Welcome to Jetson Orin NANO Developer Kit repository
05.License plate detection and recognition
TAO LPDNet and LPRNet License Plate Detection and Recognition
TAO LPDNet and LPRNet License Plate Detection and Recognition1. Case Goal and Processing Flow2. Pre-Run Checks3. Starting Detection4. USB Camera Real-Time Detection4.1 Direct Run4.2 Common Parameters5. CSI Camera Real-Time Detection5.1 Direct Run5.2 Common Parameters6. Common Issues
Applicable environment: Jetson Orin Nano, JetPack 7.2.1, DeepStream 9.1
After setting the power mode to MAXN, running the case may pop up an over-current warning; just ignore it.
1. Case Goal and Processing Flow
TrafficCamNet first detects vehicles, LPDNet locates license plate regions, and LPRNet recognizes characters on the license plate crop. The Chinese and US versions use different models and dictionaries; choose according to the region.
video/USB/CSI → TrafficCamNet → LPDNet license plate detection → LPRNet character recognition → preview/video output2. Pre-Run Checks
xxxxxxxxxxcd ~/DeepStream/tao-toolkit-shell-demols -lh video_in/cn.mp4 video_in/us.mp4ls -ld video_outConfirm the material for the target region exists. Use the Jetson desktop/VNC for real-time cameras; the camera frame should have clear, sufficiently large vehicles and license plates.
3. Starting Detection
Chinese version:
xxxxxxxxxx./run_video.sh lpr --cn
US version:
xxxxxxxxxx./run_video.sh lpr --us
The corresponding inputs are video_in/cn.mp4 and video_in/us.mp4, and the outputs are video_out/lpr_cn.mp4 and video_out/lpr_us.mp4 respectively. The license plate boxes and character text represent the results of the localization and recognition stages respectively.
Replacing the test video: put the material into the input path used by the configuration according to the region, or check the input configuration and then modify the source URI. Keep the original file before replacing; use gst-discoverer-1.0 to check the media encoding. Do not only change the command parameters while still using another region's input material/dictionary.
4. USB Camera Real-Time Detection
4.1 Direct Run
The LPR USB entry uses the specified region parameter. First check the USB camera node and formats:
xxxxxxxxxxls -l /dev/video*cd ~/DeepStream/tao-toolkit-shell-demo./run_usb.sh lpr --cnFor US license plate testing, change the command parameter to --us. The live preview does not generate the above offline MP4; recognition quality depends on the camera resolution, shutter/motion blur, and the pixel size of the license plate in the frame.

Press Ctrl+C once and wait for the application to exit and the camera to be released.
4.2 Common Parameters
Before running, you can check the formats, resolutions, and frame rates supported by the camera:
xxxxxxxxxxv4l2-ctl -d /dev/video0 --list-formats-ext
Note: the USB pipeline of Object Embedding Vector Generation is fixed to MJPEG, so among the parameters below, choose the MJPEG parameters
x# View all parameters./run_usb.sh lpr --cn --help
# Use another video device./run_usb.sh lpr --cn --device /dev/video1
# Choose from the configurations found above that the camera supports; here we use 1280×720, 60FPS./run_usb.sh lpr --cn --width 1280 --height 720 --fps 605. CSI Camera Real-Time Detection
5.1 Direct Run
The course CSI entry supports live preview, using Jetson Argus instead of the USB V4L2 device path. Connect a compatible module and confirm the driver and Argus are working, then run:
xxxxxxxxxxcd ~/DeepStream/tao-toolkit-shell-demo./run_csi.sh lpr --usThe Chinese version uses --cn. CSI mode supports live preview and does not generate the above offline MP4.

5.2 Common Parameters
Before running, you can check the formats, resolutions, and frame rates supported by the camera; this must be run in a graphical interface:
xxxxxxxxxxgst-launch-1.0 nvarguscamerasrc
xxxxxxxxxx# View all parameters./run_csi.sh lpr --us --help
# Use the second CSI camera./run_csi.sh lpr --us --sensor-id 1
# Adjust input size and frame rate./run_csi.sh lpr --us --width 1920 --height 1080 --fps 306. Common Issues
| Symptom | Check and Handling |
|---|---|
| No license plate boxes | First confirm the vehicle detection boxes are normal, then check the license plate visibility, size, and LPDNet configuration. |
| License plate boxes but no text | Check the LPRNet engine, crop, and recognition stage logs. |
| Garbled characters/region mismatch | Verify --cn/--us, the model version, and dict_ch.txt/dict_us.txt. |
| No USB image | Check the node, format, permissions, occupancy, and display session. |
| CSI does not start | Check the compatible module connection, sensor driver, Argus service, and camera occupancy. |