Khadas Docs

Amazing Khadas, always amazes you!

User Tools

Site Tools



NPU Prebuilt Demo Usage

Prebuilt example demos for interacting with the Amlogic NPU using OpenCV4

Install OpenCV4

Update your system and install the OpenCV packages.

$ sudo apt update
$ sudo apt install libopencv-dev python3-opencv

Get NPU Demo

Get the demo source: khadas/aml_npu_demo_binaries

$ git clone --recursive

The NPU demo contains three examples:

  1. detect_demo - A collection of yolo series models for camera dynamic recognition.
  2. detect_demo_picture - A collection of yolo series models that identify pictures.
  3. inceptionv3 - Identify the inception model of the picture.

Inception Model

The inception model does not have any library dependencies and can be used as is.

Enter the inceptionv3 directory.

$ cd aml_npu_demo_binaries/inceptionv3
$ ls
dog_299x299.jpg  goldfish_299x299.jpg  imagenet_slim_labels.txt  VIM3  VIM3L

imagenet_slim_labels.txt is a label file. After the result is identified, the label corresponding to the result can be queried in this file.

Depending on your board, enter the VIM3 or VIM3L directory accordingly. Here is the VIM3 as an example.

$ ls aml_npu_demo_binaries/inceptionv3/VIM3
$ inceptionv3  inception_v3.nb  
$ cd aml_npu_demo_binaries/inceptionv3/VIM3
$ ./
Create Neural Network: 59ms or 59022us
Verify Graph: 0ms or 739us
Start run graph [1] times...
Run the 1 time: 20.00ms or 20497.00us
vxProcessGraph execution time:
Total   20.00ms or 20540.00us
Average 20.54ms or 20540.00us
--- Top5 ---
2: 0.833984
795: 0.009102
974: 0.003592
408: 0.002207
393: 0.002111

By querying imagenet_slim_labels.txt, the result is a goldfish, which is also correctly identified. You can use the method above to identify other images.

Yolo Series Model

Install and uninstall libraries

The yolo series models need to install the library into the system. Both the detect_demo and detect_demo_picture examples require this procedure.

You can follow the steps to either install or uninstall the libraries.

Install libraries:

$ cd aml_npu_demo_binaries/detect_demo_picture
$ sudo ./INSTALL

Uninstall libraries:

$ cd aml_npu_demo_binaries/detect_demo_picture
$ sudo ./UNINSTALL

Type Parameter Description

The type parameter is an input parameter that must be selected whether it is to use camera dynamic recognition or to recognize pictures. This parameter is mainly used to specify the running YOLO series model.

0 : yoloface model
1 : yolov2 model
2 : yolov3 model
3 : yolov3_tiny model
4 : yolov4 model

Operating Environment for NPU demo

NPU Demo can run in X11 Desktop or framebuffer mode, just select the corresponding demo to run.

  1. The demo with fb is running in framebuffer mode.
  2. The demo with x11 is running in X11 mode.

Demo examples


$ cd aml_npu_demo_binaries/detect_demo_picture
$ ls 
1080p.bmp  detect_demo_x11  detect_demo_fb  INSTALL  lib  nn_data  UNINSTALL

Command format of the picture.

$ cd aml_npu_demo_binaries/detect_demo_picture
$ ./detect_demo_xx -m <type parameter> -p <picture_path>

Here is an example of using OpenCV4 to call the yolov3 model to recognize pictures under x11.

$ cd aml_npu_demo_binaries/detect_demo_picture
$ ./detect_demo_x11 -m 2 -p 1080p.bmp

The results of the operation are as follows.



You should use the demo of usb to use the USB camera, and the demo of mipi to use the mipi camera.


Command format for camera dynamic recognition.

$ cd aml_npu_demo_binaries/detect_demo
$ ./detect_xx_xx -d <video node> -m <type parameter>

Here is an example of using OpenCV4 to call yolov3 in the x11 environment using a USB camera.

$ cd aml_npu_demo_binaries/detect_demo
$ ./detect_demo_x11_usb -d /dev/video1 -m 2

After turning on the camera, the recognition result will be displayed on the screen.


Last modified: 2023/09/11 09:58 by hyphop