Site Search

NXP processor actual device verification guide summary site

Introducing evaluation methods using NXP processor evaluation boards WEB content currently being serialized

NXP offers a variety of processors including i.MX 6/7, i.MX 8, and the latest i.MX 9. (List of main lineup)

Board Support Packages (BSP) for Arm Coretex-A provided by NXP are

There are "Embedded Linux", "Android", "Android Automotive", and "Windows 10 IoT Enterprise".

We provide easy-to-understand explanations in Japanese on how to verify using evaluation boards. Please access them from the list below.

 

Reference NXP site link

i.MX Software and Development Tools | NXP Semiconductors

NXP GitHub | NXP Semiconductors

1 - i.MX8M Plus EVK
How to run the Machine Learning demo using the Yocto Linux BSP image

The NXP 8MPLUSLPD4-PEVK Power Measurement Evaluation Kit provides a platform for comprehensive evaluation of the i.MX 8M Plus processor.
There are several Yocto Linux BSP images available for the evaluation kit.
This time, we will introduce the steps to demonstrate machine learning using a Yocto Linux BSP image that includes a package of machine learning functions.

2 - i.MX8M Plus EVK
Measuring Power Consumption with the BCU

The 8MPLUSLPD4-PEVK power measurement evaluation kit is equipped with hardware for measuring the power consumption of the i.MX 8M Plus processor, so we will introduce the procedure for measuring power consumption using the BCU (Board Control Utilities).

3 - i.MX8M Plus EVK
Posenet demo using TensorFlow Lite quantized model (GStreamer edition)

The Yocto Linux BSP image, which includes the machine learning function package, also implements a machine learning inference environment using the GStreamer Plugin.
This time, we will run a Posenet demo using pipeline processing with GStreamer.

4 - i.MX8M Plus EVK
Object Detection demo using TensorFlow Lite quantized model

The Yocto Linux BSP image that contains the machine learning functionality package also implements the TensorFlow Lite delegate.

This makes it possible to perform inference using the TensorFlow Lite Runtime API in Python, so we will try running a demo of Object Detection using the quantized model mobilenet_ssd_v2_coco_quant_postprocess.tflite.

5 - Hailo-8 as an AI accelerator
How to connect to i.MX8M Plus and build an AI system

This article introduces how to build an AI system by connecting Hailo-8 to the host CPU i.MX8M Plus as an AI accelerator. Two versions are available: Ubuntu OS and Yocto Linux OS.

If you are interested in the edge AI system that combines i.MX8M Plus and Hailo products, please take a look.

6 - ADLINK SMARC LEC-IMX8MP
How to build a Yocto Linux development environment with machine learning capabilities

ADLINK offers a variety of pre-built OS images for the SMARC LEC-IMX8MP based on NXP’s i.MX8M Plus. Unfortunately, however, they do not offer a pre-built OS image for Yocto with machine learning capabilities.


Therefore, we have compiled the Yocto Project build instructions for an image with machine learning capabilities for SMARC LEC-IMX8MP, so please give it a try.

To access NXP-related documents, you will need to register for My NXP. Please refer to the following website to register.

NXP Semiconductors' My NXP registration benefits and registration method -Macnica (macnica.co.jp)

NXP processor beginner guide summary site

Buy Evaluation Boards at Macnica Mouser

Inquiry / Quotation

If you have any questions about this product or would like a quote, please contact us using the form below.

NXP Semiconductors Manufacturer Information Top

If you want to go back to NXP Semiconductors Manufacturer Information Top, please click below.