Mobile source nitrogen oxide emission inventory compilation method, electronic device and storage medium

Abstract

The present application discloses a method for compiling a mobile source nitrogen oxide emission inventory, an electronic device, and a storage medium, which can be applied to the field of air quality numerical simulation processing technology. The present application performs vehicle type identification, vehicle speed measurement, and license plate number identification on mobile sources in a preset area, and then calculates the instantaneous nitrogen oxide emissions of the mobile source based on the relationship between the license plate number and the preset database, combined with the vehicle type identification result or the vehicle speed measurement result, and then calculates the total nitrogen oxide emissions of the mobile source in each road section based on the instantaneous nitrogen oxide emissions, and then processes the road sections in a grid-like manner to obtain a mobile source nitrogen oxide emission inventory, so that the mobile source nitrogen oxide emission inventory can provide data support for real-time dynamic measurement of atmospheric nitrogen oxides and air quality simulation analysis, thereby improving the accuracy of the measurement results and air quality simulation analysis results.

Description

Technical Field

[0001] The present application relates to the technical field of air quality numerical simulation processing, and in particular to a method for compiling a mobile source nitrogen oxide emission inventory, an electronic device and a storage medium. Background Art

[0002] When simulating air quality models (such as CMAQ, CAMx, etc.), the air quality models are simulated based on the dynamic emissions of each pollution source. Existing mobile sources are mostly calculated using annual static total emissions, which makes it difficult to support real-time dynamic measurement of atmospheric nitrogen oxides and air quality simulation analysis.

[0003] In summary, the technical problems existing in the relevant technologies need to be improved. Summary of the invention

[0004] The main purpose of the embodiments of the present application is to propose a method for compiling a mobile source nitrogen oxide emission inventory, an electronic device and a storage medium, which can provide real-time and effective data for real-time dynamic measurement of atmospheric nitrogen oxides and air quality simulation analysis.

[0005] To achieve the above objectives, one aspect of an embodiment of the present application provides a method for compiling a mobile source nitrogen oxide emission inventory, the method comprising:

[0006] Carry out vehicle type recognition on mobile sources in a preset area to obtain vehicle type recognition results;

[0007] Measuring the speed of all mobile sources in the preset area to obtain the real-time speed of each mobile source;

[0008] Identify the license plate number of the mobile source in the preset area;

[0009] Determining that the license plate number is in a preset database, and calculating the instantaneous nitrogen oxide emission of the mobile source corresponding to the license plate number according to the real-time vehicle speed as a first instantaneous nitrogen oxide emission;

[0010] Determining that the license plate number is not located in the preset database, and calculating the instantaneous nitrogen oxide emissions of the mobile source corresponding to the license plate number as the second instantaneous nitrogen oxide emissions according to the vehicle type recognition result and the real-time vehicle speed;

[0011] Calculating the total amount of nitrogen oxide emissions of each type of mobile source on each road section according to the first instantaneous nitrogen oxide emissions or the second instantaneous nitrogen oxide emissions;

[0012] According to the gridded processing of the road sections with the total amount of nitrogen oxide emissions, a mobile source nitrogen oxide emission inventory is obtained.

[0013] In some embodiments, the step of performing vehicle type recognition on the mobile source in the preset area to obtain the vehicle type recognition result includes:

[0014] Get the frame-by-frame images of the traffic camera video stream;

[0015] Extracting a shared feature map of the frame-by-frame image of the traffic camera video stream;

[0016] Sliding the shared feature map to obtain a plurality of anchor frames;

[0017] Predict the foreground probability and offset of all anchor boxes through parallel convolution kernels;

[0018] Determine a regression correction frame to be selected from all the anchor frames according to the offset and the foreground probability;

[0019] Determine the target regression correction frame according to the intersection-over-intersection ratio of the regression correction frame to be selected and the real target frame;

[0020] Extracting regional features from the shared feature map according to the target regression correction frame;

[0021] The vehicle type of each mobile source is determined according to the regional characteristics to obtain a vehicle type recognition result.

[0022] In some embodiments, measuring the speed of all mobile sources in the preset area to obtain the real-time speed of each mobile source includes:

[0023] Performing target detection in the preset area to determine the moving source in the image;

[0024] Generate an image movement trajectory of the moving source according to multiple frames of continuous corner points of the moving source in the image;

[0025] Calculate the actual moving distance of the moving source according to the image moving trajectory;

[0026] The real-time vehicle speed of the mobile source is calculated according to the actual moving distance.

[0027] In some embodiments, identifying the license plate number of the mobile source in the preset area includes:

[0028] Acquire a moving source image within the preset area;

[0029] performing grayscale processing and image enhancement processing on the moving source image in sequence;

[0030] Performing image edge detection on the mobile source image after image enhancement processing to obtain license plate feature information of the mobile source image;

[0031] Determine the location in the moving source image according to the license plate feature information to obtain a license plate image;

[0032] Performing tilt correction on the license plate image;

[0033] Perform character segmentation on the tilt-corrected license plate image to obtain a character segmentation result;

[0034] The license plate number of the mobile source is determined according to the character segmentation result.

[0035] In some embodiments, the calculation formula of the first instantaneous nitrogen oxide emission is as follows:

[0036]

[0037]

[0038] in, represents the average emission factor of nitrogen oxides during the calculation period, represents the first instantaneous nitrogen oxide emission, represents the instantaneous concentration of the nitrogen oxides, represents the instantaneous exhaust mass flow rate, v represents the real-time vehicle speed, z represents the frequency of data acquisition, n represents the number of tests, Represents the instantaneous molar mass of nitrogen oxides.

[0039] In some embodiments, the calculating the instantaneous nitrogen oxide emissions of the mobile source corresponding to the license plate number as the second instantaneous nitrogen oxide emissions according to the vehicle type recognition result and the real-time vehicle speed includes:

[0040] Determining a vehicle model base emission factor according to the vehicle model identification result;

[0041] determining a vehicle speed correction factor according to the real-time vehicle speed;

[0042] The instantaneous nitrogen oxide emissions of the mobile source corresponding to the license plate number are calculated according to the vehicle model base emission factor and the vehicle speed correction factor as the second instantaneous nitrogen oxide emissions.

[0043] In some embodiments, the calculating the total amount of nitrogen oxide emissions of each type of mobile source on each road section according to the first instantaneous nitrogen oxide emissions or the second instantaneous nitrogen oxide emissions includes:

[0044] Linearly generalizing the road to obtain a road generalization map, wherein the road generalization map includes road location information and traffic camera location information;

[0045] Performing road node conversion on the road generalization graph to obtain a road segment graph;

[0046] According to the road position information and the traffic camera position information, a relationship correspondence matrix between roads and cameras in the road segment graph is established;

[0047] determining a corresponding road section in the relationship correspondence matrix according to a camera position corresponding to the first instantaneous nitrogen oxide emission or the second instantaneous nitrogen oxide emission;

[0048] Calculate the total amount of nitrogen oxide emissions from each type of mobile source on the road section.

[0049] In some embodiments, the gridding of the road sections according to the total amount of nitrogen oxide emissions to obtain a mobile source nitrogen oxide emission inventory includes:

[0050] Performing equidistant grid cutting processing on the area to which the road section belongs;

[0051] Calculate the nitrogen oxide emission component in each grid after grid cutting according to the total nitrogen oxide emission;

[0052] The grids are marked according to the nitrogen oxide emission components to obtain a mobile source nitrogen oxide emission inventory.

[0053] To achieve the above object, another aspect of an embodiment of the present application provides an electronic device, including:

[0054] at least one processor;

[0055] at least one memory for storing at least one program;

[0056] When the at least one program is executed by the at least one processor, the at least one processor implements the above method.

[0057] To achieve the above objective, another aspect of an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and the computer program implements the above method when executed by a processor.

[0058] The embodiments of the present application include at least the following beneficial effects: The present application provides a method for compiling a mobile source nitrogen oxide emission inventory, an electronic device and a storage medium. The scheme performs vehicle type identification, vehicle speed measurement and license plate number identification on mobile sources in a preset area, and then calculates the instantaneous nitrogen oxide emissions of the mobile source based on the relationship between the license plate number and a preset database, combined with the vehicle type identification result or the vehicle speed measurement result. Then, the total nitrogen oxide emissions of the mobile source in each road section are calculated based on the instantaneous nitrogen oxide emissions, and the road sections are gridded to obtain a mobile source nitrogen oxide emission inventory, so that the mobile source nitrogen oxide emission inventory can provide data support for real-time dynamic measurement of atmospheric nitrogen oxides and air quality simulation analysis, thereby improving the accuracy of the measurement results and air quality simulation analysis results. BRIEF DESCRIPTION OF THE DRAWINGS

[0059] Figure 1 It is a flow chart of a method for compiling a mobile source nitrogen oxide emission inventory provided in an embodiment of the present application;

[0060] Figure 2 It is a flow chart of a method for compiling a linear nitrogen oxide emission inventory of a motor vehicle based on a road network provided in an embodiment of the present application;

[0061] Figure 3 It is a flow chart of a method for compiling a gridded motor vehicle nitrogen oxide emission inventory provided in an embodiment of the present application;

[0062] Figure 4 It is a schematic diagram of the hardware structure of the electronic device provided in the embodiment of the present application. DETAILED DESCRIPTION

[0063] In order to make the purpose, technical solutions and advantages of the present application clearer, the present application is further described in detail below in conjunction with the accompanying drawings and examples. It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the embodiments of the present application. They are only examples of devices and methods consistent with some aspects of the embodiments of the present application as detailed in the attached claims.

[0064] It is understood that the terms "first", "second", etc. used in this application can be used to describe various concepts in this article, but unless otherwise specified, these concepts are not limited by these terms. These terms are only used to distinguish one concept from another. For example, without departing from the scope of the embodiments of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Depending on the context, the words "if" and "if" as used herein can be interpreted as "at the time of" or "when" or "in response to determination".

[0065] The terms "at least one", "multiple", "each", "any", etc. used in this application, at least one includes one, two or more, multiple includes two or more, each refers to each of the corresponding multiple, and any refers to any one of the multiple.

[0066] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein are only for the purpose of describing the embodiments of this application and are not intended to limit this application.

[0067] Before describing the embodiments of the present application in detail, some nouns and terms involved in the embodiments of the present application are first described. The nouns and terms involved in the embodiments of the present application are subject to the following explanations.

[0068] NOx: Nitrogen oxides.

[0069] CMAQ: The full English name is Community Multiscale Air Quality Modeling System, which is interpreted as community multi-scale air quality simulation system in Chinese.

[0070] CAMx: The full English name is Comprehensive Air quality Model with extensions, which is interpreted as the extended comprehensive air quality model in Chinese.

[0071] Faster R-CNN: The full English name is Faster Region-based Convolutional Neural Network, which is interpreted in Chinese as a deep learning algorithm for target detection.

[0072] RPN: The full name of RPN is Region Proposal Network, a neural network component used for target detection. Its main function is to generate candidate regions in the image that may contain targets for subsequent target classification and bounding box regression.

[0073] VGG16: A convolutional neural network (CNN) architecture proposed by Simonyan and Zisserman in 2014. It is a classic model proposed by the Visual Geometry Group (VGG) in the ImageNet image recognition challenge. It has a deeper network structure and is widely used in tasks such as image classification, object detection, and feature extraction.

[0074] YOLO-V5: The latest version of the YOLO (You Only Look Once) series of algorithms, a deep learning algorithm for object detection, with the main goal of achieving faster and more accurate object detection with a smaller model size.

[0075] OpenCV: The full name of Open Source Computer Vision Library, an open source computer vision library, provides a wealth of image processing and computer vision algorithms.

[0076] Hough Line Transform: Hough Line Transform is an image processing technique used to detect straight lines in an image. It was proposed by Hough in 1962 and later improved and widely used in the fields of computer vision and image processing.

[0077] RGB (Red, Green, Blue): RGB is one of the most common color models and is also the color representation method used in electronic devices (such as monitors and cameras).

[0078] YUV: YUV is a color model that separates brightness (Y) and chrominance (U, V).

[0079] CMY (Cyan, Magenta, Yellow): The CMY model is a color model used for printing colors.

[0080] HSI (Hue, Saturation, Intensity): The HSI model is a color model that decomposes color into three components: hue (H), saturation (S), and intensity (I).

[0081] Sobel operator: A commonly used edge detection operator used to detect edges in images. It is a convolution-based method that uses a set of filters to calculate the gradient magnitude and direction of each pixel in the image to find the edges in the image.

[0082] Douglas-Peucker algorithm: A commonly used line segment simplification algorithm used to simplify curves or polylines (such as roads in maps) into relatively few line segments while preserving the key feature points of the original shape as much as possible.

[0083] Ramer-Douglas-Peucker algorithm: also known as Ramer or RDP algorithm, is a line segment simplification algorithm used to simplify a curve or polyline into relatively few line segments while preserving the key feature points of the original shape as much as possible.

[0084] In related technologies, when simulating air quality models (such as CMAQ, CAMx, etc.), the dynamic emission measurement of mobile sources such as motor vehicles has always been a bottleneck problem. At present, mobile sources, mainly motor vehicles, mostly use annual static emission total measurement, which is difficult to support real-time dynamic measurement of atmospheric nitrogen oxides and air quality simulation analysis.

[0085] One of the major challenges in calculating dynamic emissions from mobile sources is the lack of effective dynamic traffic data on the road network. The calculation is based on factors such as dynamic traffic volume, road length, and emission factors. Although the calculation method of single-vehicle emission factors has been studied in depth and has effective calculation methods, due to the limitations of monitoring technology, it is difficult to obtain the full sample traffic volume of the road network through actual collection, making it difficult to calculate dynamic emissions from the road network.

[0086] In view of this, a method for compiling a mobile source nitrogen oxide emission inventory is provided in an embodiment of the present application. The method performs vehicle type identification, vehicle speed measurement and license plate number identification on mobile sources in a preset area, and then calculates the instantaneous nitrogen oxide emissions of the mobile source based on the relationship between the license plate number and a preset database in combination with the vehicle type identification result or the vehicle speed measurement result. Then, the total nitrogen oxide emissions of the mobile source in each road section are calculated based on the instantaneous nitrogen oxide emissions, and the road sections are gridded to obtain a mobile source nitrogen oxide emission inventory. The mobile source nitrogen oxide emission inventory can provide data support for real-time dynamic measurement of atmospheric nitrogen oxides and air quality simulation analysis, thereby improving the accuracy of the measurement results and air quality simulation analysis results.

[0087] The method for compiling a mobile source nitrogen oxide emission inventory provided in the embodiment of the present application relates to the field of information technology. The method for compiling a mobile source nitrogen oxide emission inventory provided in the embodiment of the present application can be applied to a terminal, can be applied to a server, or can be software running in a terminal or a server. In some embodiments, the terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, and a vehicle-mounted terminal, etc., but is not limited to this; the server side can be configured as an independent physical server, or can be configured as a server cluster or distributed system composed of multiple physical servers, and can also be configured as a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms. The server can also be a node server in a blockchain network; the software can be an application that implements the method for compiling a mobile source nitrogen oxide emission inventory, etc., but is not limited to the above forms.

[0088] The present application can be used in many general or special computer system environments or configurations. For example: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, etc. The present application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. The present application can also be practiced in distributed computing environments, in which tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules can be located in local and remote computer storage media including storage devices.

[0089] Figure 1 is an optional flowchart of the method for compiling a mobile source nitrogen oxide emission inventory provided in an embodiment of the present application. Figure 1 The method may include but is not limited to steps S101 to S106.

[0090] Step S110, performing vehicle type recognition on the mobile source in the preset area to obtain a vehicle type recognition result;

[0091] Step S120, measuring the speed of all mobile sources in the preset area to obtain the real-time speed of each mobile source;

[0092] Step S130, identifying the license plate number of the mobile source in the preset area;

[0093] Step S140, determining that the license plate number is in a preset database, and calculating the instantaneous nitrogen oxide emissions of the mobile source corresponding to the license plate number according to the real-time vehicle speed as the first instantaneous nitrogen oxide emissions;

[0094] Step S150, determining that the license plate number is not in the preset database, and calculating the instantaneous nitrogen oxide emissions of the mobile source corresponding to the license plate number according to the vehicle type recognition result and the real-time vehicle speed as the second instantaneous nitrogen oxide emissions;

[0095] Step S160, calculating the total amount of nitrogen oxide emissions of each type of mobile source on each road section according to the first instantaneous nitrogen oxide emissions or the second instantaneous nitrogen oxide emissions;

[0096] Step S170: Grid-process the road sections according to the total amount of nitrogen oxide emissions to obtain a mobile source nitrogen oxide emission inventory.

[0097] In an embodiment of the present application, when the vehicle type of a mobile source in a preset area is identified, after acquiring a frame-by-frame image of a traffic camera video stream, a shared feature map of the frame-by-frame image of the traffic camera video stream is extracted, and then the shared feature map is subjected to sliding processing to obtain a plurality of anchor frames; the foreground probability and offset of all anchor frames are predicted by parallel convolution kernels; a regression correction frame to be selected is determined from all the anchor frames according to the offset and the foreground probability; a target regression correction frame is determined according to the intersection-and-union ratio of the regression correction frame to be selected and the true target frame; regional features are extracted from the shared feature map according to the target regression correction frame; finally, the vehicle type of each mobile source is determined according to the regional features, and a vehicle type identification result is obtained.

[0098] It is understandable that the vehicle type recognition in this embodiment can be realized by the Faster R-CNN algorithm. Specifically, the Faster R-CNN algorithm generates candidate frames through the RPN network in the first stage, and performs final classification and regression through ROI Pooling and the fully connected layer in the second stage, integrating the candidate region, classification regression, feature extraction and other modules into a single convolutional network, thereby realizing the traffic flow statistics of different vehicle types (micro-light trucks, medium-sized trucks, heavy trucks, micro-passenger cars, medium-sized buses, large buses, taxis, buses, motorcycles).

[0099] The Faster R-CNN algorithm in this embodiment uses VGG16 as the basic convolutional network, which contains multiple convolutional layers and fully connected layers. By inputting the frame image of the traffic camera video stream, a feature map containing shared feature information is obtained. A 3*3 convolution kernel is used to slide on the shared feature map to generate 9 anchor boxes (Anchor boxes) for each position. These Anchor boxes have 3 ratios (1:1, 1:2, 2:1). Through two parallel 1*1 convolution kernels, the probabilities of the background and foreground of all Anchor boxes and the corresponding offsets are predicted respectively. The offset is used to adjust all Anchor boxes.

[0100] Limit the boundaries of all anchor boxes and use regression correction boxes to adjust them. The regression correction boxes are arranged from large to small according to the foreground probability. Use the non-maximum suppression algorithm (NMS) to calculate the intersection-over-union ratio and select a certain number of regression correction boxes (for example, 2000) as the regression correction boxes to be selected.

[0101] Select the regression correction frame after NMS (i.e., the regression correction frame to be selected) and calculate the intersection-over-union ratio with the real target frame. Select a certain number of regression correction frames (e.g., 128) as the target regression correction frames, and send the target regression correction frames to the ROI Pooling layer. The ROI Pooling layer applies the area corresponding to each regression correction frame to the shared feature map and extracts the regional features in the area, so that the type of mobile source (vehicle) can be identified based on the regional features.

[0102] In an embodiment of the present application, when measuring the vehicle speed of all mobile sources in a preset area, target detection can be performed in the preset area first to determine the mobile source in the image; then, an image movement trajectory of the mobile source is generated based on multiple frames of continuous corner points of the mobile source in the image; then, after calculating the actual moving distance of the mobile source based on the image movement trajectory, the real-time vehicle speed of the mobile source is calculated based on the actual moving distance.

[0103] It can be understood that, in this embodiment, the speed measurement of the mobile source (vehicle) can be carried out by using the YOLO-V5 algorithm to classify and detect all target vehicles on the lane, combining the machine vision algorithm provided by OpenCV to extract corner points and track the trajectory of the target vehicle, using the perspective model to convert the trajectory length of the target vehicle into the actual distance, and using the number of frames of the traffic camera video stream to achieve speed measurement. Specifically, the application process includes the following steps:

[0104] Step 1: The YOLO-V5 algorithm performs target detection in the preset area.

[0105] The YOLO-V5 vehicle target detection algorithm is divided into several processes: sampling, labeling, training and detection. The training process is particularly critical. The training function can be abstracted as: . W is the detection weight, train is the training function, Set For the labeled dataset, yaml The configuration file for the model. data for Set of yaml document, round is the number of training rounds, batch is the training batch, size is the normalized image size.

[0106] The image to be tested image , trained weight file W and Model yaml The file is input into the detection function, and the detection output result is: .

[0107] Step 2: OpenCV algorithm is used to extract target corner points and track trajectories.

[0108] Corner point extraction is the process of calculating the eigenvalue of each pixel and selecting the point with the largest eigenvalue. Use contour detection or corner point detection to extract the corner points of each vehicle, and use OpenCV to track the target. The corner point detection function uses: , img Pictures to be processed, r For key areas, pout Output vector for detection result, num is the maximum number of detected corner points, q To detect the quality of corner points, md is the minimum distance between two corner points, block To calculate the covariance matrix window size, usehar Indicates whether to use Harris Corner detection, usehar If it is 0, use Shi-Tomasi Operator, k For Harris Intermediate parameters for corner detection.

[0109] Trajectory tracking is to continuously track the unique corner points selected by the above targets in the traffic video stream for multiple frames to determine the movement trajectory of the target in the time series. The implementation function of trajectory tracking uses: , P 1 is the grayscale image corresponding to the extracted corner point, P 2 is the next frame of grayscale image, V 1 is the corner point vector set of the previous frame image, V 2 is the corner point vector set of the next frame image, stat is the state vector of the detection output, errTo detect the error message vector, size is the search window size, level To calculate the number of associated layers.

[0110] Step 3: The perspective transformation algorithm converts the trajectory length into the actual moving distance.

[0111] According to the traffic camera parameters and perspective relationship, the vehicle track length on the image is converted into the actual distance. In the case of oblique imaging, the "pixel-distance" function relationship can be simplified to: , S is the actual distance on the road, M For images in y The actual distance between any two segments in a direction, y i Mark points for images p i The corresponding pixel.

[0112] Step 4: Measure vehicle speed using frame conversion algorithm.

[0113] When the image triggers the speed measurement, the target tracking trajectory has n points, i.e. tracking n Frame, the initial position of the trajectory is p(x 0 , y 0 ) , the trigger position is p(x n ,y n ) , then by querying the "pixel-distance" two-dimensional array table, the actual distance corresponding to the initial point of the trajectory can be obtained as S(y 0 ) , the actual distance corresponding to the trigger point is S(y n ) Video per second fnum Frame, then the vehicle speed v The calculation formula is: .

[0114] In some embodiments, after identifying the license plate number of the mobile source in the preset area, the image of the mobile source in the preset area can be obtained by frame extraction from the traffic camera video stream, and then the license plate recognition of multiple target vehicles is realized through preprocessing, license plate positioning, character segmentation, character recognition and other steps, and the unique identification (vehicle ID) for subsequent data retrieval is realized. Specifically, the processing process after obtaining the mobile source image in this embodiment includes the following steps:

[0115] Step 1: Grayscale processing of the image.

[0116] In computer systems, RGB, YUV, CMY and HSI are commonly used color spaces. Different color spaces are just different ways of representing the same physical quantity. Grayscale is the process of converting the acquired color image into a grayscale image. Its purpose is to reduce the amount of data for digital image processing. The image acquired by the camera is in the RGB color space, where R represents red, G represents green, and B represents blue. The grayscale value is h ,Right now , w r , w g , w b The weights of red, green, and blue respectively.

[0117] Step 2: Image enhancement processing.

[0118] The main function of image enhancement is to effectively improve the quality and recognition of the image, making subsequent image processing easier. Spatial domain enhancement and frequency domain enhancement are commonly used enhancement techniques. Spatial domain filtering is achieved through convolution operations, in which the filter (kernel) is convolved with the image. The specific formula is: , G(x,y) is the pixel value after filtering, w(s,t) is the weight of the filter, f(x,y) is the original image.

[0119] Step 3: Image edge detection.

[0120] Using the Sobel operator to detect edges, by applying the Sobel operator to the image to process the edges in the x and y directions, it not only helps to reduce interference information, but also highlights the edge features of the license plate characters, providing a good foundation for subsequent license plate positioning. G x ) and the vertical convolution kernel ( G y ), calculate the gradient of the image in the horizontal and vertical directions respectively. The gradient is calculated as follows: The direction formula of the gradient is: .

[0121] Step 4: License plate location.

[0122] After the license plate image is preprocessed, the image meets the conditions for rectangle inspection, and the cv2.findContours function in OpenCV is used to perform rectangle detection on the image. In accordance with the national regulations on the license plate size, we set the aspect ratio of the candidate area between 3.0 and 3.5 to ensure that the framed area contains the license plate information and can effectively remove invalid areas.

[0123] Step 5: License plate image tilt correction.

[0124] In OpenCV, the HoughLines function is used to call the standard HoughLine transformation and the multi-scale HoughLine transformation, while the HoughLinesP function is used to call the cumulative probability HoughLine transformation. These functions are mainly used to detect straight lines in an image. A straight line can be represented on a two-dimensional coordinate axis as: ,in, m is the slope of the line, b is a straight line with y The principle of Hough transform is to represent a straight line as a point in polar coordinate space. In polar coordinate space, a straight line can be represented as (r,θ) ,in r is the distance from the origin to the line, θ is the angle of the line. In the Hough transform, each point in the two-dimensional image is projected into the polar coordinate space to form a curve. (r,θ) The points will intersect on the curve, and these intersection points represent the possible straight lines in the original image. In the cumulative probability Hough transform (HoughLinesP), unlike the standard Hough transform, it allows the detection of partially matched straight lines instead of just complete matches.

[0125] Step 6: License plate character segmentation.

[0126] The vertical projection method is used for image segmentation, and the projection information of each column of pixels in the image is used to determine the segmentation position between characters. By analyzing the vertical projection, the blank area between characters can be found, thereby effectively achieving character segmentation. The vertical projection method is a method for achieving character segmentation by analyzing the vertical projection of an image. Its basic principle is to find the blank area between characters by counting the pixels in each column of the image, thereby determining the segmentation position of the characters. The mathematical formula of the vertical projection method can be expressed in the following way: , P y (x) is the projection value of the image in the vertical direction, I(x,y) is the image at coordinates (x, y) The pixel value at H is the height of the image. The formula calculates the sum of the pixel values ​​of all rows on this column to form the value of the vertical projection. By analyzing this projection value, the segmentation position between characters can be found.

[0127] Step 7: License plate character segmentation.

[0128] If the first character of the license plate is a province in Chinese, the Chinese character template list is used for matching; if the second character is an English character, the English character template is used for matching; if the other characters are a mixture of English and numbers, the English and number template lists are used for matching. The matching process includes reading the address of the template and comparing it with the image, formatting and converting the template, thresholding the image, obtaining the size of the image to be detected, adjusting the template to the same size as the image, and finally performing template matching to obtain the license plate number corresponding to the mobile source.

[0129] In this embodiment, after obtaining the license plate number, it is possible to query whether the license plate number is located in a preset database, and the preset database may be an integrated database of land, vehicle and person.

[0130] Specifically, the "Heaven, Earth, Vehicle and People" integrated database is a comprehensive motor vehicle supervision system widely used nationwide, which includes remote online monitoring of heavy-duty diesel vehicles, non-road mobile machinery supervision, environmental protection supervision of new vehicles, regular emission inspections of motor vehicles, remote sensing monitoring supervision, on-site inspections of mobile sources, oil and gas recovery supervision, one vehicle one file, one map of the motor vehicle situation and other functions, breaking through the data barriers of the "Heaven", "Earth", "Vehicle" and "People" supervision modules.

[0131] When located in the database, the nitrogen oxide emissions of mobile sources can be calculated based on the real-time operating conditions method; when not located in the database, the nitrogen oxide emissions can be calculated using the vehicle speed corrected emission factor method.

[0132] Specifically, the real-time working condition method Nitrogen oxides The process of emission accounting is as follows:

[0133] Based on the real-time driving conditions of the vehicle, operating parameters such as original concentration and exhaust flow are selected from the data read from the integrated database of the sky, earth, vehicle and people, and the nitrogen oxide emissions of the passing vehicle are calculated by integrating the instantaneous values.

[0134]

[0135] is the average emission factor of nitrogen oxides within the calculation period (usually 1 minute), g / km ; is the first instantaneous nitrogen oxide emission of nitrogen oxides; is the instantaneous concentration of nitrogen oxides, ppm ; is the instantaneous exhaust mass flow rate, kg / s ; is the real-time vehicle speed, km / s ; z is the data sampling frequency; n is the number of tests, Represents the instantaneous molar mass of nitrogen oxides.

[0136] The process of calculating nitrogen oxide emissions using the speed-corrected emission factor method is as follows:

[0137] Determine the vehicle model base emission factor based on the vehicle model identification results;

[0138] Determine a vehicle speed correction factor according to the real-time vehicle speed;

[0139] The instantaneous nitrogen oxide emissions of the vehicle corresponding to the license plate number are calculated according to the vehicle model base emission factor and the vehicle speed correction factor as the second instantaneous nitrogen oxide emissions. The calculation formula for the second instantaneous nitrogen oxide emissions is as follows:

[0140]

[0141] The calculation period (usually 1 minute) Nitrogen oxides The average emission factor for g / km ; For car models i The baseline emission factor (which can be determined from Table 1) is g / km ; is the vehicle speed correction factor (which can be determined from Table 2), dimensionless, and is determined based on the instantaneous vehicle speed.

[0142] Table 1 Baseline emission factors by vehicle type ( g / km )

[0143]

[0144] Table 2 Vehicle speed correction factor

[0145]

[0146] In the embodiment of the present application, the total amount of nitrogen oxide emissions of each type of mobile source on each road section is calculated according to the first instantaneous nitrogen oxide emissions or the second instantaneous nitrogen oxide emissions. For example, Figure 2 is a flow chart of a method for compiling a linear nitrogen oxide emission inventory for motor vehicles based on a road network provided in an embodiment of the present application, Figure 2 The method may include but is not limited to steps 210 to 240:

[0147] Step 210: Linearly generalize the road to obtain a road generalization map, wherein the road generalization map includes road location information and traffic camera location information. Specifically, this embodiment extracts vector information of roads and traffic cameras based on road network data. Linearly generalize the road, use the Douglas-Peucker algorithm or the Ramer-Douglas-Peucker algorithm to simplify the road segments into fewer segments, while retaining key shapes and turning points. The location of the traffic camera is retained to ensure that its vector information is not affected by the generalization.

[0148] Step 220: node the road generalization graph to obtain a road segment graph. Specifically, in this embodiment, for each road, it is divided into multiple segments or road segments according to intersections and turning points. A corresponding algorithm or method, such as an intersection detection algorithm or curve fitting, is used to node the road into an indefinite number of segments.

[0149] Step 230: Based on the road location information and the traffic camera location information, a relationship correspondence matrix between the roads and cameras in the road segment graph is established using the minimum distance method and the total balance principle, and the representative camera of each road is recorded.

[0150] Step 240: According to the camera position corresponding to the first instantaneous nitrogen oxide emission or the second instantaneous nitrogen oxide emission, the corresponding road section is determined in the relationship correspondence matrix, and the total nitrogen oxide emission of each type of mobile source on the road section is calculated. Specifically, the calculation formula of the total nitrogen oxide emission is as follows:

[0151]

[0152] The road number is i Total nitrogen oxide emissions, g ; is the average emission factor of nitrogen oxides from vehicles, g / km ; The road number is i The traffic volume, pcu / h ; The road number is i Length, km .

[0153] In the embodiment of the present application, the road sections are gridded according to the total amount of nitrogen oxide emissions to obtain a mobile source nitrogen oxide emission inventory. For example, Figure 3 is a flowchart of a method for compiling a gridded motor vehicle nitrogen oxide emission inventory provided in an embodiment of the present application. Figure 3 The method may include but is not limited to steps 310 to 330:

[0154] Step 310: Determine the range and resolution of the grid to be created based on the road network data and the technical requirements of the air quality model simulation. Then use the meshgrid tool to create an evenly spaced grid to divide the road network area into multiple grid cells.

[0155] Step 320: Cut the area to which the road section belongs; specifically, in this embodiment, all road sections in the area to which the road section belongs are cut to form multiple subdivided emission sections. The cutting can be performed based on the starting point and end point of the road, or based on key points on the road. The nitrogen oxide emission of each subdivided emission section can be determined by calculation based on parameters such as vehicle type, vehicle speed, and section length.

[0156] Step 330: Calculate the nitrogen oxide emission component in each grid after the grid cutting process according to the total nitrogen oxide emission. Specifically, this embodiment superimposes and reconstructs statistics on the subdivided emission sections according to the grid IDs to which they belong, and obtains the nitrogen oxide emission component of each grid. The calculation formula is: , For Grid i of Nitrogen oxides Emissions ( g ), For Grid i Internal subdivision emission section j of Nitrogen oxides Emissions( g ). Then mark the grids according to the nitrogen oxide emission components to obtain a mobile source nitrogen oxide emission inventory. Specifically, the marking process of this embodiment can be marked by color. For example, different colors are assigned to each grid according to the size of the nitrogen oxide emission component, and gradient colors are used to represent different levels of emissions. According to the range of emissions, they can be divided into three levels: low, medium, and high, and then represented by green, yellow, and red, respectively, so that the data can be made clearer.

[0157] In summary, the method of this embodiment has the following beneficial effects:

[0158] First, this embodiment solves the problem of lack of effective road network dynamic traffic data faced by dynamic emission measurement of mobile sources such as motor vehicles by introducing regional target detection algorithms, machine vision algorithms and image recognition technologies.

[0159] Second, this embodiment realizes real-time statistics and measurement of key data such as vehicle model, traffic flow, speed and license plate by acquiring traffic camera video streams and integrated database information of the sky, earth and vehicles in real time, thereby improving the real-time and accuracy of emission calculation.

[0160] Third, this embodiment combines regional target detection, machine vision and image recognition technology, and comprehensively utilizes multi-source data (video streams, database information, etc.), thereby achieving comprehensive acquisition of data required for emission measurement and improving data diversity and comprehensiveness.

[0161] Fourth, this embodiment uses geographic information technology to perform networked distribution of emissions, providing a real-time motor vehicle emissions inventory for the air quality model, further improving the accuracy and practicality of simulation and analysis.

[0162] The embodiment of the present application also provides an electronic device, the electronic device includes a memory and a processor, the memory stores a computer program, and the processor implements the above method when executing the computer program. The electronic device can be any intelligent terminal including a tablet computer, a car computer, etc.

[0163] It can be understood that the contents of the above method embodiments are all applicable to the present device embodiments, the functions specifically implemented by the present device embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0164] See also Figure 4 , Figure 4 The hardware structure of an electronic device of another embodiment is illustrated, and the electronic device includes:

[0165] The processor 410 may be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of the present application;

[0166] The memory 420 can be implemented in the form of a read-only memory (ROM), a static storage device, a dynamic storage device, or a random access memory (RAM). The memory 420 can store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program codes are stored in the memory 420, and the processor 410 calls and executes the methods of the embodiments of this application;

[0167] Input / output interface 430, used to implement information input and output;

[0168] Communication interface 440, used to realize communication interaction between the device and other devices, which can be realized through wired mode (such as USB, network cable, etc.) or wireless mode (such as mobile network, WIFI, Bluetooth, etc.);

[0169] bus 450 , which transmits information between the various components of the device (e.g., processor 410 , memory 420 , input / output interface 430 , and communication interface 440 );

[0170] The processor 410 , the memory 420 , the input / output interface 430 , and the communication interface 440 are connected to each other in communication within the device via the bus 450 .

[0171] An embodiment of the present application further provides a computer-readable storage medium, which stores a computer program, and the computer program implements the above method when executed by a processor.

[0172] It can be understood that the contents of the above method embodiments are all applicable to the present storage medium embodiments, the functions specifically implemented by the present storage medium embodiments are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments.

[0173] The memory, as a non-transient computer-readable storage medium, can be used to store non-transient software programs and non-transient computer executable programs. In addition, the memory may include a high-speed random access memory, and may also include a non-transient memory, such as at least one disk storage device, a flash memory device, or other non-transient solid-state storage device. In some embodiments, the memory may optionally include a memory remotely disposed relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0174] The embodiments described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. Those skilled in the art will appreciate that with the evolution of technology and the emergence of new application scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0175] Those skilled in the art will appreciate that the technical solutions shown in the figures do not constitute a limitation on the embodiments of the present application, and may include more or fewer steps than shown in the figures, or a combination of certain steps, or different steps.

[0176] The device embodiments described above are merely illustrative, and the units described as separate components may or may not be physically separated, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0177] Those skilled in the art will appreciate that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices may be implemented as software, firmware, hardware, or a suitable combination thereof.

[0178] The terms "first", "second", "third", "fourth", etc. (if any) in the specification of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0179] It should be understood that in the present application, "at least one (item)" means one or more, and "plurality" means two or more. "And / or" is used to describe the association relationship of associated objects, indicating that three relationships may exist. For example, "A and / or B" can mean: only A exists, only B exists, and A and B exist at the same time, where A and B can be singular or plural. The character " / " generally indicates that the objects associated before and after are in an "or" relationship. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b or c can mean: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, c can be single or multiple.

[0180] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the above units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0181] The units described above as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0182] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0183] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including multiple instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of various embodiments of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, referred to as ROM), random access memory (Random Access Memory, referred to as RAM), disk or optical disk and other media that can store programs.

[0184] The preferred embodiments of the present invention are described above with reference to the accompanying drawings, but the scope of the rights of the present invention is not limited thereto. Any modification, equivalent substitution and improvement made by a person skilled in the art without departing from the scope and essence of the present invention should be within the scope of the rights of the present invention.

Claims

1. A method for compiling a mobile source nitrogen oxide emission inventory, characterized in that: The method comprises: Carry out vehicle type recognition on mobile sources in a preset area to obtain vehicle type recognition results; Measuring the speed of all mobile sources in the preset area to obtain the real-time speed of each mobile source; Identify the license plate number of the mobile source in the preset area; Determining that the license plate number is in a preset database, and calculating the instantaneous nitrogen oxide emission of the mobile source corresponding to the license plate number according to the real-time vehicle speed as a first instantaneous nitrogen oxide emission; Determining that the license plate number is not located in the preset database, and calculating the instantaneous nitrogen oxide emissions of the mobile source corresponding to the license plate number as the second instantaneous nitrogen oxide emissions according to the vehicle type recognition result and the real-time vehicle speed; Calculating the total amount of nitrogen oxide emissions of each type of mobile source on each road section according to the first instantaneous nitrogen oxide emissions or the second instantaneous nitrogen oxide emissions; According to the grid-processing road sections of the total amount of nitrogen oxide emissions, a mobile source nitrogen oxide emission inventory is obtained; The step of calculating the total amount of nitrogen oxide emissions of each type of mobile source on each road section according to the first instantaneous nitrogen oxide emissions or the second instantaneous nitrogen oxide emissions includes: Linearly generalizing the road to obtain a road generalization map, wherein the road generalization map includes road location information and traffic camera location information; Performing road node conversion on the road generalization graph to obtain a road segment graph; According to the road position information and the traffic camera position information, a relationship correspondence matrix between roads and cameras in the road segment graph is established; determining a corresponding road section in the relationship correspondence matrix according to a camera position corresponding to the first instantaneous nitrogen oxide emission or the second instantaneous nitrogen oxide emission; Calculate the total amount of nitrogen oxide emissions from each type of mobile source on the road section.

2. The method according to claim 1, characterized in that The step of performing vehicle type recognition on the mobile source in the preset area to obtain a vehicle type recognition result includes: Get the frame images of the traffic camera video stream; Extracting a shared feature map of the frame-by-frame image of the traffic camera video stream; Sliding the shared feature map to obtain a plurality of anchor frames; Predict the foreground probability and offset of all anchor boxes through parallel convolution kernels; Determine a regression correction frame to be selected from all the anchor frames according to the offset and the foreground probability; Determine the target regression correction frame according to the intersection and union ratio of the regression correction frame to be selected and the real target frame; Extracting regional features from the shared feature map according to the target regression correction frame; The vehicle type of each mobile source is determined according to the regional characteristics to obtain a vehicle type recognition result.

3. The method according to claim 1, characterized in that The measuring of the vehicle speed of all mobile sources in the preset area to obtain the real-time vehicle speed of each mobile source includes: Performing target detection in the preset area to determine the moving source in the image; Generate an image movement trajectory of the moving source according to multiple frames of continuous corner points of the moving source in the image; Calculate the actual moving distance of the moving source according to the image moving trajectory; The real-time vehicle speed of the mobile source is calculated according to the actual moving distance.

4. The method according to claim 1, characterized in that The identifying the license plate number of the mobile source in the preset area includes: Acquire a moving source image within the preset area; performing grayscale processing and image enhancement processing on the moving source image in sequence; Performing image edge detection on the mobile source image after image enhancement processing to obtain license plate feature information of the mobile source image; Determine the location in the moving source image according to the license plate feature information to obtain a license plate image; Performing tilt correction on the license plate image; Perform character segmentation on the tilt-corrected license plate image to obtain a character segmentation result; The license plate number of the mobile source is determined according to the character segmentation result.

5. The method according to claim 1, characterized in that The calculation formula of the first instantaneous nitrogen oxide emission is as follows: in, represents the average emission factor of nitrogen oxides during the calculation period, represents the first instantaneous nitrogen oxide emission, represents the instantaneous concentration of the nitrogen oxides, represents the instantaneous exhaust mass flow rate, v represents the real-time vehicle speed, z represents the frequency of data acquisition, n represents the number of tests, Represents the instantaneous molar mass of nitrogen oxides.

6. The method according to claim 1, characterized in that The calculating, according to the vehicle type recognition result and the real-time vehicle speed, the instantaneous nitrogen oxide emission of the mobile source corresponding to the license plate number as the second instantaneous nitrogen oxide emission includes: Determining a vehicle model base emission factor according to the vehicle model identification result; determining a vehicle speed correction factor according to the real-time vehicle speed; The instantaneous nitrogen oxide emissions of the mobile source corresponding to the license plate number are calculated according to the vehicle model base emission factor and the vehicle speed correction factor as the second instantaneous nitrogen oxide emissions.

7. The method according to claim 1, characterized in that The gridding of the road sections according to the total amount of nitrogen oxide emissions to obtain a mobile source nitrogen oxide emission inventory includes: Performing equidistant grid cutting processing on the area to which the road section belongs; Calculate the nitrogen oxide emission component in each grid after grid cutting according to the total nitrogen oxide emission; The grids are marked according to the nitrogen oxide emission components to obtain a mobile source nitrogen oxide emission inventory.

8. An electronic device, characterized in that: include: at least one processor; at least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the method according to any one of claims 1 to 7.

9. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images