Artificial intelligence image processing method based on intelligent traffic

Abstract

The invention relates to the technical field of image processing application, and particularly discloses an artificial intelligence image processing method based on intelligent traffic, which comprises the steps of image acquisition and preprocessing, feature extraction, artificial intelligence model construction and training, intelligent analysis and decision support, intelligent image processing and feedback, result output and application and the like. According to the artificial intelligence image processing method based on the intelligent traffic, information extracted from features is used for establishing an artificial intelligence analysis model, the intelligent analysis capability of an intelligent traffic management system is remarkably enhanced, and in the third step, the extracted features and historical data are subjected to mapping calculation, so that the intelligent analysis capability of the intelligent traffic management system is improved. The constructed artificial intelligence analysis model can learn and identify the mode in the traffic image so as to predict the optimal processing method, and the establishment of the model enables the intelligent traffic management system not only to process the current image data, but also to perform learning and prediction based on historical data, thereby improving the intelligent level of the system.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing, and in particular to an artificial intelligence image processing method based on smart transportation. Background Art

[0002] With the acceleration of urbanization and the continuous increase in the number of vehicles, problems such as traffic congestion and frequent accidents have become increasingly prominent. Traditional traffic management methods can no longer meet the needs of modern society for traffic safety and efficiency. Therefore, intelligent transportation systems have emerged. Intelligent transportation systems use advanced image recognition technology to analyze and process images in traffic scenes, and automatically obtain and process information such as traffic flow and traffic violations, thereby improving the efficiency and accuracy of traffic management.

[0003] Although intelligent transportation systems have made certain progress in image processing technology, existing image processing methods still have some defects. Existing image processing methods are highly dependent on data and require a large amount of labeled data to train the model, which not only increases the cost of data collection and processing, but also limits the generalization ability of the model in new scenarios. Existing image processing methods have problems of low efficiency and insufficient accuracy, and have high requirements on the accuracy of image data, which limits the ability to process large amounts of image data and affects overall work efficiency. Summary of the invention

[0004] The purpose of the present invention is to solve the problem that existing image processing methods are highly dependent on data and require a large amount of labeled data to train models, which not only increases the cost of data collection and processing, but also limits the generalization ability of the model in new scenarios. Existing image processing methods have the problems of low efficiency and insufficient accuracy, and have high requirements on the accuracy of image data, which limits the ability to process large amounts of image data and affects the overall work efficiency. An artificial intelligence image processing method based on smart transportation is proposed.

[0005] The purpose of the present invention can be achieved by the following technical solution: An artificial intelligence image processing method based on smart transportation, comprising the following steps: Step 1: Image acquisition and preprocessing: collect traffic element images and preprocess the collected traffic element images; Step 2: Feature extraction: Extract feature information useful for identification and classification from the traffic element image preprocessed in step 1; Step 3: Construction and training of artificial intelligence model: Based on the image feature information of traffic elements extracted in step 2, the big data cloud server and forward propagation neural network algorithm are used to map and calculate the historical data and processing methods of traffic images in the smart traffic management system to build an artificial intelligence analysis model for traffic images; Step 4: Intelligent analysis and decision support: Call the traffic image artificial intelligence analysis model obtained in step 3 from the big data cloud server, perform intelligent analysis on the traffic element image preprocessed in step 1, and select the best processing method; Step 5: Intelligent processing and feedback of images: According to the best processing method in step 4, intelligent processing is performed on the traffic element images, and the processing results are fed back to the smart traffic management system in real time; Step 6: Result output and application: Push the image data and analysis results processed in step 5 to the smart traffic management system to provide decision support for the smart traffic management system.

[0006] Preferably, in step 1, the process of collecting traffic element images includes the following steps: S11: Equipment deployment: Deploy image sensor equipment such as cameras, radars, laser scanners, etc. on traffic roads, facilities, and tools; S12: Real-time monitoring: Use real-time monitoring of traffic conditions to capture dynamic changes to ensure the timeliness and accuracy of data; S13: Data transmission: The collected image data needs to be transmitted to the background through a wired or wireless network. Compression and encryption are required during the data transmission process to ensure transmission efficiency and security; S14: After decompressing and decrypting the transmitted image data, the image data is stored on the server for subsequent analysis and processing.

[0007] Preferably, the traffic element image includes the following aspects: vehicles, pedestrians, traffic signs and signals, roads, license plates, and road environment.

[0008] Preferably, in step one, preprocessing of the traffic element image includes but is not limited to image correction, image denoising, image enhancement, image normalization, image cropping, image segmentation, image sharpening, and image fusion.

[0009] Preferably, in step 2, the feature information extracted includes but is not limited to color features, texture features, shape features, spatial relationship features, semantic features, and optical flow features.

[0010] Preferably, in step three, the construction and training of the artificial intelligence model includes model design, data preparation, model training, and model optimization.

[0011] Preferably, model design is to design a forward propagation neural network model based on the extracted feature information, which is used to learn and identify patterns in traffic images; data preparation is used to collect and organize historical data and processing methods of traffic images in the intelligent traffic management system to provide a data set for training the model; model training is to use the computing power of the big data cloud server to train the model so that it can identify features in traffic images based on historical data and predict the best processing method; model optimization is to optimize the performance of the model and improve its accuracy by adjusting the model parameters and structure.

[0012] Preferably, in step 4, intelligent analysis and decision support includes the following steps: S41: Use artificial intelligence models to perform pattern recognition and classification on pre-processed traffic element images; S42: Analyze traffic flow, traffic events and traffic conditions based on the recognition results to determine the best strategy for traffic management; S43: Combine the analysis results with the real-time data of the traffic management system to provide real-time decision support and traffic optimization suggestions.

[0013] Preferably, in step 5, the intelligent processing and feedback of the image includes the following steps: S51: further analyzing and processing the traffic element image according to the preferred processing method, such as traffic flow control, accident detection and response, traffic signal optimization, etc.; S52: Display the processing results to traffic management personnel through a visual interface for manual review and intervention; S53: The processing results and feedback information are stored in a database for further learning and optimization of the model.

[0014] Preferably, in step six, the result output and application includes the following steps: S61: Format the processed image data and analysis results to facilitate integration and use in the intelligent traffic management system; S62: Push data and results to the smart traffic management system through API or other interfaces to achieve automated traffic management; S63: Regularly evaluate and update pushed data and results to ensure that the decision support of the smart traffic management system is always up-to-date and accurate.

[0015] Compared with the prior art, the present invention has the following beneficial effects: (1) This artificial intelligence image processing method based on smart transportation uses the feature extraction information to establish an artificial intelligence analysis model. This method significantly enhances the intelligent analysis capability of the smart transportation management system. In step three, by mapping the extracted features with historical data, the constructed artificial intelligence analysis model can learn and identify patterns in traffic images, thereby predicting the best processing method. The establishment of this model enables the smart transportation management system to not only process current image data, but also learn and predict based on historical data, thereby improving the intelligence level of the system. As the best processing method in step five pushes the processed data of the image to the smart transportation management system, the system accumulates more and more data, which provides a basis for artificial intelligence analysis. The model provides richer data support. With the accumulation of data and the continuous optimization of the model, the analysis ability of the intelligent traffic management system will become more powerful, and it will be able to more accurately predict and handle various traffic conditions, thereby improving the efficiency and safety of traffic management. This method realizes the continuous optimization and adaptation of the intelligent traffic management system through continuous image processing and data push. As more and more image features of the best processing methods are extracted and used in the artificial intelligence analysis model, the performance of the model will be continuously improved, making the next artificial intelligence analysis model more powerful. This continuous optimization and adaptive ability enables the intelligent traffic management system to continuously adapt to changes in traffic conditions over time and improve the accuracy of its prediction and response; (2) This artificial intelligence image processing method based on smart transportation has achieved a comprehensive improvement in the level of intelligent traffic management through the close cooperation of six steps. First, in step one, by deploying image sensor equipment and real-time monitoring of traffic conditions, real-time acquisition and preprocessing of traffic element images are achieved, providing a high-quality data foundation for subsequent intelligent analysis. The feature extraction in step two further extracts key information from the preprocessed image. In step three, the artificial intelligence model constructed using the big data cloud server and the forward propagation neural network algorithm can learn from historical data and identify patterns in traffic images, providing a powerful intelligent analysis tool for traffic management. The intelligent analysis and decision support in steps 4 and 5, as well as the intelligent processing and feedback of images, enable the traffic management system to respond to traffic conditions in real time and make decisions quickly, such as traffic flow control and accident response, which greatly reduces the need for human intervention and improves the automation level and response speed of traffic management. Finally, the result output and application in step 6 pushes the processed image data and analysis results to the intelligent traffic management system, providing real-time decision support for traffic management departments and helping to formulate more effective traffic management strategies. Overall, the coordination of these six steps realizes the intelligence of the entire process from data collection to decision support, greatly improving the efficiency and effectiveness of traffic management. (3) The artificial intelligence image processing method based on smart transportation first extracts features from the collected images, then uses these features to map and calculate with the data of the smart transportation management system, establishes an artificial intelligence analysis model, and finally uses the model to intelligently analyze and obtain the best processing method to process the collected images, thereby achieving high efficiency and high accuracy in image processing. In step two, the feature extraction speed is fast and the requirements are low, which helps to quickly process a large amount of image data and improve overall work efficiency. In step five, the image is intelligently processed again using the best processing method, making the processing results more accurate and better meeting the needs of smart transportation management. This two-stage image processing process can provide high-quality image processing results while maintaining high efficiency. This is particularly important for traffic management systems that require rapid response, such as in traffic congestion or accident detection, to quickly identify problems and take action while ensuring the accuracy of recognition and avoiding false alarms or omissions. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0017] Figure 1 The figure is a flow chart of the processing method of the present invention. DETAILED DESCRIPTION

[0018] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0019] See also Figure 1 As shown, an artificial intelligence image processing method based on smart transportation includes the following steps: Step 1: Image acquisition and preprocessing: collect traffic element images and preprocess the collected traffic element images; Step 2: Feature extraction: Extract feature information useful for identification and classification from the traffic element image preprocessed in step 1; Step 3: Construction and training of artificial intelligence model: Based on the image feature information of traffic elements extracted in step 2, the big data cloud server and forward propagation neural network algorithm are used to map and calculate the historical data and processing methods of traffic images in the smart traffic management system to build an artificial intelligence analysis model for traffic images; Step 4: Intelligent analysis and decision support: Call the traffic image artificial intelligence analysis model obtained in step 3 from the big data cloud server, perform intelligent analysis on the traffic element image preprocessed in step 1, and select the best processing method; Step 5: Intelligent processing and feedback of images: According to the best processing method in step 4, intelligent processing is performed on the traffic element images, and the processing results are fed back to the smart traffic management system in real time; Step 6: Result output and application: Push the image data and analysis results processed in step 5 to the smart traffic management system to provide decision support for the smart traffic management system.

[0020] Preferably, in step 1, the process of collecting traffic element images includes the following steps: S11: Equipment deployment: Deploy image sensor equipment such as cameras, radars, laser scanners, etc. on traffic roads, facilities, and tools; S12: Real-time monitoring: Use real-time monitoring of traffic conditions to capture dynamic changes to ensure the timeliness and accuracy of data; S13: Data transmission: The collected image data needs to be transmitted to the background through a wired or wireless network. Compression and encryption are required during the data transmission process to ensure transmission efficiency and security; S14: After decompressing and decrypting the transmitted image data, the image data is stored on the server for subsequent analysis and processing.

[0021] Preferably, the traffic element image includes the following aspects: vehicles, pedestrians, traffic signs and signals, roads, license plates, and road environment, wherein vehicles include various types of vehicles, such as cars, trucks, buses, and motorcycles; traffic signs and signals include traffic lights, traffic signs, and roadblocks; roads include lane lines, road conditions, and road boundaries; and the road environment includes buildings, green belts, and traffic facilities around the road.

[0022] Preferably, in step one, the preprocessing of the traffic element image includes but is not limited to image correction, image denoising, image enhancement, image normalization, image cropping, image segmentation, image sharpening, and image fusion. Image correction uses perspective transformation and geometric correction technology to correct image tilt, distortion and other problems caused by shooting angles, which can improve the geometric accuracy of the image and provide accurate image data for subsequent feature extraction and analysis. Image denoising uses mean filtering, median filtering, and Gaussian filtering methods to eliminate unnecessary noise in the image, improve image quality, restore image clarity and detail information, and make the image more realistic, natural, and easy to analyze and process. Image enhancement uses histogram equalization and contrast enhancement technology to improve the contrast, brightness and color of the image. It is easier to observe and analyze, improves the visual effect of the image, makes the target features more prominent, and facilitates subsequent identification and analysis. Image normalization scales the pixel values ​​of the image to a small, specified range to eliminate the impact of dimension, which can speed up model training, improve model accuracy, reduce the impact of geometric transformation, and improve the generalization ability of the model. Image cropping removes areas of no interest in the image, retains only key features or areas of interest, reduces unnecessary calculations, and improves processing speed, making the image more focused and highlighting the subject. Image segmentation uses threshold segmentation, region growing, and deep learning methods to divide the image into different regions or objects, help identify and locate objects in the image, and facilitate subsequent target identification and analysis. Image sharpening uses Unsharp Masking and Laplacian sharpening technology to enhance the edges and details of the image, improve the clarity of the image, make the edges more obvious, and improve image quality. Image fusion uses pixel-level, feature-level, or decision-level fusion to combine the information of multiple images, improve the visual effect and information content of the image, and enhance the accuracy and reliability of target detection and recognition.

[0023] Preferably, in step 2, the extracted feature information includes but is not limited to color features, texture features, shape features, spatial relationship features, semantic features, and optical flow features, wherein the color features extract the color information in the image by converting the color space (such as RGB to HSV) and calculating the color histogram, which can distinguish different traffic signs and vehicle types and improve the recognition accuracy; the texture features use methods such as gray-level co-occurrence matrix (GLCM) or local binary pattern (LBP) to describe the texture information of the image; the texture features can effectively capture the local structural information of the image, which is very useful for distinguishing different types of road conditions and traffic signs; the shape features extract the shape information of the object in the image by edge detection (such as Sobel and Canny operators) and contour tracking technology, which is crucial for identifying the contours and postures of traffic participants such as vehicles and pedestrians; and the spatial relationship The spatial relationship features help to understand the patterns of traffic flow and the interactions between traffic participants by analyzing the relative positions and distances between different objects in the image, such as using clustering algorithms or graphic models. The semantic features use deep learning models (such as CNN) to automatically learn the high-level semantic information of the image. The semantic features can provide rich contextual information and help improve the depth and accuracy of traffic scene understanding. The optical flow features are extracted by calculating the motion vectors of pixels in the image sequence. The optical flow features are of great significance for analyzing the dynamic changes of traffic flow and detecting the motion status of vehicles. By extracting image features, the dimension of the data can be reduced, the processing speed and efficiency can be improved, and the recognition and prediction accuracy of the model can be improved. By selecting the most representative features, the risk of overfitting of the model can be reduced, making the decision-making process of the model more transparent and explainable.

[0024] Preferably, in step three, the construction and training of the artificial intelligence model includes model design, data preparation, model training, and model optimization.

[0025] Preferably, the model design is to design a forward propagation neural network model based on the extracted feature information, which is used to learn and recognize patterns in traffic images, and adopt deep learning technology, especially convolutional neural network, to extract image features through convolutional layers, and the output of each layer is used as the input of the next layer, so as to abstract and extract image features layer by layer, and also includes pre-training the model on the image data set to initialize the network parameters to improve the convergence speed of the network; data preparation is used to collect and organize historical data and processing methods of traffic images in the intelligent traffic management system, and provide a data set for the training model, including manually collecting data from multiple sources, constructing a unimodal traffic data set and a multimodal traffic data set, and adopting histogram equalization, The data is preprocessed by image resizing to ensure the consistency and quality of the model input; model training is to use the computing power of the big data cloud server to train the model. During the training process, the model is iteratively trained using the training set data so that it can identify features in traffic images based on historical data and predict the best processing method. Data enhancement techniques are used in training, including rotation, translation, shearing and scaling, to expand the training data set and improve the generalization ability of the model; model optimization is to optimize the performance of the model and improve its accuracy by adjusting the model parameters and structure, including using the Dropout layer to prevent overfitting and configuring appropriate learning rates and optimizers to improve the stability and convergence speed of training.

[0026] Preferably, in step 4, intelligent analysis and decision support includes the following steps: S41: Use artificial intelligence models to perform pattern recognition and classification on pre-processed traffic element images; S42: Analyze traffic flow, traffic events and traffic conditions based on the recognition results to determine the best strategy for traffic management; S43: Combine the analysis results with the real-time data of the traffic management system to provide real-time decision support and traffic optimization suggestions.

[0027] Preferably, in step 5, the intelligent processing and feedback of the image includes the following steps: S51: further analyzing and processing the traffic element image according to the preferred processing method, such as traffic flow control, accident detection and response, traffic signal optimization, etc.; S52: Display the processing results to traffic management personnel through a visual interface for manual review and intervention; S53: The processing results and feedback information are stored in a database for further learning and optimization of the model.

[0028] Preferably, in step six, the result output and application includes the following steps: S61: Format the processed image data and analysis results to facilitate integration and use in the intelligent traffic management system; S62: Push data and results to the smart traffic management system through API or other interfaces to achieve automated traffic management; S63: Regularly evaluate and update pushed data and results to ensure that the decision support of the smart traffic management system is always up-to-date and accurate.

[0029] The preferred embodiments of the present invention disclosed above are only used to help illustrate the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to the specific implementation methods described. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. An artificial intelligence image processing method based on smart transportation, characterized in that: The following steps are involved: Step 1: Image acquisition and preprocessing: collect traffic element images and preprocess the collected traffic element images; Step 2: Feature extraction: Extract feature information useful for identification and classification from the traffic element image preprocessed in step 1; Step 3: Construction and training of artificial intelligence model: Based on the image feature information of traffic elements extracted in step 2, the big data cloud server and forward propagation neural network algorithm are used to map and calculate the historical data and processing methods of traffic images in the smart traffic management system to build an artificial intelligence analysis model for traffic images; Step 4: Intelligent analysis and decision support: Call the traffic image artificial intelligence analysis model obtained in step 3 from the big data cloud server, perform intelligent analysis on the traffic element image preprocessed in step 1, and select the best processing method; Step 5: Intelligent processing and feedback of images: According to the best processing method in step 4, intelligent processing is performed on the traffic element images, and the processing results are fed back to the smart traffic management system in real time; Step 6: Result output and application: Push the image data and analysis results processed in step 5 to the smart traffic management system to provide decision support for the smart traffic management system.

2. The artificial intelligence image processing method based on smart transportation according to claim 1 is characterized in that: In step 1, the process of collecting traffic element images includes the following steps: S11: Equipment deployment: Deploy image sensor equipment such as cameras, radars, laser scanners, etc. on traffic roads, facilities, and tools; S12: Real-time monitoring: Use real-time monitoring of traffic conditions to capture dynamic changes to ensure the timeliness and accuracy of data; S13: Data transmission: The collected image data needs to be transmitted to the background through a wired or wireless network. Compression and encryption are required during the data transmission process to ensure transmission efficiency and security; S14: After decompressing and decrypting the transmitted image data, the image data is stored on the server for subsequent analysis and processing.

3. The artificial intelligence image processing method based on smart transportation according to claim 2 is characterized in that: Traffic element images include the following aspects: vehicles, pedestrians, traffic signs and signals, roads, license plates, and road environment.

4. The artificial intelligence image processing method based on smart transportation according to claim 1 is characterized in that: In step one, the preprocessing of the traffic element image includes but is not limited to image correction, image denoising, image enhancement, image normalization, image cropping, image segmentation, image sharpening, and image fusion.

5. The artificial intelligence image processing method based on smart transportation according to claim 1 is characterized in that: In step 2, the extracted feature information includes but is not limited to color features, texture features, shape features, spatial relationship features, semantic features, and optical flow features.

6. The artificial intelligence image processing method based on smart transportation according to claim 1 is characterized in that: In step three, the construction and training of the artificial intelligence model includes model design, data preparation, model training, and model optimization.

7. The artificial intelligence image processing method based on smart transportation according to claim 6 is characterized in that: Model design is to design a forward propagation neural network model based on the extracted feature information to learn and identify patterns in traffic images; data preparation is used to collect and organize historical data and processing methods of traffic images in the intelligent traffic management system to provide a data set for training the model; model training is to use the computing power of the big data cloud server to train the model so that it can identify features in traffic images based on historical data and predict the best processing method; model optimization is to optimize the performance of the model and improve its accuracy by adjusting the model parameters and structure.

8. The artificial intelligence image processing method based on smart transportation according to claim 1 is characterized in that: In step 4, intelligent analysis and decision support includes the following steps: S41: Use artificial intelligence models to perform pattern recognition and classification on pre-processed traffic element images; S42: Analyze traffic flow, traffic events and traffic conditions based on the recognition results to determine the best strategy for traffic management; S43: Combine the analysis results with the real-time data of the traffic management system to provide real-time decision support and traffic optimization suggestions.

9. The artificial intelligence image processing method based on smart transportation according to claim 1 is characterized in that: In step 5, the intelligent processing and feedback of the image includes the following steps: S51: further analyzing and processing the traffic element image according to the preferred processing method, such as traffic flow control, accident detection and response, traffic signal optimization, etc.; S52: Display the processing results to traffic management personnel through a visual interface for manual review and intervention; S53: The processing results and feedback information are stored in a database for further learning and optimization of the model.

10. The artificial intelligence image processing method based on smart transportation according to claim 1 is characterized in that: In step six, the result output and application includes the following steps: S61: Format the processed image data and analysis results to facilitate integration and use in the intelligent traffic management system; S62: Push data and results to the smart traffic management system through API or other interfaces to achieve automated traffic management; S63: Regularly evaluate and update pushed data and results to ensure that the decision support of the smart traffic management system is always up-to-date and accurate.

Images