ONNX 模型库
返回模型

说明文档

突尼斯车牌 - 阿拉伯文本检测 (YOLOv8s)

该模型使用 YOLOv8s 检测突尼斯车牌中的阿拉伯文字 "تونس"（突尼斯）。

模型描述

  • 模型类型：YOLOv8s（小型）
  • 任务：目标检测
  • 类别：1 个类别 - "tunis"（阿拉伯文本区域）
  • 用途：检测和定位突尼斯车牌中的 "تونس" 文字，用于 OCR 预处理

使用场景

该模型设计用于车牌 OCR 的预处理步骤：

  1. 检测阿拉伯文本 "تونس" 区域
  2. 遮罩或裁剪该区域
  3. 对剩余的数字字符进行 OCR，以获得更高的准确率

训练详情

  • 基础模型：YOLOv8s 预训练权重
  • 图像尺寸：512x512
  • 框架：Ultralytics YOLOv8
  • 训练数据集：突尼斯车牌图像

使用方法

from ultralytics import YOLO

# Load the model
model = YOLO("yassine-mhirsi/tunis-word-detection-yolov8s")

# Run inference
results = model.predict("path/to/license_plate.jpg", conf=0.5)

# Process results
for result in results:
    boxes = result.boxes
    for box in boxes:
        print(f"Confidence: {box.conf[0]:.2f}")
        print(f"Bounding Box: {box.xyxy[0]}")

模型文件

  • best.pt - 训练得到的最佳权重
  • last.pt - 最后的检查点
  • 包含训练指标和可视化结果

示例

突尼斯车牌

引用

如果您使用此模型，请引用：

@misc{tunis-word-detection-yolov8s,
  authors = {Yassine Mhirsi,Malek Messaoudi},
  title = {Tunisian License Plate Arabic Text Detection},
  year = {2025},
  publisher = {Hugging Face},
  howpublished = {\url{https://huggingface.co/Safe-Drive-TN/tunis-word-detection-yolov8s/}}
}

许可证

MIT 许可证

yassine-mhirsi/tunis-word-detection-yolov8s

作者 yassine-mhirsi

object-detection ultralytics
↓ 1 ♥ 0

创建时间: 2025-12-03 09:12:13+00:00

更新时间: 2025-12-03 09:12:14+00:00

在 Hugging Face 上查看

文件 (73)

.gitattributes
README.md
runs/detect/predict/0_jpg.rf.2df34217a35c8b0e86730cdcdd635647.jpg
runs/detect/predict/100_jpg.rf.aaedf56d49016b0342c4b68049cbc360.jpg
runs/detect/predict/136_jpg.rf.1bfc19834f5938efa4455c2c38472f89.jpg
runs/detect/predict/13_jpg.rf.5993150c8babb983de007300c6ba649f.jpg
runs/detect/predict/14_jpg.rf.61a1d074477c285e6c76c7bd400aaeae.jpg
runs/detect/predict/16_jpg.rf.c29e176417d3ea9e48cf92f2663fa748.jpg
runs/detect/predict/18_jpg.rf.e4db82008cd0a7f3c39b6612de5c7e6f.jpg
runs/detect/predict/24_jpg.rf.1f5327bfa02b91d10b80138088b79fcf.jpg
runs/detect/predict/28_jpg.rf.80de86c719ceb0cc25a09ef0b4587231.jpg
runs/detect/predict/32_jpg.rf.2566677f3441af250fd2cb1ac40fb879.jpg
runs/detect/predict/33_jpg.rf.172accaf395df0743f03a186f8f8eb46.jpg
runs/detect/predict/37_jpg.rf.6c98f266ed845fa99fa822e1529a84fb.jpg
runs/detect/predict/39_jpg.rf.89a0593255f340fc8bb22adc5d00e9c7.jpg
runs/detect/predict/40_jpg.rf.12f226a09f865560f149ade9bb6ea5f9.jpg
runs/detect/predict/44_jpg.rf.ffa28229298a27604283fa1428dc851a.jpg
runs/detect/predict/48_jpg.rf.2a2485afc5e60f76b024c8d7d5d073fc.jpg
runs/detect/predict/52_jpg.rf.3c7c3e3ac6dfcfed36a2c6b223ba2319.jpg
runs/detect/predict/57_jpg.rf.21756804e9e44f9aec2f10f72dc11f95.jpg
runs/detect/predict/58_jpg.rf.4edbe7b70313bd0f705fffade17d4bf8.jpg
runs/detect/predict/62_jpg.rf.aef6430f2773087680be2ebd723d5cc4.jpg
runs/detect/predict/65_jpg.rf.d7b492d8d839a28abec02bddb298b7a6.jpg
runs/detect/predict/68_jpg.rf.865e177a226867fa7516fec94e14345f.jpg
runs/detect/predict/73_jpg.rf.a230d99b28262930899c9de25ba24aba.jpg
runs/detect/predict/84_jpg.rf.da99f1243eb7d2e5debf0b4195d9be61.jpg
runs/detect/predict/95_jpg.rf.e1f35786aca1d1f18b5669077e8eebe0.jpg
runs/detect/val/BoxF1_curve.png
runs/detect/val/BoxPR_curve.png
runs/detect/val/BoxP_curve.png
runs/detect/val/BoxR_curve.png
runs/detect/val/confusion_matrix.png
runs/detect/val/confusion_matrix_normalized.png
runs/detect/val/val_batch0_labels.jpg
runs/detect/val/val_batch0_pred.jpg
runs/detect/val/val_batch1_labels.jpg
runs/detect/val/val_batch1_pred.jpg
runs/detect/val/val_batch2_labels.jpg
runs/detect/val/val_batch2_pred.jpg
runs/detect/val2/BoxF1_curve.png
runs/detect/val2/BoxPR_curve.png
runs/detect/val2/BoxP_curve.png
runs/detect/val2/BoxR_curve.png
runs/detect/val2/confusion_matrix.png
runs/detect/val2/confusion_matrix_normalized.png
runs/detect/val2/val_batch0_labels.jpg
runs/detect/val2/val_batch0_pred.jpg
runs/detect/val2/val_batch1_labels.jpg
runs/detect/val2/val_batch1_pred.jpg
tunis_detector/yolov8s_run/BoxF1_curve.png
tunis_detector/yolov8s_run/BoxPR_curve.png
tunis_detector/yolov8s_run/BoxP_curve.png
tunis_detector/yolov8s_run/BoxR_curve.png
tunis_detector/yolov8s_run/args.yaml
tunis_detector/yolov8s_run/confusion_matrix.png
tunis_detector/yolov8s_run/confusion_matrix_normalized.png
tunis_detector/yolov8s_run/labels.jpg
tunis_detector/yolov8s_run/results.csv
tunis_detector/yolov8s_run/results.png
tunis_detector/yolov8s_run/train_batch0.jpg
tunis_detector/yolov8s_run/train_batch1.jpg
tunis_detector/yolov8s_run/train_batch2.jpg
tunis_detector/yolov8s_run/val_batch0_labels.jpg
tunis_detector/yolov8s_run/val_batch0_pred.jpg
tunis_detector/yolov8s_run/val_batch1_labels.jpg
tunis_detector/yolov8s_run/val_batch1_pred.jpg
tunis_detector/yolov8s_run/val_batch2_labels.jpg
tunis_detector/yolov8s_run/val_batch2_pred.jpg
tunis_detector/yolov8s_run/weights/best.onnx ONNX
tunis_detector/yolov8s_run/weights/best.pt
tunis_detector/yolov8s_run/weights/last.pt
yolo11n.pt
yolov8s.pt