WPG launches AI image recognition vehicle identification solution based on NXP i.MX8QM

On May 13, 2021, WPG Holdings, a leading semiconductor component distributor dedicated to the Asia-Pacific market, announced that its subsidiary Pinjia has launched AI image recognition and vehicle recognition solutions based on NXP's i.MX8QM.

Today's society is gradually developing into an economic system centered on multimedia and highly dependent on data and automation. As an important part of the system, the automotive industry is also experiencing unprecedented intelligent upgrades under the promotion of many technologies. As autonomous driving and assisted driving technologies become more mature, how to help customers develop AI applications has become a new issue.

The AI ​​image recognition and vehicle identification solution based on NXP i.MX8QM launched by WPG uses the eIQ 2.0 software development environment, integrates multiple different algorithms and provides corresponding APIs for customers to develop and use. The core chip i.MX8QM of this solution can stably handle complex and heavy resource consumption such as image recognition, machine learning, and data computing.

Figure 1-Block diagram of the AI ​​image recognition and vehicle identification solution based on NXP i.MX8QM launched by GPX

The i.MX8QM chip is equipped with 4 A53 cores, 2 A72 cores and 2 built-in GC7000XSVX GPUs. It uses NXP's advanced technology and not only has a flexible and fast startup mechanism, but also provides display fault transfer function. In addition, this chip has passed ISO26262 and ASIL-B certification, so it can better ensure the safety level of the vehicle system.

This solution is developed based on NXP native BSP 5.4.24_2.1.0, adds Python elements, and can use GPU/NPU to improve the computing efficiency of AI neural networks, making the application scenarios more complete. In addition, this solution is also equipped with an integrated development environment for users to quickly enter the development field, and also provides currently popular open source learning modules such as OpenCV, TensorFlow, TensorFlowLite, caffe, etc., which can help users save development costs and development time.

Figure 2-Application diagram of AI image recognition and vehicle identification solutions launched by VP of Business Intelligence based on NXP i.MX8QM

Core technology advantages:

Automotive Gade，ASIL-B；

16x Vec4-Shader GPU，32 compute units OpenGL® ES 3.2 and Vulkan® support Tessellation and Geometry Shading；

2xArm A72 core + 4 A53 core；

MIPI CSI can connect to two high-definition cameras at the same time;

Pinjia provides cross-platform (PC to I.MX) ML (Machine Learning) applications.

Program Specifications:

Python 3.7；

TensorFlow 2.1；

TensorFlowLite 2.1；

OpenCV 4.2.0；

ArmNN 19.08。