local picture of the non-motor vehicle, namely the minimum circumscribed rectangular frame of the picture with the scribing configuration is extracted, and the mask processing is carried out on the outside of the region of the minimum circumscribed rectangular frame of the extracted local picture.
Transmitting the processed pictures to a task end in a json data format, and sending the pictures to a non-motor vehicle random placement classification model pre-trained in the step S4, wherein the input data mainly comprises: information such as ID of video capture device, ID of current frame, input data type, data format, image data, size of image, frame rate of video stream, ID of server, task ID, current time, video ID, etc.: by way of example only, and not by way of limitation, corresponding input data is selected in practical applications.
Step S4: and classifying the partial images processed by the mask by using the pre-trained non-motor vehicle random placement classification model to generate a non-motor vehicle random placement label and confidence.
In the embodiment of the invention, the training process of the non-motor vehicle disordered classification model comprises the following steps: acquiring an image of an actual traffic scene, carrying out data enhancement on the image of the actual traffic scene, carrying out scribing configuration on the frame extraction image with the data enhancement, and marking a minimum circumscribed rectangular frame and a vehicle label of a non-motor vehicle target detection area of the traffic scene to synthesize a training data set; extracting a minimum circumscribed rectangular frame and a vehicle label of a target detection area of the non-motor vehicle in a training data set, inputting the minimum circumscribed rectangular frame as a regression parameter true value into a network model, calculating a regression loss function, and training the network model; inputting information of the minimum circumscribed rectangular frame and the vehicle label into the network model, calculating a loss value, adjusting a learning rate, and performing cyclic training until the loss value is smaller than a first preset threshold value or the cyclic frequency is larger than a preset value (the first preset threshold value and the preset value can be set according to expert experience values and are not limited here), stopping training, and obtaining a trained non-motor vehicle disordered classification model.
In the training process of the non-motor vehicle disordered classification model, the actual application scene is collected, such as: the method comprises the steps that image data of a non-motor vehicle under a city management monitoring camera can be selected from about 300 video points of a plurality of cities, videos with recording time of 1 minute are recorded from six early points to eight late points at intervals of twenty minutes, the recording time is about one week, representative pictures are extracted from about 10 ten thousand obtained videos, multi-angle samples are collected from a plurality of cities, a plurality of scenes and a plurality of time periods, each picture is required to be clear, human eyes are easy to distinguish, and when the pictures are actually collected, the corresponding representative pictures are selected according to actual requirements. Meanwhile, the data enhancement is carried out on the selected picture, and the data enhancement method comprises the following steps: the data set is further enlarged by methods of random rotation, random blurring, random color transformation, normalization and the like, which are only used as examples, but not limited thereto, and corresponding data enhancement modes are selected according to actual demands in practical application. Carrying out scribing configuration on the data-enhanced frame extraction image, marking a minimum circumscribed rectangular frame and a vehicle label of a non-motor vehicle target detection area of a traffic scene, and synthesizing a training data set;
in an embodiment of the present invention, placing a tag includes: orderly placement, irregular placement and no non-motor vehicles. For example: the bicycle placing areas in the scene are far, and the situation is not marked; distinction between tidy and irregular: the conditions of unordered placement and serious front-rear irregular placement of the bicycle are considered as irregular, and the method is not limited to the example, and the division standard of the corresponding placement label is selected according to actual requirements in practical application.
In the embodiment of the invention, based on the deep learning theory, a swish activation function of an EfficientNet B3 model is adopted to conduct classification training on the disordered arrangement of the non-motor vehicles. Extracting a minimum circumscribed rectangular frame and a vehicle label of a target detection area of the non-motor vehicle in a training data set, inputting the minimum circumscribed rectangular frame as a regression parameter true value into a network model, calculating a regression loss function, and training the network model; inputting information of the minimum circumscribed rectangular frame and the vehicle label into the network model, calculating a loss value, adjusting a learning rate, and performing cyclic training until the loss value is smaller than a first preset threshold value or the cyclic frequency is larger than a preset value (the first preset threshold value and the preset value can be set according to expert experience values and are not limited here), stopping training, and obtaining a trained non-motor vehicle disordered classification model.
Step S5: judging whether the non-motor vehicle to be detected has a random arrangement phenomenon or not according to the random arrangement label and the confidence level of the non-motor vehicle.
In the embodiment of the invention, when the tag is not orderly placed and the confidence is larger than the second preset threshold, outputting early warning information of not orderly placed, and selecting a corresponding threshold according to actual requirements in actual application; the early warning information comprises: the location and time of the non-motor vehicles are not limited by the examples.
The method for judging the random arrangement of the non-motor vehicle provided by the embodiment of the invention comprises the steps of obtaining a video stream of a scene to be detected, which is shot by image acquisition equipment in real time, wherein the video stream comprises the following components: a multi-frame image; carrying out scribing configuration of a preset detection area on a frame extraction image in a scene video stream to be detected; extracting a minimum external rectangular frame of the frame extraction image scribing configuration and performing mask processing to obtain a local image after mask processing; classifying the partial images processed by the mask by using a pre-trained non-motor vehicle random placement classification model to generate a random placement label and confidence of the non-motor vehicle; judging whether the non-motor vehicle to be detected has a random arrangement phenomenon or not according to the random arrangement label and the confidence level of the non-motor vehicle. According to the embodiment of the invention, through training of the non-motor vehicle disordered placement classification model, the scribing configuration of the preset detection area is carried out according to the frame extraction images in different scenes to be detected, the minimum external rectangular frame of the scribing configuration of the frame extraction images is extracted, mask processing is carried out, the local image after mask processing is obtained, the local image is input into the non-motor vehicle disordered placement classification model, and whether the non-motor vehicle to be detected has disordered placement phenomenon is judged according to the disordered placement label and the confidence level of the non-motor vehicle. The method provided by the embodiment of the invention improves the accuracy of judging the random arrangement of the non-motor vehicle, saves resources and is flexible and easy to use.
Example 2
An embodiment of the present invention provides a system for determining a non-motor vehicle being placed in disorder, as shown in fig. 2, including:
the image acquisition module 1 is used for acquiring a video stream of a scene to be detected, which is shot by the image acquisition equipment in real time, wherein the video stream comprises: a multi-frame image; this module performs the method described in step S1 in embodiment 1, and will not be described here again.
The scribing configuration module 2 is used for carrying out scribing configuration of a preset detection area on the frame extraction image in the scene video stream to be detected; this module performs the method described in step S2 in embodiment 1, and will not be described here.
The image processing module 3 is used for extracting the minimum external rectangular frame of the frame extraction image scribing configuration and performing mask processing to obtain a local image after the mask processing; this module performs the method described in step S3 in embodiment 1, and will not be described here.
The model classification module 4 is used for classifying the partial images processed by the mask by utilizing a pre-trained non-motor vehicle disordered placement classification model to generate a non-motor vehicle disordered placement label and confidence; this module performs the method described in step S4 in embodiment 1, and will not be described here.
The model judging module 5 is used for judging whether the non-motor vehicle to be detected has a random arrangement phenomenon or not according to the random arrangement label and the confidence level of the non-motor vehicle; this module performs the method described in step S5 in embodiment 1, and will not be described here.
The embodiment of the invention provides a judging system for random placement of a non-motor vehicle, which is used for acquiring a video stream of a scene to be detected, which is shot by image acquisition equipment in real time, through an image acquisition module, wherein the video stream comprises the following components: a multi-frame image; the scribing configuration module is used for carrying out scribing configuration of a preset detection area on the frame extraction image in the scene video stream to be detected; extracting a minimum external rectangular frame of the frame extraction image scribing configuration from an image processing module, and performing mask processing to obtain a local image after mask processing; in a model classification module, classifying the partial images processed by the mask by using a pre-trained non-motor vehicle random placement classification model to generate a non-motor vehicle random placement label and confidence; judging whether the non-motor vehicle to be detected has a random arrangement phenomenon or not according to the random arrangement label and the confidence level of the non-motor vehicle. The accuracy of judging the random arrangement of the non-motor vehicles is improved, and meanwhile, the resources are saved, and the non-motor vehicle random arrangement judging device is flexible and easy to use. According to the embodiment of the invention, through training of the non-motor vehicle disordered placement classification model, the scribing configuration of the preset detection area is carried out according to the frame extraction images in different scenes to be detected, the minimum external rectangular frame of the scribing configuration of the frame extraction images is extracted, mask processing is carried out, the local image after mask processing is obtained, the local image is input into the non-motor vehicle disordered placement classification model, and whether the non-motor vehicle to be detected has disordered placement phenomenon is judged according to the disordered placement label and the confidence level of the non-motor vehicle. The method provided by the embodiment of the invention improves the accuracy of judging the random arrangement of the non-motor vehicle, saves resources and is flexible and easy to use.
Example 3
An embodiment of the present invention provides a terminal, as shown in fig. 3, including: at least one processor 401, such as a CPU (Central Processing Unit ), at least one communication interface 403, a memory 404, at least one communication bus 402. Wherein communication bus 402 is used to enable connected communications between these components. The communication interface 403 may include a Display screen (Display) and a Keyboard (Keyboard), and the optional communication interface 403 may further include a standard wired interface and a wireless interface. The memory 404 may be a high-speed RAM memory (Random Access Memory) or a nonvolatile memory (nonvolatile memory), such as at least one magnetic disk memory. The memory 404 may also optionally be at least one storage device located remotely from the aforementioned processor 401. Wherein the processor 401 may execute the non-motor vehicle shuffle determination method of embodiment 1. A set of program codes is stored in the memory 404, and the processor 401 calls the program codes stored in the memory 404 for executing the judging method of the non-motor vehicle shuffle in embodiment 1. The communication bus 402 may be a peripheral component interconnect standard (peripheral component interconnect, PCI) bus or an extended industry standard architecture (extended industry standard architecture, EISA) bus, among others. Communication bus 402 may be divided into an address bus, a data bus, a control bus, and the like. For ease of illustration, only one line is shown in fig. 3, but not only one bus or one type of bus. Wherein the memory 404 may include volatile memory (English) such as random-access memory (RAM); the memory may also include a nonvolatile memory (english: non-volatile memory), such as a flash memory (english: flash memory), a hard disk (english: hard disk drive, abbreviated as HDD) or a solid-state drive (english: SSD); memory 404 may also include a combination of the above types of memory. The processor 401 may be a central processor (English: central processing unit, abbreviated: CPU), a network processor (English: network processor, abbreviated: NP) or a combination of CPU and NP.
Wherein the memory 404 may include volatile memory (English) such as random-access memory (RAM); the memory may also include a nonvolatile memory (english: non-volatile memory), such as a flash memory (english: flash memory), a hard disk (english: hard disk drive, abbreviated as HDD) or a solid state disk (english: solid-state drive, abbreviated as SSD); memory 404 may also include a combination of the above types of memory.
The processor 401 may be a central processor (English: central processing unit, abbreviated: CPU), a network processor (English: network processor, abbreviated: NP) or a combination of CPU and NP.
Wherein the processor 401 may further comprise a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a Programmable Logic Device (PLD), or a combination thereof (English: programmable logic device). The PLD may be a complex programmable logic device (English: complex programmable logic device, abbreviated: CPLD), a field programmable gate array (English: field-programmable gate array, abbreviated: FPGA), a general-purpose array logic (English: generic array logic, abbreviated: GAL), or any combination thereof.
Optionally, the memory 404 is also used for storing program instructions. The processor 401 may invoke program instructions to implement the non-motor vehicle shuffle determination method as in embodiment 1 of the present application.
The embodiment of the invention also provides a computer readable storage medium, and the computer readable storage medium stores computer executable instructions, and the computer executable instructions can execute the method for judging the non-motor vehicle is out of order in the embodiment 1. Wherein the storage medium may be a magnetic Disk, an optical Disk, a Read-Only Memory (ROM), a random access Memory (Random Access Memory, RAM), a Flash Memory (Flash Memory), a Hard Disk (HDD), or a Solid State Drive (SSD); the storage medium may also comprise a combination of memories of the kind described above.
It is apparent that the above examples are given by way of illustration only and are not limiting of the embodiments. Other variations or modifications of the above teachings will be apparent to those of ordinary skill in the art. It is not necessary here nor is it exhaustive of all embodiments. And obvious variations or modifications thereof are contemplated as falling within the scope of the present invention.
Claims (5)
1. A method for judging the disordered placement of a non-motor vehicle is characterized by comprising the following steps:
acquiring a video stream of a scene to be detected, which is shot by image acquisition equipment in real time, wherein the video stream comprises: a multi-frame image;
carrying out scribing configuration of a preset detection area on a frame extraction image in a scene video stream to be detected;
extracting a minimum external rectangular frame of the frame extraction image scribing configuration and performing mask processing to obtain a local image after mask processing;
classifying the partial images processed by the mask by using a pre-trained non-motor vehicle random placement classification model to generate a random placement label and confidence of the non-motor vehicle; the placing tag comprises: orderly and irregularly placed, and no non-motor vehicle exists;
judging whether the non-motor vehicle to be detected has a random arrangement phenomenon or not according to the random arrangement label and the confidence level of the non-motor vehicle; outputting early warning information of irregular placement when the irregular placement is larger than a second preset threshold value;
the training process for the non-motor vehicle disordered classification model comprises the following steps:
acquiring an image of an actual traffic scene, carrying out data enhancement on the image of the actual traffic scene, carrying out scribing configuration on the frame extraction image with the data enhancement, and marking a minimum circumscribed rectangular frame and a vehicle label of a non-motor vehicle target detection area of the traffic scene to synthesize a training data set;
extracting a minimum circumscribed rectangular frame and a vehicle label of a target detection area of the non-motor vehicle in a training data set, inputting the minimum circumscribed rectangular frame as a regression parameter true value into a network model, calculating a regression loss function, and training the network model;
inputting information of the minimum circumscribed rectangular frame and a vehicle label into an input network model, calculating a loss value, adjusting a learning rate, and performing cyclic training until the loss value is smaller than a first preset threshold value or the cycle number is larger than a preset value, stopping training, and obtaining a trained non-motor vehicle disordered placement classification model; wherein, adopt EfficientNet B3 model, carry on the classification training to the unordered arrangement of the non-motor vehicle.
2. The method for determining the disorder of the non-motor vehicle according to claim 1, wherein the pre-warning information further comprises: places where non-motor vehicles are placed in disorder and time for placing in disorder.
3. A non-motor vehicle misplacement determination system, comprising:
the image acquisition module is used for acquiring a video stream of a scene to be detected, which is shot by the image acquisition equipment in real time, wherein the video stream comprises: a multi-frame image;
the scribing configuration module is used for carrying out scribing configuration of a preset detection area on the frame extraction image in the scene video stream to be detected;
the image processing module is used for extracting the minimum external rectangular frame of the frame-drawing image scribing configuration and performing mask processing to obtain a local image after the mask processing;
the model classification module is used for classifying the partial images processed by the mask by using a pre-trained non-motor vehicle random placement classification model to generate a non-motor vehicle random placement label and confidence; the placing tag comprises: orderly and irregularly placed, and no non-motor vehicle exists;
the model judging module is used for judging whether the non-motor vehicle to be detected has a random arrangement phenomenon or not according to the random arrangement label and the confidence level of the non-motor vehicle; outputting early warning information of irregular placement when the irregular placement is larger than a second preset threshold value;
the training process for the non-motor vehicle disordered classification model comprises the following steps: acquiring an image of an actual traffic scene, carrying out data enhancement on the image of the actual traffic scene, carrying out scribing configuration on the frame extraction image with the data enhancement, and marking a minimum circumscribed rectangular frame and a vehicle label of a non-motor vehicle target detection area of the traffic scene to synthesize a training data set; extracting a minimum circumscribed rectangular frame and a vehicle label of a target detection area of the non-motor vehicle in a training data set, inputting the minimum circumscribed rectangular frame as a regression parameter true value into a network model, calculating a regression loss function, and training the network model; inputting information of the minimum circumscribed rectangular frame and a vehicle label into an input network model, calculating a loss value, adjusting a learning rate, and performing cyclic training until the loss value is smaller than a first preset threshold value or the cycle number is larger than a preset value, stopping training, and obtaining a trained non-motor vehicle disordered placement classification model; wherein, adopt EfficientNet B3 model, carry on the classification training to the non-motor vehicle disorder arrangement; and classifying and training the non-motor vehicle by adopting a swish activation function of an EfficientNet B3 model.
4. A terminal, comprising: at least one processor, and a memory communicatively coupled to the at least one processor, wherein the memory stores instructions executable by the at least one processor to cause the at least one processor to perform the non-motorized vehicle shuffle determination method of any of claims 1-2.
5. A computer-readable storage medium storing computer instructions for causing the computer to execute the non-motor vehicle shuffle determination method of any one of claims 1-2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011087082.XA CN112163544B (en) | 2020-10-12 | 2020-10-12 | Method and system for judging random placement of non-motor vehicles |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011087082.XA CN112163544B (en) | 2020-10-12 | 2020-10-12 | Method and system for judging random placement of non-motor vehicles |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112163544A CN112163544A (en) | 2021-01-01 |
| CN112163544B true CN112163544B (en) | 2024-02-20 |
Family
ID=73866547
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011087082.XA Active CN112163544B (en) | 2020-10-12 | 2020-10-12 | Method and system for judging random placement of non-motor vehicles |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112163544B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113055823B (en) * | 2021-04-01 | 2023-07-11 | 超级视线科技有限公司 | Method and device for managing shared bicycle based on road side parking |
| CN113221800A (en) * | 2021-05-24 | 2021-08-06 | 珠海大横琴科技发展有限公司 | Monitoring and judging method and system for target to be detected |
| CN113762229B (en) * | 2021-11-10 | 2022-02-08 | 山东天亚达新材料科技有限公司 | Intelligent identification method and system for building equipment in building site |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109344782A (en) * | 2018-10-11 | 2019-02-15 | 珠海市杰理科技股份有限公司 | Shared parked vehicle monitoring method, device |
| CN110096981A (en) * | 2019-04-22 | 2019-08-06 | 长沙千视通智能科技有限公司 | A kind of video big data traffic scene analysis method based on deep learning |
| CN110705404A (en) * | 2019-09-20 | 2020-01-17 | 北京文安智能技术股份有限公司 | Method, device and system for detecting random discharge of shared bicycle |
| CN111444845A (en) * | 2020-03-26 | 2020-07-24 | 江苏集萃华科智能装备科技有限公司 | Non-motor vehicle illegal parking identification method, device and system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11244171B2 (en) * | 2014-01-22 | 2022-02-08 | Conduent Business Services Llc | Video-based system for automated detection of double parking violations |
-
2020
- 2020-10-12 CN CN202011087082.XA patent/CN112163544B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109344782A (en) * | 2018-10-11 | 2019-02-15 | 珠海市杰理科技股份有限公司 | Shared parked vehicle monitoring method, device |
| CN110096981A (en) * | 2019-04-22 | 2019-08-06 | 长沙千视通智能科技有限公司 | A kind of video big data traffic scene analysis method based on deep learning |
| CN110705404A (en) * | 2019-09-20 | 2020-01-17 | 北京文安智能技术股份有限公司 | Method, device and system for detecting random discharge of shared bicycle |
| CN111444845A (en) * | 2020-03-26 | 2020-07-24 | 江苏集萃华科智能装备科技有限公司 | Non-motor vehicle illegal parking identification method, device and system |
Non-Patent Citations (1)
| Title |
|---|
| 基于改进实例分割网络的步行道违停车辆检测算法;郑雅羽等;《高技术通讯》;第1-5节 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112163544A (en) | 2021-01-01 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112163544B (en) | Method and system for judging random placement of non-motor vehicles | |
| CN109272509B (en) | Target detection method, device and equipment for continuous images and storage medium | |
| CN111382704B (en) | Vehicle line pressing violation judging method and device based on deep learning and storage medium | |
| EP3806064A1 (en) | Method and apparatus for detecting parking space usage condition, electronic device, and storage medium | |
| CN102867418B (en) | Method and device for judging license plate identification accuracy | |
| CN112163543A (en) | Method and system for detecting illegal lane occupation of vehicle | |
| CN109711407B (en) | License plate recognition method and related device | |
| CN110111581A (en) | Target identification method, device, computer equipment and storage medium | |
| CN110781980B (en) | Training method of target detection model, target detection method and device | |
| CN112329881B (en) | License plate recognition model training method, license plate recognition method and device | |
| CN110705370B (en) | Road condition identification method, device, equipment and storage medium based on deep learning | |
| CN110458126B (en) | Pantograph state monitoring method and device | |
| CN110991310B (en) | Portrait detection method, device, electronic equipment and computer readable medium | |
| CN111047874B (en) | Intelligent traffic violation management method and related product | |
| US20190311492A1 (en) | Image foreground detection apparatus and method and electronic device | |
| CN110599453A (en) | Panel defect detection method and device based on image fusion and equipment terminal | |
| CN110782409A (en) | Method for removing shadow of multi-motion object | |
| CN114155551A (en) | Improved pedestrian detection method and device based on YOLOv3 under complex environment | |
| CN113963316A (en) | Target event determination method and device, storage medium and electronic device | |
| CN112906471A (en) | Traffic signal lamp identification method and device | |
| CN116311205A (en) | License plate recognition method, license plate recognition device, electronic equipment and storage medium | |
| CN112990350B (en) | Target detection network training method and target detection network-based coal and gangue identification method | |
| CN115311630A (en) | Method and device for generating distinguishing threshold, training target recognition model and recognizing target | |
| CN112597924A (en) | Electric bicycle track tracking method, camera device and server | |
| CN110443244B (en) | Graphics processing method and related device |
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 |