apcc-geoslegend / anpr-with-yolov4

A.I. parking barrier gate using YOLOv4 and DynamiKontrol module.

A.I. Parking Barrier Gate

Usage

Download the pretrained model from:

https://drive.google.com/file/d/1b3rYgP48z_NGvSuNoMKDvxXzYmray_Qr/view?usp=sharing

and run

python parking_gate.py

Dependency

• Python 3
• OpenCV 4+ (cv2.dnn module required)
• DynamiKontrol https://dk.m47rix.com

Reference

• Thanks to Junggyun for sharing pretrained models and code.

---

ANPR-with-Yolo-v4

ANPR (Automatic Number-Plate Recognition) : 차량번호판 자동 인식 프로그램

Yolo (You Only Look Once) : One-Stage Object Detector

About Darknet : http://pjreddie.com/darknet/

Download Model

• .weights : https://drive.google.com/file/d/1b3rYgP48z_NGvSuNoMKDvxXzYmray_Qr/view?usp=sharing
• .mlmodel : https://drive.google.com/file/d/1eREdAVoiOVlAiPOxv5c_Sc5EBPybOIUq/view?usp=sharing

Classes

• car
• license_plate

Training

Labeling Tool : https://github.com/AlexeyAB/Yolo_mark

Darknet (Yolov4) : https://github.com/AlexeyAB/darknet

Cloud Service	GPU	Traing Data	훈련 횟수	시간
GCP(Google Cloud Platform)	Nvidia Tesla P100	2600여장	4000회	5h

./darknet detector train data/obj.data cfg/yolov4_ANPR.cfg yolov4.conv.137 -gpu 0

Usage (test)

1. git clone https://github.com/AlexeyAB/darknet
2. cd darknet
3. 사용하는 환경에 맞게 Makefile 설정 vi Makefile

GPU=0		# GPU 사용 시 1로 변경
CUDNN=0		# cuDNN 사용 시 1로 변경 (NVIDIA)
CUDNN_HALF=0
OPENCV=0	# OpenCV 사용 시 1로 변경
AVX=0
OPENMP=0
LIBSO=1         # libdarknet.so 생성

...
...

4. make

• 기본 패키지 : make, gcc, pkg-config (없다면 sudo apt-get install …로 설치)

5. data/*, cfg/yolov4-ANPR.cfg, backup/yolov4-ANPR.weights 다운로드

image

./darknet detector test data/obj.data cfg/yolov4-ANPR.cfg backup/yolov4-ANPR.weights data/(이미지파일.jpg)

반드시 .jpg 이미지 사용

video

./darknet detector demo data/obj.data cfg/yolov4-ANPR.cfg backup/yolov4-ANPR.weights data/(동영상파일.mp4)

webcam

./darknet detector demo data/obj.data cfg/yolov4-ANPR.cfg backup/yolov4-ANPR.weights

Example

Prediction Image

./darknet detector test data/obj.data cfg/yolov4-ANPR.cfg backup/yolov4-ANPR.weights data/testfile.jpg

Loading weights from backup/yolov4-ANPR.weights...
 seen 64, trained: 256 K-images (4 Kilo-batches_64)
Done! Loaded 162 layers from weights-file
data/testfile.jpg: Predicted in 9325.005000 milli-seconds.
car: 63%
car: 98%
license_plate: 96%
car: 47%
car: 61%
car: 30%

Prediction Video

• ./darknet detector demo data/obj.data cfg/yolov4-ANPR.cfg backup/yolov4-ANPR.weights data/testvideo.jpg
• python darknet_video.py

Demo Video Link (1) : https://drive.google.com/file/d/1DGmF2bwtDMe1y-wNuv_YT827Vr6Y8Q2m/view?usp=sharing

Demo Video Link (2) : https://drive.google.com/file/d/1nJjIQFcrYRYSJ0n9FK0-x_Fk6HrULsZY/view?usp=sharing

References

• Paper
  • You Only Look Once : Unified, Real-Time Object Detection
  • YOLO9000 : Better, Faster, Stronger
  • YOLOv3 : An Incremental Improvement
  • YOLOv4：Optimal Speed and Accuracy of Object Detection
• Keras-yolov3
  • weights to h5 : https://github.com/qqwweee/keras-yolo3/blob/master/convert.py
  • weights to mlmodel : https://gist.github.com/TakaoNarikawa/aef13571eec97d78603688eef05b5389
  • Mish : https://qiita.com/TakaoNarikawa/items/e4521fd8c7a522e9d4fd
• Core ML
  • https://gist.github.com/TakaoNarikawa
  • https://github.com/Ma-Dan/YOLOv3-CoreML

Presentation

• 발표자료 : https://drive.google.com/file/d/1yhhIZ0ZU5MIZZar-WTBGgEHwLkOHcewe/view?usp=sharing

• 발표영상 : https://www.youtube.com/watch?v=H3-SVf0Ps4c

• Capstone Design : https://github.com/kwanghoon/CapstoneDesign