Huawei’s New Sensor Patent Helps Improve Driving Safety

In the field of artificial intelligence, data is the key point to driving model learning and optimizing. Whether it is computer vision, natural language processing, or autonomous driving, datasets provide the necessary foundation for algorithms. high-quality data can not only improve the performance of algorithms, but also promote the whole industries innovation and development. By collecting and annotating large amounts of data, researchers can train out more accurate and intelligent models to achieve more efficient prediction and decision-making capabilities.

The automatic driving system is generally composed of cameras, radars and processors, and realizes automatic driving functions through the combination of software and hardware. However, the camera has a certain blind area of vision, and the detection range of the radar is also limited, so it is difficult to accurately detect all the surrounding obstacles.

Recently, Huawei published a new patent on autonomous driving technology, which integrates the distribution information of obstacles they perceive by fusing two types of sensors: camera and radar. After fusion, the areas the vehicle can pass through will be presented from the system in probabilistic form. Huawei’s new patent can make the vehicle more comprehensively perceive obstacles around the vehicle, thereby effectively improving the safety of autonomous driving technology.

Huawei’s patent principle can be roughly divided into two steps. The first step is that the system obtains the first probability distribution of obstacles through the camera, and the second step is to obtain the second probability distribution of obstacles according to the echo time and echo width of the radar echo signal. According to the joint analysis of the first probability distribution and the second probability distribution of the obstacles, the system obtains the drivable area of the vehicle represented by the probability, and the probability represents the probability that the vehicle is impassable.

Since its establishment, Nexdata has been deeply involved in the field of AI data services and has accumulated rich experience, and fully understands the importance of high-quality data to the AI industry. Nexdata provides multi-sensor data collection services, and supports 2D, 3D, and 2D-3D fusion data annotation through the self-developed data annotation platform. Our data solutions can help autonomous vehicles quickly complete the whole process of perception-information processing-decision-making, helping the research and development of autonomous driving technology.

Besides customized data collection, Nexdata provides ready-to-go datasets for autonomous driving applications, such as occupant behavior, facial expressions, drivers’ gaze tracking, gesture detection, identity authentication, and etc. Our data solutions could best assist the training of algorithms to accurately identify and analyze the identity information, intentions, and behaviors of drivers and passengers.

Multi-race — Driver Behavior Collection Data

The data includes multiple ages, multiple time periods and multiple races (Caucasian, Black, Indian). The driver behaviors includes dangerous behavior, fatigue behavior and visual movement behavior.

Passenger Behavior Recognition Data

The data includes multiple age groups, multiple time periods and multiple races (Caucasian, Black, Indian). The passenger behaviors include passenger normal behavior, passenger abnormal behavior(carsick behavior, sleepy behavior, lost items behavior).

Multi-race 7 Expressions Recognition Data

The data includes male and female. The age distribution ranges from child to the elderly. For each person, 7 images were collected. The data diversity includes different facial postures, different expressions, different light conditions and different scenes.

3D Faces Collection Data

The data is collected from 1,078 Chinese people. Each subject was collected once a week, 6 times in total, so the time span is 6 weeks.

18 Gestures Recognition Data

This data diversity includes multiple scenes, 18 gestures, 5 shooting angels, multiple ages and multiple light conditions. For annotation, gesture 21 landmarks (each landmark includes the attribute of visible and visible), gesture type and gesture attributes were annotated.

End

If you need data services, please feel free to contact us: info@nexdata.ai.

Facing with growing demand for data, companies and researchers need to constantly explore new data collection and annotation methods. AI technology can better cope with fast changing market demands only by continuously improving the quality of data. With the accelerated development of data-driven intelligent trends, we have reason to look forward to a more efficient, intelligent, and secure future.