CN111310652B - Flow identification method and system based on classification regression double-domain model - Google Patents
Flow identification method and system based on classification regression double-domain model Download PDFInfo
- Publication number
- CN111310652B CN111310652B CN202010090623.8A CN202010090623A CN111310652B CN 111310652 B CN111310652 B CN 111310652B CN 202010090623 A CN202010090623 A CN 202010090623A CN 111310652 B CN111310652 B CN 111310652B
- Authority
- CN
- China
- Prior art keywords
- target
- classification
- regression
- picture
- value
- 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
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
- G06V20/53—Recognition of crowd images, e.g. recognition of crowd congestion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- 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
- G06F18/2415—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on parametric or probabilistic models, e.g. based on likelihood ratio or false acceptance rate versus a false rejection rate
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Mathematical Optimization (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computational Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Evolutionary Biology (AREA)
- Mathematical Analysis (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Probability & Statistics with Applications (AREA)
- Pure & Applied Mathematics (AREA)
- Algebra (AREA)
- Operations Research (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- Multimedia (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Image Analysis (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The embodiment of the invention provides a flow identification method based on a classification regression double-domain model, which comprises the following steps: collecting a target picture in a preset area; inputting the target picture into a classification regression double-domain model, and outputting a target classification value and a target regression value of the target picture through the classification regression double-domain model, wherein the classification double-domain model comprises a pre-trained classifier and a pre-trained regressor, and the classifier and the regressor are used for identifying the flow in the target picture; judging whether the target classification value is larger than a preset classification value or not; outputting a flow identification result based on the target regression value when the target classification value is larger than a preset classification value; and outputting the flow identification result based on the target classification value when the target classification value is not greater than a preset classification value. The embodiment of the invention can accurately identify the number of individuals in a complex scene.
Description
Technical Field
The embodiment of the invention relates to the field of image recognition, in particular to a flow recognition method, a system, computer equipment and a computer readable storage medium based on a classification regression double-domain model.
Background
Along with the progress of science and technology and the improvement of living standard of people, social activities of people are gradually increased, traffic flow or traffic flow congestion in markets, transportation hubs, large-scale activity sites and large-scale public places is more and more serious, potential safety hazards caused by the traffic or traffic congestion are more and more serious, currently, how to automatically and real-timely estimate the number of people (or other objects needing to be counted) in a complex scene has important research value, and has deep guiding significance for providing effective event decisions for public affair staff, but the current flow identification model has generally lower individual number identification precision in the complex scene such as the markets, the transportation hubs and the like.
Therefore, how to accurately identify the flow of people, vehicles and the like in complex scenes such as markets, transportation hubs and the like by using the model becomes one of the technical problems to be solved at present.
Disclosure of Invention
In view of the foregoing, it is necessary to provide a flow identification method, system, computer device and computer readable storage medium based on a classification regression dual-domain model, so as to solve the technical problem that the accuracy of identifying the predicted value of the individual number model is low in a complex scene.
In order to achieve the above object, an embodiment of the present invention provides a method for identifying traffic based on a classification regression dual-domain model, where the method includes the steps of:
collecting a target picture in a preset area;
inputting the target picture into a classification regression double-domain model, and outputting a target classification value and a target regression value of the target picture through the classification regression double-domain model, wherein the classification double-domain model comprises a pre-trained classifier and a pre-trained regressor, and the classifier and the regressor are used for identifying the flow in the target picture;
judging whether the target classification value is larger than a preset classification value or not;
outputting a flow identification result based on the target regression value when the target classification value is larger than a preset classification value;
and outputting the flow identification result based on the target classification value when the target classification value is not greater than a preset classification value.
Illustratively, the step of outputting the target classification value and the target regression value of the target picture through the classification regression dual-domain model includes:
classifying the target picture through the classifier to obtain a plurality of confidence degrees corresponding to the target picture, wherein each confidence degree is used for representing the probability that the target picture belongs to one of a plurality of picture categories;
determining a target picture category of the target picture according to the picture category corresponding to the highest confidence coefficient in the plurality of confidence coefficients;
and determining a target classification value of the target picture according to the target picture category, wherein the target classification value is used for representing the number of individuals in the target picture.
Illustratively, the step of outputting the target classification value and the target regression value of the target picture through the classification regression dual-domain model includes:
carrying out regression processing on the target picture through a regressing device to obtain a numerical value corresponding to the number of individuals in the target picture;
and performing integer processing on the numerical value to obtain an integer numerical value, and determining the integer numerical value as the target regression value.
Illustratively, the method further comprises: the flow identification result is sent to target equipment, so that the target equipment executes corresponding operation according to the flow identification result;
or generating a corresponding operation instruction according to the flow identification result so as to control the target equipment to execute corresponding operation through the operation instruction.
Illustratively, the method further comprises the step of training the classifier:
acquiring an initial classifier;
acquiring a plurality of initial classified pictures, wherein each initial classified picture is allocated with a first quantity label for representing the quantity of individuals;
dividing the plurality of initial classified pictures into n+2 picture categories according to the first number labels corresponding to each initial classified picture, and configuring corresponding picture category labels, wherein N is a positive integer, and the n+2 picture categories comprise a picture category with the number of 0 individuals, a picture category with the number of N individuals, and a picture category with the number of one individual being greater than N;
configuring a plurality of classification training samples according to the initial classification pictures corresponding to the picture category labels; a kind of electronic device with high-pressure air-conditioning system
And training the initial classifier through the plurality of classification training samples to obtain a trained classifier.
Illustratively, the method further comprises the training step of the regressor:
acquiring an initial regressive device;
acquiring a plurality of initial regression pictures, wherein each initial regression picture is allocated with a second number label for representing the number of individuals;
configuring a plurality of regression training samples according to the initial regression pictures corresponding to the second quantity labels; a kind of electronic device with high-pressure air-conditioning system
And training the initial regressive device through the multiple regression training samples to obtain a trained regressive device.
Illustratively, the method further comprises: the flow identification result is sent to target equipment, and the target equipment is controlled to perform corresponding operation according to the flow identification result and an operation instruction preset by the target equipment
In order to achieve the above object, an embodiment of the present invention further provides a traffic identification system based on a classification regression dual-domain model, including:
the acquisition module is used for acquiring a target picture in a preset area;
the input module is used for inputting the target picture into a classification regression double-domain model, and outputting a target classification value and a target regression value of the target picture through the classification regression double-domain model, wherein the classification double-domain model comprises a pre-trained classifier and a pre-trained regressor, and the classifier and the regressor are used for identifying the flow in the target picture;
the judging module is used for judging whether the target classification value is larger than a preset classification value or not;
the output module is used for outputting a flow identification result based on the target regression value when the target classification value is larger than a preset classification value; and outputting the flow identification result based on the target classification value when the target classification value is not greater than a preset classification value.
Illustratively, the input module is further configured to:
classifying the target picture through the classifier to obtain a plurality of confidence degrees corresponding to the target picture, wherein each confidence degree is used for representing the probability that the target picture belongs to one of a plurality of picture categories;
determining a target picture category of the target picture according to the picture category corresponding to the highest confidence coefficient in the plurality of confidence coefficients;
and determining a target classification value of the target picture according to the target picture category, wherein the target classification value is used for representing the number of individuals in the target picture.
Illustratively, the input module is further configured to:
carrying out regression processing on the target picture through a regressing device to obtain a numerical value corresponding to the number of individuals in the target picture;
and performing integer processing on the numerical value to obtain an integer numerical value, and determining the integer numerical value as the target regression value.
To achieve the above object, an embodiment of the present invention further provides a computer device, where the computer device includes a memory, a processor, and a computer program stored on the memory and executable on the processor, where the computer program is executed by the processor to implement the steps of the method for identifying traffic based on the classification regression double domain model as described above.
To achieve the above object, an embodiment of the present invention further provides a computer readable storage medium having stored therein a computer program executable by at least one processor to cause the at least one processor to perform the steps of the method for traffic identification based on a classification regression double domain model as described above.
The flow identification method, the system, the computer equipment and the computer readable storage medium based on the classification regression double-domain model provided by the embodiment of the invention provide an effective prediction and identification method for individuals and the number of individuals appearing in the picture; the embodiment of the invention combines the classifier and the regressor, so that the model can accurately identify the number of individuals in a complex scene.
Drawings
Fig. 1 is a flow chart of a flow identification method based on a classification regression dual-domain model according to an embodiment of the invention.
Fig. 2 is a schematic program module diagram of a flow identification system based on a classification regression dual-domain model according to a second embodiment of the present invention.
Fig. 3 is a schematic diagram of a hardware structure of a third embodiment of the computer device of the present invention.
Detailed Description
The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention. 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.
It should be noted that the description of "first", "second", etc. in this disclosure is for descriptive purposes only and is not to be construed as indicating or implying a relative importance or implying an indication of the number of technical features being indicated. Thus, a feature defining "a first" or "a second" may explicitly or implicitly include at least one such feature. In addition, the technical solutions of the embodiments may be combined with each other, but it is necessary to base that the technical solutions can be realized by those skilled in the art, and when the technical solutions are contradictory or cannot be realized, the combination of the technical solutions should be considered to be absent and not within the scope of protection claimed in the present invention.
In the following embodiments, an exemplary description will be made with the computer device 2 as an execution subject.
Example 1
Referring to fig. 1, a flow chart of steps of a flow identification method based on a classification regression dual-domain model according to an embodiment of the invention is shown. It will be appreciated that the flow charts in the method embodiments are not intended to limit the order in which the steps are performed. An exemplary description will be made below with the computer device 2 as an execution subject. Specifically, the following is described.
Step S100, collecting target pictures in a preset area.
The target picture may be collected by a picture collecting device, and the preset area may be a mall, a street, a road, a subway, a parking lot, a scenic spot or other places of activity, for example, a picture of a subway entrance may be obtained by photographing through a camera of the subway entrance.
Step S102, inputting the target picture into a classification regression double-domain model, and outputting a target classification value and a target regression value of the target picture through the classification regression double-domain model, wherein the classification double-domain model comprises a pre-trained classifier and a pre-trained regressor, and the classifier and the regressor are used for identifying the flow in the target picture.
For example, the target classification value and the target regression value may be a numerical value, which are used to represent the number of individuals in the target image, and here, taking a preset area "subway entrance" as an example, the subway entrance image shot at the subway entrance is identified by using a regression double-domain model, so as to obtain the target classification value and the target regression value.
Illustratively, the step S102 may further include:
step S102a1, performing classification processing on the target picture by using the classifier to obtain a plurality of confidence coefficients corresponding to the target picture.
For example, each confidence is used to represent a probability that the target picture belongs to one of a preset plurality of picture categories. Each picture can be classified according to the number of individuals in the picture and a preset threshold value, and the pictures with the same number of individuals are classified into one type, wherein the threshold value is the maximum individual identification value which can be identified by the classifier under the condition of ensuring the identification precision. For example, when the preset threshold is 10, the pictures may be classified into 12 picture categories, where the 12 picture categories may include a picture category of 0 individuals, a picture category of 1 to 10 individuals, and a picture category of greater than 10 individuals, so that the target picture is classified by the classifier to obtain 12 confidence degrees corresponding to the target picture, and each confidence degree corresponds to one picture category.
By way of example, taking the preset area "subway entrance" as an example, the picture class of the subway entrance picture may be predicted by the MSE function in the classifier to obtain a plurality of probabilities corresponding to the subway entrance picture, where each probability corresponds to a picture class.
Step S102a2, determining a target picture category of the target picture according to the picture category corresponding to the highest confidence level in the plurality of confidence levels.
The method includes selecting a maximum probability of the probabilities, and taking a picture category corresponding to the maximum probability as a first picture category of the subway entrance picture.
Step S102a3, determining a target classification value of the target picture according to the picture category of the subway entrance picture and the target picture category, where the target classification value is used to represent the number of individuals in the target picture.
For example, according to the value corresponding to the first picture category, determining the number of individuals in the subway entrance picture, and taking the value of the number of individuals as a target classification value.
Illustratively, the step S102 may further include:
step S102b1, performing regression processing on the target picture by using a regressive device, so as to obtain a numerical value corresponding to the number of individuals in the target picture.
By way of example, taking the preset area "shopping mall" as an example, the number of individuals in the shopping mall picture can be predicted in the regressor through the cross entropy function, so as to obtain a specific numerical value of the number of individuals in the shopping mall picture.
Step S102b2, performing an integer processing on the numerical value to obtain an integer numerical value, and determining the integer numerical value as the target regression value.
And processing the specific numerical value of the individual quantity in the large-scale market picture to obtain an integer numerical value, namely rounding off the non-integer to obtain an integer value, and taking the integer value as the target regression value.
Step S104, judging whether the target classification value is larger than a preset classification value; outputting a flow identification result based on the target regression value when the target classification value is larger than a preset classification value; and outputting the flow identification result based on the target classification value when the target classification value is not greater than a preset classification value.
For example, the flow identification result may be the number of individuals in a preset area; for example, the number of passengers at a subway entrance, the number of visitors at a scenic spot entrance, the number of guests at a large-sized venue entrance, the number of vehicles at a parking lot entrance, and the like may be mentioned.
The preset classification value may be a threshold value, which may be a fixed value, configured according to the number of classes identifiable by the classifier; the number of the classes which can be identified by the classifier can be adjusted according to the identification effects (identification precision and identification efficiency) of the classifier and the regressor; for example, the predictive effect of the classifier is superior to the regressive when the number of individuals in the preset area is not more than 20, and the predictive effect of the regressive is superior to the classifier when the number of individuals in the preset area is more than 20, the value may be set to 20.
It will be appreciated that the number of identifiable categories of the classifier may be 22, and at this time, the maximum identifiable number of individuals of the classifier is 20, that is, the picture category with the number of individuals of 0 in one picture subtracted from the picture category with the number of more than 20 individuals in one picture is subtracted, so the number may be configured to be 20.
Judging whether the target classification value is larger than 20 by the classifier:
in the first case, when the target classification value is not more than 20, the target classification value is used as a flow identification result; it will be appreciated that when the target classification value is not greater than 20, the result obtained by the classifier is more accurate than the result obtained by the regressor, so the target classification value is used as the flow rate identification result.
Secondly, when the target classification value is larger than 20, the regression result is used as a flow identification result; it will be appreciated that when the target classification value is greater than 20, the result obtained by the regressor is more accurate than the result obtained by the classifier, so the regressed result is used as the flow identification result.
The flow identification method based on the classification regression double-domain model further comprises the following steps: the flow identification result is sent to target equipment, so that the target equipment executes corresponding operation according to the flow identification result; or generating a corresponding operation instruction according to the flow identification result so as to control the target equipment to execute corresponding operation through the operation instruction.
Illustratively, the target device may be an automated valve in a public location, such as: an automatic valve at the entrance of a parking lot, an automatic valve at the entrance of a subway, an automatic valve at the entrance of a scenic spot, an automatic valve at the entrance of a large-scale activity place, and the like. And controlling the opening operation and the closing operation of the automatic valve according to the flow identification result and the preset operation instruction of the automatic valve.
For example, the flow identification method based on the classification regression double-domain model may further include training steps S200 to S208 of the classifier:
step S200, an initial classifier is acquired.
The initial classifier may be, for example, an initial deep convolutional neural network model that may be used as a classifier.
In step S202, a plurality of initial classification pictures are acquired, and each initial classification picture is assigned a first number label indicating the number of individuals.
Illustratively, the plurality of initially classified pictures are obtained from an existing database; the first image acquisition device such as a camera can acquire a plurality of images, and then a first number label which represents the number of individuals is allocated to each image so as to obtain a plurality of initial classification images. For example, the plurality of pictures may be images shot by a certain camera, or may be video frames, where the camera may be each monitoring camera in a subway station, each monitoring camera at a scenic spot entrance, each monitoring camera at an underground parking garage, each monitoring camera at a large-scale activity place entrance, and the like.
Step S204, dividing the plurality of initial classified pictures into n+2 picture categories according to the first number of labels corresponding to each initial classified picture, and configuring corresponding picture category labels, where N is a positive integer, and the n+2 picture categories include a picture category with an individual number of 0, a picture category with an individual number of N, and a picture category with an individual number of greater than N.
It should be noted that, when the number of individuals in the preset area is not greater than N, the prediction effect of the classifier is better than that of the regressive, and when the number of individuals in the preset area is greater than N, the prediction effect of the regressive is better than that of the classifier, that is, the classifier can maximally identify the picture category as N in the preset area.
Step S206, configuring a plurality of classification training samples according to the initial classification pictures corresponding to the picture class labels.
The initial classification picture carrying the picture category label is taken as a classification training sample, so that a plurality of classification training samples are obtained.
Step S208, training the initial classifier through the plurality of classification training samples to obtain a trained classifier.
Illustratively, a deep convolutional neural network model that may be used as a classifier is trained by the plurality of classification training samples until the model converges to arrive at the classifier.
Illustratively, the initial classifier has a cross entropy function as a classification objective function, the cross entropy function:
wherein N represents the number of training samples, w is a network parameter, p and q represent the prediction probability and the training experience probability respectively, y represents the training label value,representing model predictions.
For example, the flow identification method based on the classification regression double-domain model may further include training steps S300 to S306 of the regressor:
step S300, an initial regressor is obtained.
The initial regressor may be, for example, an initial deep convolutional neural network model that may be used as a regressor.
In step S302, a plurality of initial regression pictures are obtained, and each initial regression picture is assigned a second number label indicating the number of individuals.
Illustratively, the plurality of initial regression pictures are obtained from an existing database; the method can also obtain a plurality of pictures through an image acquisition device such as a camera, and then allocate a first number label for representing the number of individuals to each picture so as to obtain a plurality of initial regression pictures. For example, the plurality of pictures may be images shot by a certain camera, or may be video frames, where the camera may be each monitoring camera in a subway station, each monitoring camera at a scenic spot entrance, each monitoring camera at an underground parking garage, each monitoring camera at a large-scale activity place entrance, and the like.
Step S304, configuring a plurality of regression training samples according to the initial regression pictures corresponding to the second number of labels.
The initial regression pictures carrying the second number of labels are taken as regression training samples, so that a plurality of regression training samples are obtained.
And step S306, training the initial regressive device through the plurality of regression training samples to obtain a trained regressive device.
Illustratively, a deep convolutional neural network model that can be used as a regressor is trained by the plurality of regression training samples until the model converges to obtain the regressor.
Illustratively, the initial regressor has an MSE (Mean Square Error mean square error) function as a regression objective function, the MSE function:
wherein n represents the number of training samples, y i Representing the actual value of the value,representing the predicted value.
Example two
Fig. 2 is a schematic program module diagram of a flow identification system based on a classification regression dual-domain model according to a second embodiment of the present invention. The classification regression dual-domain model-based flow identification system 20 may include or be partitioned into one or more program modules that are stored in a storage medium and executed by one or more processors to perform the present invention and may implement the classification regression dual-domain model-based flow identification method described above. Program modules in accordance with embodiments of the present invention refer to a series of computer program instruction segments capable of performing particular functions, and are more suited to describing the execution of the classification regression dual-domain model-based flow identification system 20 in a storage medium than the program itself. The following description will specifically describe functions of each program module of the present embodiment:
the acquisition module 200 is configured to acquire a target picture in a preset area.
The input module 202 is configured to input the target picture into a classification regression dual-domain model, and output a target classification value and a target regression value of the target picture through the classification regression dual-domain model, where the classification dual-domain model includes a pre-trained classifier and a pre-trained regressor, and the classifier and the regressor are both configured to identify traffic in the target picture.
Illustratively, the input module 202 is further configured to: classifying the target picture through the classifier to obtain a plurality of confidence degrees corresponding to the target picture, wherein each confidence degree is used for representing the probability that the target picture belongs to one of a plurality of picture categories; determining a target picture category of the target picture according to the picture category corresponding to the highest confidence coefficient in the plurality of confidence coefficients; and determining a target classification value of the target picture according to the target picture category, wherein the target classification value is used for representing the number of individuals in the target picture.
Illustratively, the input module 202 is further configured to: carrying out regression processing on the target picture through a regressing device to obtain a numerical value corresponding to the number of individuals in the target picture; and performing integer processing on the numerical value to obtain an integer numerical value, and determining the integer numerical value as the target regression value.
The judging module 204 is configured to judge whether the target classification value is greater than a preset classification value.
The output module 206 is configured to output a flow identification result based on the target regression value when the target classification value is greater than a preset classification value; and outputting the flow identification result based on the target classification value when the target classification value is not greater than a preset classification value. And the method is used for judging whether the classification value is larger than a preset classification value.
Illustratively, the flow identification system 20 based on the classification regression dual-domain model further includes: a transmitting module 208, where the transmitting module 208 is configured to: the flow identification result is sent to target equipment, so that the target equipment executes corresponding operation according to the flow identification result; or generating a corresponding operation instruction according to the flow identification result so as to control the target equipment to execute corresponding operation through the operation instruction.
Illustratively, the flow identification system 20 based on the classification regression dual-domain model further includes: a training module 210, the training module 210 being configured to: acquiring an initial classifier; acquiring a plurality of initial classified pictures, wherein each initial classified picture is allocated with a first quantity label for representing the quantity of individuals; dividing the plurality of initial classified pictures into n+2 picture categories according to the first number labels corresponding to each initial classified picture, and configuring corresponding picture category labels, wherein N is a positive integer, and the n+2 picture categories comprise a picture category with the number of 0 individuals, a picture category with the number of N individuals, and a picture category with the number of one individual being greater than N; configuring a plurality of classification training samples according to the initial classification pictures corresponding to the picture category labels; and training the initial classifier through the plurality of classification training samples to obtain a trained classifier.
Illustratively, the training module 210 is further configured to: acquiring an initial regressive device; acquiring a plurality of initial regression pictures, wherein each initial regression picture is allocated with a second number label for representing the number of individuals; configuring a plurality of regression training samples according to the initial regression pictures corresponding to the second quantity labels; and training the initial regressive device through the multiple regression training samples to obtain a trained regressive device.
Example III
Referring to fig. 3, a hardware architecture diagram of a computer device according to a third embodiment of the present invention is shown. In this embodiment, the computer device 2 is a device capable of automatically performing numerical calculation and/or information processing according to a preset or stored instruction. The computer device 2 may be a rack server, a blade server, a tower server, or a rack server (including a stand-alone server, or a server cluster made up of multiple servers), or the like. As shown, the computer device 2 includes, but is not limited to, at least a memory 21, a processor 22, a network interface 23, and a traffic recognition system 20 based on a classification regression two-domain model that are communicatively coupled to each other via a system bus.
In this embodiment, the memory 21 includes at least one type of computer-readable storage medium including flash memory, a hard disk, a multimedia card, a card memory (e.g., SD or DX memory, etc.), a Random Access Memory (RAM), a Static Random Access Memory (SRAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a programmable read-only memory (PROM), a magnetic memory, a magnetic disk, an optical disk, and the like. In some embodiments, the memory 21 may be an internal storage unit of the computer device 2, such as a hard disk or a memory of the computer device 2. In other embodiments, the memory 21 may also be an external storage device of the computer device 2, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card) or the like, which are provided on the computer device 2. Of course, the memory 21 may also include both internal storage units of the computer device 2 and external storage devices. In this embodiment, the memory 21 is generally used to store an operating system and various application software installed on the computer device 2, for example, the program code of the flow identification system 20 based on the classification regression dual-domain model in the second embodiment. Further, the memory 21 may be used to temporarily store various types of data that have been output or are to be output.
The processor 22 may be a central processing unit (Central Processing Unit, CPU), controller, microcontroller, microprocessor, or other data processing chip in some embodiments. The processor 22 is typically used to control the overall operation of the computer device 2. In this embodiment, the processor 22 is configured to execute the program code or the processing data stored in the memory 21, for example, execute the flow identification system 20 based on the classification regression double-domain model, so as to implement the flow identification method based on the classification regression double-domain model of the first embodiment.
The network interface 23 may comprise a wireless network interface or a wired network interface, which network interface 23 is typically used for establishing a communication connection between the computer apparatus 2 and other electronic devices. For example, the network interface 23 is used to connect the computer device 2 to an external terminal through a network, establish a data transmission channel and a communication connection between the computer device 2 and the external terminal, and the like. The network may be an Intranet (Intranet), the Internet (Internet), a global system for mobile communications (Global System of Mobile communication, GSM), wideband code division multiple access (Wideband Code Division Multiple Access, WCDMA), a 4G network, a 5G network, bluetooth (Bluetooth), wi-Fi, or other wireless or wired network.
It is noted that fig. 3 only shows a computer device 2 having components 20-23, but it is understood that not all of the illustrated components are required to be implemented, and that more or fewer components may alternatively be implemented.
In the present embodiment, the flow identification system 20 based on the classification regression two-domain model stored in the memory 21 may also be divided into one or more program modules, which are stored in the memory 21 and executed by one or more processors (the processor 22 in the present embodiment) to complete the present invention.
For example, fig. 2 shows a schematic diagram of a program module of the flow identification system 20 according to the second embodiment of the present invention, where the flow identification system 20 based on the classification regression dual-domain model may be divided into an acquisition module 200, an input module 202, a judgment module 204, an output module 206, a transmission module 208, and a training module 210. The program modules referred to herein are meant to be a series of computer program instruction segments capable of performing a specific function, more suitably than a program, describing the execution of the flow identification system 20 in the computer device 2 based on the classification regression double domain model. The specific functions of the program modules 200-210 are described in detail in the second embodiment, and are not described herein.
Example IV
The present embodiment also provides a computer-readable storage medium such as a flash memory, a hard disk, a multimedia card, a card-type memory (e.g., SD or DX memory, etc.), a Random Access Memory (RAM), a Static Random Access Memory (SRAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a programmable read-only memory (PROM), a magnetic memory, a magnetic disk, an optical disk, a server, an App application store, etc., on which a computer program is stored, which when executed by a processor, performs the corresponding functions. The computer readable storage medium of the present embodiment is used for the traffic recognition system 20 based on the classification regression double-domain model, and when executed by the processor, implements the traffic recognition method based on the classification regression double-domain model of the first embodiment.
The foregoing embodiment numbers of the present invention are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.
From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment.
The foregoing description is only of the preferred embodiments of the present invention, and is not intended to limit the scope of the invention, but rather is intended to cover any equivalents of the structures or equivalent processes disclosed herein or in the alternative, which may be employed directly or indirectly in other related arts.
Claims (10)
1. A traffic identification method based on a classification regression double-domain model, the method comprising:
collecting a target picture in a preset area;
inputting the target picture into a classification regression double-domain model, and outputting a target classification value and a target regression value of the target picture through the classification regression double-domain model, wherein the classification regression double-domain model comprises a pre-trained classifier and a pre-trained regressor, and the classifier and the regressor are used for identifying the flow in the target picture;
judging whether the target classification value is larger than a preset classification value or not;
outputting a flow identification result based on the target regression value when the target classification value is larger than a preset classification value;
and outputting the flow identification result based on the target classification value when the target classification value is not greater than a preset classification value.
2. The method for identifying traffic based on a classification regression double-domain model according to claim 1, wherein the step of outputting the target classification value and the target regression value of the target picture through the classification regression double-domain model comprises:
executing steps a 1-a 3 through the classifier to obtain a target classification value of the target picture:
a1: classifying the target picture to obtain a plurality of confidence degrees corresponding to the target picture, wherein each confidence degree is used for representing the probability that the target picture belongs to one of a plurality of picture categories;
a2: determining a target picture category of the target picture according to the picture category corresponding to the highest confidence coefficient in the plurality of confidence coefficients;
a3: and determining a target classification value of the target picture according to the target picture category, wherein the target classification value is used for representing the number of individuals in the target picture.
3. The method for identifying traffic based on a classification regression double-domain model according to claim 1, wherein the step of outputting the target classification value and the target regression value of the target picture through the classification regression double-domain model comprises:
executing steps b 1-b 2 by the regressor to obtain a target regression value of the target picture:
b1: carrying out regression processing on the target picture to obtain a numerical value corresponding to the number of individuals in the target picture;
b2: and performing integer processing on the numerical value to obtain an integer numerical value, and determining the integer numerical value as the target regression value.
4. The classification regression dual-domain model based flow identification method of claim 1, further comprising:
the flow identification result is sent to target equipment, so that the target equipment executes corresponding operation according to the flow identification result; or (b)
And generating a corresponding operation instruction according to the flow identification result so as to control the target equipment to execute corresponding operation through the operation instruction.
5. The method for traffic identification based on a classification regression double-domain model according to claim 1, further comprising the step of training the classifier:
acquiring an initial classifier;
acquiring a plurality of initial classified pictures, wherein each initial classified picture is allocated with a first quantity label for representing the quantity of individuals;
dividing the plurality of initial classified pictures into n+2 picture categories according to the first number labels corresponding to each initial classified picture, and configuring corresponding picture category labels, wherein N is a positive integer, and the n+2 picture categories comprise a picture category with the number of 0 individuals, a picture category with the number of N individuals, and a picture category with the number of one individual being greater than N;
configuring a plurality of classification training samples according to the initial classification pictures corresponding to the picture category labels; a kind of electronic device with high-pressure air-conditioning system
Training the initial classifier through the plurality of classification training samples to obtain the classifier.
6. The method for identifying traffic based on a classification regression double-domain model according to claim 1, further comprising the step of training the regressor:
acquiring an initial regressive device;
acquiring a plurality of initial regression pictures, wherein each initial regression picture is allocated with a second number label for representing the number of individuals;
configuring a plurality of regression training samples according to the initial regression pictures corresponding to the second quantity labels; a kind of electronic device with high-pressure air-conditioning system
And training the initial regressive device through the multiple regression training samples to obtain the regressive device.
7. A classification regression dual-domain model-based flow identification system, comprising:
the acquisition module is used for acquiring a target picture in a preset area;
the input module is used for inputting the target picture into a classification regression double-domain model, and outputting a target classification value and a target regression value of the target picture through the classification regression double-domain model, wherein the classification regression double-domain model comprises a pre-trained classifier and a pre-trained regressor, and the classifier and the regressor are used for identifying the flow in the target picture;
the judging module is used for judging whether the target classification value is larger than a preset classification value or not;
the output module is used for outputting a flow identification result based on the target regression value when the target classification value is larger than a preset classification value; and outputting the flow identification result based on the target classification value when the target classification value is not greater than a preset classification value.
8. The classification regression dual-domain model based flow identification system of claim 7 wherein the input module is further configured to:
classifying the target picture through the classifier to obtain a plurality of confidence degrees corresponding to the target picture, wherein each confidence degree is used for representing the probability that the target picture belongs to one of a plurality of picture categories;
determining a target picture category of the target picture according to the picture category corresponding to the highest confidence coefficient in the plurality of confidence coefficients;
and determining a target classification value of the target picture according to the target picture category, wherein the target classification value is used for representing the number of individuals in the target picture.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and running on the processor, characterized in that the computer program when executed by the processor implements the steps of the classification regression double domain model based flow identification method of any one of claims 1 to 6.
10. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored therein a computer program, which is executed by at least one processor, to cause the at least one processor to perform the steps of the classification regression double-domain model-based flow identification method according to any one of claims 1 to 6.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010090623.8A CN111310652B (en) | 2020-02-13 | 2020-02-13 | Flow identification method and system based on classification regression double-domain model |
| PCT/CN2020/093034 WO2021159634A1 (en) | 2020-02-13 | 2020-05-28 | Classification and regression dual-domain model based traffic identification method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010090623.8A CN111310652B (en) | 2020-02-13 | 2020-02-13 | Flow identification method and system based on classification regression double-domain model |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111310652A CN111310652A (en) | 2020-06-19 |
| CN111310652B true CN111310652B (en) | 2023-07-25 |
Family
ID=71161561
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010090623.8A Active CN111310652B (en) | 2020-02-13 | 2020-02-13 | Flow identification method and system based on classification regression double-domain model |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN111310652B (en) |
| WO (1) | WO2021159634A1 (en) |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104134078A (en) * | 2014-07-22 | 2014-11-05 | 华中科技大学 | Automatic selection method for classifiers in people flow counting system |
| CN107527024A (en) * | 2017-08-08 | 2017-12-29 | 北京小米移动软件有限公司 | Face face value appraisal procedure and device |
| CN109241914A (en) * | 2018-09-11 | 2019-01-18 | 广州广电银通金融电子科技有限公司 | A kind of Small object pedestrian detection method under complex scene |
| CN110059581A (en) * | 2019-03-28 | 2019-07-26 | 常熟理工学院 | People counting method based on depth information of scene |
| CN110096979A (en) * | 2019-04-19 | 2019-08-06 | 佳都新太科技股份有限公司 | Construction method, crowd density estimation method, device, equipment and the medium of model |
| CN110708285A (en) * | 2019-08-30 | 2020-01-17 | 中国平安人寿保险股份有限公司 | Flow monitoring method, device, medium and electronic equipment |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20160033800A (en) * | 2014-09-17 | 2016-03-29 | 삼성전자주식회사 | Method for counting person and counting apparatus |
| CN106599856A (en) * | 2016-12-19 | 2017-04-26 | 四川长虹电器股份有限公司 | Combined face detection, positioning and identification method |
| US10949475B2 (en) * | 2018-05-14 | 2021-03-16 | Ebay Inc. | Search system for providing web crawling query prioritization based on classification operation performance |
-
2020
- 2020-02-13 CN CN202010090623.8A patent/CN111310652B/en active Active
- 2020-05-28 WO PCT/CN2020/093034 patent/WO2021159634A1/en active Application Filing
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104134078A (en) * | 2014-07-22 | 2014-11-05 | 华中科技大学 | Automatic selection method for classifiers in people flow counting system |
| CN107527024A (en) * | 2017-08-08 | 2017-12-29 | 北京小米移动软件有限公司 | Face face value appraisal procedure and device |
| CN109241914A (en) * | 2018-09-11 | 2019-01-18 | 广州广电银通金融电子科技有限公司 | A kind of Small object pedestrian detection method under complex scene |
| CN110059581A (en) * | 2019-03-28 | 2019-07-26 | 常熟理工学院 | People counting method based on depth information of scene |
| CN110096979A (en) * | 2019-04-19 | 2019-08-06 | 佳都新太科技股份有限公司 | Construction method, crowd density estimation method, device, equipment and the medium of model |
| CN110708285A (en) * | 2019-08-30 | 2020-01-17 | 中国平安人寿保险股份有限公司 | Flow monitoring method, device, medium and electronic equipment |
Non-Patent Citations (1)
| Title |
|---|
| 基于SVM的先分类再回归方法及其在产量预测中的应用;董毅等;计算机应用(第09期);第12-45页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| WO2021159634A1 (en) | 2021-08-19 |
| CN111310652A (en) | 2020-06-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11868926B2 (en) | Systems and methods for managing public place in smart city | |
| CN106529711B (en) | User behavior prediction method and device | |
| EP4012606A1 (en) | Method for estimating and presenting passenger flow, system, and computer readable storage medium | |
| JP6939911B2 (en) | Methods and devices for adaptive vehicle control | |
| CN110245704B (en) | Service processing method and device, storage medium and electronic equipment | |
| CN105825350A (en) | Video analysis-based intelligent tourism early warning decision-making system and use method thereof | |
| CN104239386A (en) | Method and system for prioritizion of facial recognition matches | |
| CN113095434A (en) | Target detection method and device, electronic equipment and storage medium | |
| KR102333143B1 (en) | System for providing people counting service | |
| Hakim et al. | Implementation of an image processing based smart parking system using Haar-Cascade method | |
| CN109584579B (en) | Traffic signal lamp control method based on face recognition and computer equipment | |
| CN112233428B (en) | Traffic flow prediction method, device, storage medium and equipment | |
| CN110602446A (en) | Garbage recovery reminding method and system and storage medium | |
| CN112863195B (en) | Vehicle state determination method and device | |
| CN110728211A (en) | Subway security check queuing guide method and device and storage medium | |
| CN114049658A (en) | Floating population management method and device based on face recognition, computer equipment and storage medium | |
| CN111310652B (en) | Flow identification method and system based on classification regression double-domain model | |
| CN108447273A (en) | A kind of remaining parking stall prediction technique based on intelligent parking system | |
| CN113052048A (en) | Traffic incident detection method and device, road side equipment and cloud control platform | |
| CN110046535B (en) | Intelligent travel time prediction system, method and storage medium based on machine learning | |
| CN116486332A (en) | Passenger flow monitoring method, device, equipment and storage medium | |
| CN113593256B (en) | Unmanned aerial vehicle intelligent driving-away control method and system based on city management and cloud platform | |
| CN116311015A (en) | Road scene recognition method, device, server, storage medium and program product | |
| CN115147752A (en) | Video analysis method and device and computer equipment | |
| CN113592902A (en) | Target tracking method and device, computer equipment and storage medium |
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 |