Sample For Car License Recognization

• Description
• Performance
• Prerequisition
• Download
• Build and Run
• Notice

Description

This sample is to show how to use graded models for detection and classification with DeepStream SDK version not less than 5.0.1. The models in this sample are all TAO3.0 models.

PGIE(car detection) -> SGIE(car license plate detection) -> SGIE(car license plate recognization)

This pipeline is based on three TAO models below

• Car detection model https://ngc.nvidia.com/catalog/models/nvidia:tao:trafficcamnet
• LPD (car license plate detection) model https://ngc.nvidia.com/catalog/models/nvidia:tao:lpdnet
• LPR (car license plate recognization/text extraction) model https://ngc.nvidia.com/catalog/models/nvidia:tao:lprnet

More details for TAO3.0 LPD and LPR models and TAO training, please refer to TAO document.

Performance

Below table shows the end-to-end performance of processing 1080p videos with this sample application.

Prerequisition

• DeepStream SDK 5.0.1

  Make sure deepstream-test1 sample can run successful to verify your DeepStream installation

• tao-converter

  Download x86 or Jetson tao-converter which is compatible to your platform from the following links.

Download

1. Download Project with SSH or HTTPS

    // SSH
    git clone [email protected]:NVIDIA-AI-IOT/deepstream_lpr_app.git
    // or HTTPS
    git clone https://github.com/NVIDIA-AI-IOT/deepstream_lpr_app.git

2. Prepare Models and TensorRT engine

For DS 6.0 version, this step can be ignored.

    cd deepstream_lpr_app/

For US car plate recognition

    ./download_us.sh
    // DS5.0.1 gst-nvinfer cannot generate TRT engine for LPR model, so generate it with tao-converter
    ./tao-converter -k nvidia_tlt -p image_input,1x3x48x96,4x3x48x96,16x3x48x96 \
           models/LP/LPR/us_lprnet_baseline18_deployable.etlt -t fp16 -e models/LP/LPR/us_lprnet_baseline18_deployable.etlt_b16_gpu0_fp16.engine

For Chinese car plate recognition

    ./download_ch.sh
    // DS5.0.1 gst-nvinfer cannot generate TRT engine for LPR model, so generate it with tao-converter
    ./tao-converter -k nvidia_tlt -p image_input,1x3x48x96,4x3x48x96,16x3x48x96 \
           models/LP/LPR/ch_lprnet_baseline18_deployable.etlt -t fp16 -e models/LP/LPR/ch_lprnet_baseline18_deployable.etlt_b16_gpu0_fp16.engine

Build and Run

    make
    cd deepstream-lpr-app

For US car plate recognition

    cp dict_us.txt dict.txt

For Chinese car plate recognition

    cp dict_ch.txt dict.txt

Start to run the application

    ./deepstream-lpr-app <1:US car plate model|2: Chinese car plate model> \
         <1: output as h264 file| 2:fakesink 3:display output> <0:ROI disable|1:ROI enable> \
         <input mp4 file name> ... <input mp4 file name> <output file name>

A sample of US car plate recognition:

./deepstream-lpr-app 1 2 0 us_car_test2.mp4 us_car_test2.mp4 output.264

A sample of Chinese car plate recognition:

./deepstream-lpr-app 2 2 0 ch_car_test.mp4 ch_car_test.mp4 output.264

Notice

1. This sample application only support mp4 files which contain H264 videos as input files.
2. For Chinese plate recognition, please make sure the OS supports Chinese language.
3. The second argument of deepstream-lpr-app should be 2(fakesink) for performance test.
4. The trafficcamnet and LPD models are all INT8 models, the LPR model is FP16 model.