Intelligent Transport Systems are getting into the trend. They are all about a different, smarter approach to the transportation system, including traffic management, road security, and many other aspects. So, many countries are either implementing or considering the idea of intelligent traffic systems.
However, the development of intelligent transportation systems is a challenging task. It requires strong expertise in technologies, including IoT, Computer Vision, Machine Learning, etc. Also, the system needs to be as precise as possible.
So, how to build a secure and cost-efficient system? What benefits does the development of intelligent transportation systems bring? We will answer these and other questions.
In the article, you will discover:
- What is an intelligent transportation system?;
- Types of intelligent traffic systems;
- Why do you need the Intelligent Transport System?;
- Trends that power Intelligent Transport Systems;
- What are the best practices for the development of intelligent transportation systems?
An Intelligent Transport System (ITS) applies sensing, analysis, control, and communication technologies to improve safety, mobility, and efficiency.
An intelligent traffic system includes a wide array of applications. They process and share information to ease congestion, enhance traffic management, and increase transportation benefits to commercial users and the public in general.
Typically, tech experts classify Intelligent Transport Systems into three categories.
- Mobility ITS:
They are aimed to provide the shortest route between origin-destination pairs considering factors such as distance, time, energy consumption, etc., in a data-rich travel environment. These applications can help monitor and manage transportation system performance by adjusting traffic signals, dynamically managing transit operations, or dispatching emergency maintenance services.
- Safety ITS:
Safety applications reduce crashes by providing advisories and warnings. These applications include vehicle safety applications, emergency management. A safety Intelligent Transport System can, for instance, give a speed warning at a slippery roadway.
- Environment ITS:
The instant traffic congestion data can help make informed decisions that in-turn decrease the environmental impact of day-to-day trips. It gets possible to avoid congestion by taking alternate routes or rescheduling the trips, which can make the trips more eco-friendly.
Also, Intelligent Transport Systems are classified according to the location of the system itself. Thus, there are outdoor and indoor Intelligent Transport Systems.
Indoor systems are the ones located in the vehicle and transfer data to the driver. Outdoor systems are usually located above the road.
But why do we need Intelligent Transport Systems, and what benefits they can bring?
Many countries opt for implementing Intelligent Transport Systems to improve their transportation systems. There are a number of reasons why an Intelligent Transport System is needed.
First of all, an Intelligent Transport System allows us to monitor and adjust road network performance in real-time. Also, data previously collected by the costly physical infrastructure can be provided through new data sources, such as cameras.
Analysis performed based on historic data can be undertaken by Intelligent Transport Systems through real-time data analytics for Intelligent Transportation Systems.
Road users’ choices can now be influenced through a wide range of publication channels such as mobile devices or in-car systems.
Also, there are different stakeholders that can benefit from Intelligent Transport Systems:
Different types of Intelligent Traffic Systems have different purposes. However, the principle of their work is similar. So, how to make an Intelligent Transport System work?
Intelligent Transport System or ITS is a combination of leading-edge information and communication technologies used to improve the safety, efficiency, and sustainability of transportation networks and reduce traffic congestion, thus improving drivers’ experiences.
Developers of Intelligent Transport Systems use the latest tech trends such as Computer Vision, Deep Learning, edge computing, IoT, etc., to power such solutions. So, how does an Intelligent Transport System work? Let’s find out.
Typically, the work on an Intelligent Transport System comprises different stages such as data collection, data annotation, data analysis, training, and testing AI models.
Let’s take a look at these stages in more detail, based on a real-life case study:
A company that develops intelligent traffic systems (ITS) for government, police, and traffic departments has partnered with N-iX. They needed substantial computer vision expertise (from discovery to implementation) for one of their projects.
Our experts are currently working on an outdoor Intelligent Transport System aimed to track whether there are no traffic rules offenses. The client has cameras located on specific road parts that detect the car, speed, color, and size. Our task was to develop models that will detect a specific anomaly based on the image received from the cameras.
As that is an outdoor system, we came across a few challenges that we needed to solve:
- Different weather conditions (rain, snow, sun, clouds)
- Different light conditions (day, night, evening)
- Windshield reflections (hard to see passengers, drivers)
- Left, right-sided cars (hard to detect a fastened seat belt)
An Intelligent Transport System: Key stages:
- Data collection:
All data collection for training models is performed on the client's side. The data (snapshots with different time intervals) is retrieved from Jetson TX2 devices (with connected cameras) and sent directly to AWS S3, where it is stored, and from which we upload the images for further annotation and analysis.
- Data annotation:
This stage can be both manual and automated and is required for further system training.
It typically consists of several phases:
- With the help of the CVAT tool, we perform manual annotation. The tool also allows uploading automatic annotation using models we already trained on smaller datasets. For instance, we annotate images according to the windshield/person detector, seatbelt/offense classifier. In simple terms, we annotate if there are any offenses on the images: eating, drinking, talking over the phone, excessive head rotation, hands on the wheel, or not.
- We manually annotate or fix automatic annotations.
- We download correct annotations in COCO format from CVAT.
- We use data analytics for Intelligent Transportation Systems to study the annotated data and use it for retraining models and making the results more accurate.
- Data analysis:
Data undergoes steps such as error rectification, data cleaning, data synthesis, and adaptive logical analysis during this stage. Inconsistencies in data are identified with specific software and rectified. After that, data is pooled for analysis and analyzed further to train models, predict traffic scenarios, and detect anomalies.
To ensure that the model is trained on accurate data, we analyze annotations using the Pandas tool. It is crucial to understand how many images we have collected, and how often specific objects (i.e., windshields, drivers, passengers are presented on the images, what the minimum/maximum size of these objects is. How many images we have with different cases(e.g., with a seatbelt fastened or not fastened, and whether the dataset with these cases is balanced).
- Image enhancement:
Also, we helped our client with image enhancement, as we work with outdoor cameras during different weather conditions and dayparts. So, we needed to reduce reflections, brightness, darkness, and other factors that affect image quality.
- Training, validating, and testing models:
- Collected, annotated, and analyzed data is used as the training dataset. Retraining models as the training dataset grows to increase the accuracy of models.
- Evaluating the accuracy of models on the testing dataset to see if they need to be retrained to provide more accurate results.
The development of an Intelligent Transport System requires strong expertise in many tech domains. So, what tech trends enable Intelligent Transport Systems?
- Deep Learning
Deep learning is a type of Machine Learning that is used for working with unstructured data, such as images and video. That’s why it is widely used in Intelligent transport systems for image detection (e.g., windshield, person detection) and classification(e.g., offense, seat belt) in Intelligent Transport Systems.
Here are some of the critical Deep Learning tools that we use in Intelligent transport systems:
Tensorflow Object Detection API and Transfer Learning Toolkit: for object detection (cars, pedestrians, traffic lights, etc.);
Keras: for tasks classification;
Deep Stream: allows the implementation of video surveillance systems on the road, but it is limited only to specific deep learning models and camera models. So it may not be a good fit in some cases.
- Computer Vision
The Computer Vision tool, OpenCV, is used for image enhancement and preprocessing (resizing, cropping, normalizing, removing distortions). Also, Computer Vision is needed for recognizing different weather conditions.
- Internet of Things
IoT is used for different purposes in Intelligent Transport Systems. For instance, an AI-powered embedded system Jetson TX2 is a single board computer to which a camera is connected. It is used for processing the data on-premise to ensure better security of personal data. What’s more, it doesn't require transferring large volumes of data and therefore can do with 4G. Only the processed results and alerts are sent to the cloud. That is why it is a perfect fit for Intelligent Transport Systems.
- Define goals and have a clear understanding of business KPIs.
N-iX offers a Discovery Stage to help you understand what to start with, the risks and challenges, how to overcome them, what tech stack and architecture are the best fit for your specific case (taking into account the types of cameras and hardware you are currently using). As a result, you mitigate potential risks and get all the deliverables to kick off the project successfully.
- When training models, ML engineers, and Data scientists need to have a clear understanding of the data and challenges associated with retrieving clean, valid, and complete data.
ML engineers and Data scientists need to ensure the dataset for training is balanced (each class should be present in the dataset in equal parts). For instance, to ensure a model detects correctly if the seatbelt is fastened or not, you need to have a balanced dataset to train it on. It means you need an equal number of images when:
- a seatbelt is fastened;
- a seatbelt is not fastened.
It allows training models that will be more accurate.
- Prepare a nice and clean dataset and use transfer learning. It is critical to focus on data quality, not quantity.
You can start training backbone models (available from Google, Amazon) without collecting thousands of data/images by yourself as they have already been trained on billions of data. And they can already detect and identify colors, patterns, lines, etc. All you need is to customize them and start training on custom data (even beginning with 100 images) and then retrain them as the pool of your images increases.
- Prototype fast and increase the training dataset step by step (for various conditions, in different locations), re-train models as the training dataset is growing, compare the accuracy of newly trained models to the previous ones, and retrain as much as you need to reach the required level of accuracy.
- Use efficient models to fit in memory and performance limits and use multitask learning to save memory and improve performance. It means that you can use a single backbone model as a feature extractor but with different tails for tasks (for example: do seatbelt/offense classification with a single model but different tails).
- Use the automation of pipelines and implement MLOps. That will allow you to implement continuously repeatable cycles of training, testing, deploying, monitoring, and operating the ML models, do more experiments, and increase the accuracy of models more time-and-cost efficiently.
- Build a PoC with baseline accuracy. Then, by using incremental learning, improve models gradually to reach a baseline accuracy of 100%.
It is also critical that you ensure a high security level on such projects, as the system deals with sensitive information such as vehicles’ registration numbers. So, how to achieve security on such projects?
How to ensure security in Intelligent Transport Systems
Security is of paramount importance when working on Intelligent Transport Systems. Here are some of the best practices we at N-iX follow:
- The team works remotely on AWS machines without an Internet connection. The only access on machines is to data (S3 Storage) and code (ssh-based connection to bitbucket repo).
- All packages are prepared and installed offline after the IT Security Team transfers data to the machine.
- We use Docker containers to isolate different environments.
Why choose N-iX for building an Intelligent Transport System?
- N-iX has strong expertise in IoT, Computer vision, Big Data, Data Science, and other technologies that enable Intelligent Transport Systems;
- Using our IoT expertise, we help our clients build IoT systems that boost operational efficiency and create connecting devices, sensors, and advanced analytical tool for smart homes, vehicles, personal devices, wearables, industrial factories, and more;
- N-iX has built long-term strategic partnerships (5+ years) with such industry leaders as Lebara, Gogo, Currencycloud, and many others;
- The company has robust cloud expertise and is an AWS partner, Google Cloud partner, and Azure partner;
- The company has 2,000+ top-notch professionals on board that can help you develop your Intelligent Transport System.