By clicking “Accept”, you agree to the storing of cookies on your device to enhance site navigation, analyze site usage, and assist in our marketing efforts. View our Privacy Policy for more information.

Data annotation for urban mobility

July 6, 2022

What is urban mobility?

It refers to the varied forms of daily and recurrent movements within urban or city settlements. These include the various means of transportation. It excludes movements which go beyond the borders of that particular urban setting. Examples of these excluded transportation means are planes and boats.

What is data annotation?

Data annotation is the process of tagging or labelling various forms of data or content such as images, text, audio,or video. This is done for machine learning models to recognise them and in addition make predictions.

What are the types of data annotation?

Text Annotation

Text annotations refer to the tagging of data which highlight specific texts such as phrases or keywords. In some cases it also includes the tagging of word sentiments and texts. Words such as "excited" are used to train machines to analyse or recognise the sentiment behind them.

Image Annotation

Image annotation is the act of labelling an image in a training model for the purpose of training the machine models. The image annotation process begins with a manual annotation where the images are labelled. They are further processed by a machine learning model.

Video Annotation

It is the process of capturing various images in a video, with the use of frame-by-frame annotated lines. It makes it easy for the machines to recognise the moving objects.

Audio Annotation

It involves the classification of audio components which come from people, animals,the environment among others. It also involves the identification of sound origins. It is very useful for Natural Language Processing,chatboxes, and virtual assistants. An example is text-to-speech translation.

Key-Point Annotation

It involves tagging or labelling "key" locations on particular images with the use of dots. For example, in annotating a human face, the mouth,eyes,ears and other unique facial features are among the keypoints to be tagged or annotated. It allows machine learning models to recognise human faces as is in the case of facial unlocking. It is also used to recognise the expressed sentiment.

How does data annotation affect and improve urban mobility?

Automated Vehicle Driving

Automated vehicles make extensive use of video and image annotations. The calibrated sensors used help  vehicles to easily make visual interpretations of images; whether in motion or static. 

Automated Number Plate Recognition 

The advent of ANPR has brought a level of vigilance and carefulness on the roads. It is a fact that 90% of road crashes emerge from human errors. The ability of ANPR systems to capture number plates as well as vehicles simply by recognising how they are handled by the drivers instills a level of discipline among drivers. This further eases road safety for pedestrians. 

Parking Systems and Payments

The use of video annotations help to monitor vehicles entering or exiting a parking space and in addition record the period of parking for a vehicle. This system of monitoring helps to easily find or penalize drivers who do not follow the rules.

Weather Forecasting

Recent technology uses annotation tools to detect weather and climate patterns. These forecasts specially aid industries in knowing when to minimise or maximize their operations. Pedestrians will know when to use raincoats or umbrellas when commuting by foot. Commercial drivers will be in the known as to when to pick up passengers to certain locations within the community and when to put the movement on hold.

Challenges of  urban mobility 

Traffic Jams

Traffic jams have become common in the urban settings as more city-dweller continue to   buy cars. The rural-urban migration situation in almost every area means a boom in their populations. As more people make their income they will prefer to get their own means of transport if they can afford it rather than joining public transport. It leads to the increase of vehicle usage on the roads resulting in these jams.

Public Transport Crowding 

Usually the seating capacity of commercial vehicles is outweighed by the demand for it. 

Atmospheric Pollution 

The World Health Organisation estimates that 4.2 million deaths occur globally as a result of air pollution which in itself stems from many factors, especially human activity. Some of the natural causes are dust and volcanic eruptions. However the major causes of atmospheric pollution in the urban setting include industrial waste from factories. Factories use a lot of toxic chemicals in their activities. Further fossil fuel usage (including firewood, charcoal and kerosene stoves) for cooking is another menace. It is a major cause of cardiovascular and respiratory disease.

Best use case of data annotation for urban mobility

Automated vehicle driving

The features of automated vehicles and semi-automated vehicles have in a very useful way helped to cope with urban mobility.

Since many automated vehicles will depend on electricity to function it will in a long run avoid or lower the menace of air pollution which is produced by normal vehicles.

It also has the adaptive cruise control feature which is able to detect oncoming traffic and automatically set the right amount of speed the vehicle is supposed to move by. This is a remarkable feature which helps to avoid unforeseen danger.

The reverse park assistance feature of automated vehicles make use of installed cameras which are located at the rear of the vehicles. The camera is then able to project a wide angle view of the vicinity on the vehicle dashboard. In some cases the projected path of the vehicle is labelled as the steering wheel is turning. It is very useful in situations where the parking maneuvers are complicated

Light Detection and Ranging (LIDAR), with the use of lasers is able to create a 3D map of the environment around an automated vehicle in a timely and accurate manner helping the vehicle to prepare ahead.

Street sign recognition is able to recognise the various traffic signs on roads and what they depict. It includes signs which read    "No U-turn", among others. It then displays it to the driver using a screen found on the instrument cluster. It uses a forward-facing camera found behind the windshield for the sole purpose of finding and reading road signs.

Automatic emergency braking systems make use of cameras to apply scene classification in identifying an oncoming collision. In this situation it automatically slows down the vehicle movement by applying its brakes.

You might also like
this new related posts

Want to find out more
about AI as well as our Data Labeling tools and services?

Isahit has a wide range of solutions and tools that will help you train your algorithms. Click below to learn more!