深度学习目标检测使用yolov8训练使用路障类数据集之_道路积水检测数据集并构建一个基于YOLOv8的道路积水检测系统 来检测识别道路积水

道路积水检测数据集,23097张,yolo和voc在这里插入图片描述

1类,标注数量:
0 道路积水: 133261
image num: 23097在这里插入图片描述
积水啦,怎么办
在这里插入图片描述
构建基于YOLOv8的道路积水检测系统。两种标注方式(YOLO和VOC)。以下是完整的代码实现,包括数据加载、模型训练、评估和推理。

1. 安装依赖

首先,确保您已经安装了所需的库,特别是YOLOv8依赖的库。

pip install torch torchvision ultralytics pyqt5 opencv-python lxml

2. 数据准备

假设您的数据集已经按照YOLO和VOC的标准格式进行了标注。具体来说,数据集目录结构如下:

datasets/
└── road_flooding_detection/
    ├── images/
    │   ├── train/
    │   └── val/
    ├── labels_yolo/
    │   ├── train/
    │   └── val/
    └── annotations_voc/
        ├── train/
        └── val/
Font/
models/
runs/
save_data/
test-file/
TestFiles/
UIProgram/
2.py
CameraTest.py
CITATION.cff
Config.py
detect_tools.py
imgTest.py
installPackages.py
MainProgram.py
requirements.txt
setup.py
test.py
train.py
VideoTest.py
yolov8n.pt
使用教程.xt
项目文件说明.png

3. 文件内容

3.1 datasets/road_flooding_detection/ 目录

假设您的数据集已经按照YOLO和VOC标准格式进行了标注。具体来说,每个图像对应一个同名的.txt文件(YOLO格式)或XML文件(VOC格式),而标签文件是JSON格式的COCO标准注释文件。

3.2 Config.py

配置文件,用于定义数据集路径、模型路径等。

[<title="Config.py">]
# Config.py
DATASET_PATH = 'datasets/road_flooding_detection/'
MODEL_PATH = 'runs/detect/train/weights/best.pt'
IMG_SIZE = 640
BATCH_SIZE = 16
EPOCHS = 50
CONF_THRESHOLD = 0.5
3.3 train.py

训练YOLOv8模型的脚本。

[<title="train.py">]
from ultralytics import YOLO
import os

# Load a model
model = YOLO('yolov8n.pt')  # You can also use other versions like yolov8s.pt, yolov8m.pt, etc.

# Define dataset configuration
dataset_config = f"""
train: {os.path.join(os.getenv('DATASET_PATH', 'datasets/road_flooding_detection/'), 'images/train')}
val: {os.path.join(os.getenv('DATASET_PATH', 'datasets/road_flooding_detection/'), 'images/val')}
annotations:
  train: {os.path.join(os.getenv('DATASET_PATH', 'datasets/road_flooding_detection/'), 'labels_yolo/train')}
  val: {os.path.join(os.getenv('DATASET_PATH', 'datasets/road_flooding_detection/'), 'labels_yolo/val')}
"""

# Save dataset configuration to a YAML file
with open('road_flooding.yaml', 'w') as f:
    f.write(dataset_config)

# Train the model
results = model.train(data='road_flooding.yaml', epochs=int(os.getenv('EPOCHS', 50)), imgsz=int(os.getenv('IMG_SIZE', 640)), batch=int(os.getenv('BATCH_SIZE', 16)))
3.4 detect_tools.py

用于检测的工具函数。

[<title="detect_tools.py">]
from ultralytics import YOLO
import cv2
import numpy as np

def load_model(model_path):
    return YOLO(model_path)

def detect_objects(frame, model, conf_threshold=0.5):
    results = model(frame, conf=conf_threshold)
    detections = []
    for result in results:
        boxes = result.boxes.cpu().numpy()
        for box in boxes:
            r = box.xyxy[0].astype(int)
            cls = int(box.cls[0])
            conf = round(float(box.conf[0]), 2)
            label = f"{model.names[cls]} {conf}"
            detections.append((r, label))
    return detections

def draw_detections(frame, detections):
    for (r, label) in detections:
        cv2.rectangle(frame, (r[0], r[1]), (r[2], r[3]), (0, 255, 0), 2)
        cv2.putText(frame, label, (r[0], r[1] - 10), cv2.FONT_HERSHEY_SIMPLEX, 0.9, (0, 255, 0), 2)
    return frame
3.5 UIProgram/MainProgram.py

主程序,使用PyQt5构建图形界面。

[<title="UIProgram/MainProgram.py">]
import sys
import cv2
from PyQt5.QtWidgets import QApplication, QMainWindow, QLabel, QVBoxLayout, QWidget, QPushButton
from PyQt5.QtGui import QImage, QPixmap
from PyQt5.QtCore import Qt, QTimer
from detect_tools import load_model, detect_objects, draw_detections
import os

class VideoWindow(QMainWindow):
    def __init__(self):
        super().__init__()

        self.setWindowTitle("Road Flooding Detection")
        self.setGeometry(100, 100, 800, 600)

        self.central_widget = QWidget()
        self.setCentralWidget(self.central_widget)

        self.layout = QVBoxLayout()
        self.central_widget.setLayout(self.layout)

        self.label = QLabel()
        self.layout.addWidget(self.label)

        self.start_button = QPushButton("Start Detection")
        self.start_button.clicked.connect(self.start_detection)
        self.layout.addWidget(self.start_button)

        self.cap = None
        self.timer = QTimer()
        self.timer.timeout.connect(self.update_frame)

        self.model = load_model(os.getenv('MODEL_PATH', 'runs/detect/train/weights/best.pt'))

    def start_detection(self):
        if not self.cap:
            self.cap = cv2.VideoCapture(0)  # Use webcam
        self.timer.start(30)

    def update_frame(self):
        ret, frame = self.cap.read()
        if not ret:
            return

        detections = detect_objects(frame, self.model, conf_threshold=float(os.getenv('CONF_THRESHOLD', 0.5)))
        frame = draw_detections(frame, detections)

        rgb_image = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
        h, w, ch = rgb_image.shape
        bytes_per_line = ch * w
        qt_image = QImage(rgb_image.data, w, h, bytes_per_line, QImage.Format_RGB888)
        pixmap = QPixmap.fromImage(qt_image)
        self.label.setPixmap(pixmap.scaled(800, 600, Qt.KeepAspectRatio))

if __name__ == "__main__":
    app = QApplication(sys.argv)
    window = VideoWindow()
    window.show()
    sys.exit(app.exec_())
3.6 requirements.txt

列出所有依赖项。

[<title="requirements.txt">]
torch
torchvision
ultralytics
pyqt5
opencv-python
lxml
3.7 setup.py

用于安装项目的脚本。

[<title="setup.py">]
from setuptools import setup, find_packages

setup(
    name='road_flooding_detection',
    version='0.1',
    packages=find_packages(),
    install_requires=[
        'torch',
        'torchvision',
        'ultralytics',
        'pyqt5',
        'opencv-python',
        'lxml'
    ],
    entry_points={
        'console_scripts': [
            'train=train:main',
            'detect=UIProgram.MainProgram:main'
        ]
    }
)
3.8 README.md

项目说明文档。

[<title="README.md">]
# Road Flooding Detection System

This project uses YOLOv8 and PyQt5 to create a real-time road flooding detection system. The system detects road flooding in images and videos.

## Installation

1. Clone the repository:
   ```bash
   git clone https://github.com/yourusername/road-flooding-detection.git
   cd road-flooding-detection
  1. Install dependencies:

    pip install -r requirements.txt
    
  2. Set up environment variables (optional):

    export DATASET_PATH=./datasets/road_flooding_detection/
    export MODEL_PATH=./runs/detect/train/weights/best.pt
    export IMG_SIZE=640
    export BATCH_SIZE=16
    export EPOCHS=50
    export CONF_THRESHOLD=0.5
    

Training

To train the YOLOv8 model:

python train.py

Running the GUI

To run the graphical user interface:

python UIProgram/MainProgram.py

Usage Tutorial

See 使用教程.xt for detailed usage instructions.


### 4. 运行步骤

1. **确保数据集路径正确**:
   - 将您的数据集放在 `datasets/road_flooding_detection` 目录下。
   - 确保图像和对应的标签文件存在并且格式正确。

2. **安装必要的库**:
   - 确保您已经安装了所需的库,如 `torch`, `torchvision`, `ultralytics`, `pyqt5`, `opencv-python`, `lxml` 等。
   - 您可以使用以下命令安装这些库:
     ```bash
     pip install -r requirements.txt
     ```

3. **运行代码**:
   - 首先运行训练代码来训练YOLOv8模型:
     ```bash
     python train.py
     ```
   - 然后运行GUI代码来启动检测系统:
     ```bash
     python UIProgram/MainProgram.py
     ```

希望这些信息能帮助您顺利构建基于YOLOv8和PyQt5的道路积水检测系统
评论
添加红包

请填写红包祝福语或标题

红包个数最小为10个

红包金额最低5元

当前余额3.43前往充值 >
需支付:10.00
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包
实付
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明:

1.余额是钱包充值的虚拟货币,按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载,可以购买VIP、付费专栏及课程。

余额充值