深度学习框架yolov5训练森林火灾数据集 建立建立基于深度学习yolov5+pyqt 界面的森林火灾检测系统

基于yolov5的森林火灾识别系统，基于深度学习的森林火灾检测系统，森林火灾识别系统。森林火灾识别
以下文字及代码仅供参考。

项目构建基础：
模型：YOLOV5
软件：Pycharm+Anaconda
环境：python=3.10 opencv_python PyQt5

1

预实现功能：

系统实现了对于多种火灾的识别检测功能：包括通过选择图片、视频、摄像头、文件夹图片文件进行实时识别；检测速度快、识别精度较高。
①选择图片识别森林火灾。
②选择视频识别森林火灾。
③摄像头检测识别森林火灾。
文件夹图片文件识别森林火灾

基于YOLOv5的森林火灾识别系统是一个非常实用的项目，可以实时检测和识别森林火灾。以下是构建这个系统的详细步骤，包括数据准备、环境搭建、模型训练、模型推理等。

1. 环境搭建

安装依赖

确保安装了以下依赖：

conda create -n fire_detection python=3.10
conda activate fire_detection
pip install opencv-python pyqt5 yolov5

2. 数据准备

数据集

• 数据集：包含2000张图片的数据集，每张图片都有标注。
• 格式转换：将标注文件转换为YOLOv5所需的格式（.txt 文件）。

数据划分

• 使用 train_test_split 将数据集划分为训练集和验证集。

import os
import random
from sklearn.model_selection import train_test_split

def split_data(image_dir, label_dir, output_dir):
    images = [f for f in os.listdir(image_dir) if f.endswith('.jpg')]
    labels = [f for f in os.listdir(label_dir) if f.endswith('.txt')]

    # Ensure the same order for images and labels
    images.sort()
    labels.sort()

    # Split into train and test sets
    train_images, val_images, train_labels, val_labels = train_test_split(images, labels, test_size=0.2, random_state=42)

    # Save the splits
    with open(os.path.join(output_dir, 'train.txt'), 'w') as f:
        for img, lbl in zip(train_images, train_labels):
            f.write(f'{img} {lbl}\n')

    with open(os.path.join(output_dir, 'val.txt'), 'w') as f:
        for img, lbl in zip(val_images, val_labels):
            f.write(f'{img} {lbl}\n')

# Example usage
image_dir = 'data/images'
label_dir = 'data/labels'
output_dir = 'data/splits'
split_data(image_dir, label_dir, output_dir)

3. 模型训练

配置文件

创建一个配置文件 fire_detection.yaml：

train: data/train/images/
val: data/val/images/
test: data/test/images/

nc: 1  # number of classes
names: ['fire']

训练命令

使用以下命令进行训练：

python yolov5/train.py --img 640 --batch 16 --epochs 100 --data fire_detection.yaml --weights yolov5s.pt --name fire_detection

4. 模型推理

推理代码

创建一个Python脚本 detect_fire.py 进行推理：

import torch
import cv2
from yolov5.utils.general import non_max_suppression, scale_coords
from yolov5.utils.torch_utils import select_device

def detect_fire(image_path, model_path='runs/train/fire_detection/weights/best.pt'):
    device = select_device('')
    model = torch.load(model_path)['model'].float().to(device).eval()

    image = cv2.imread(image_path)
    image_rgb = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
    image_tensor = torch.from_numpy(image_rgb).permute(2, 0, 1).unsqueeze(0).float() / 255.0
    image_tensor = image_tensor.to(device)

    with torch.no_grad():
        pred = model(image_tensor)[0]
        pred = non_max_suppression(pred, 0.4, 0.5)

    for *xyxy, conf, cls in reversed(pred[0]):
        x1, y1, x2, y2 = map(int, xyxy)
        cv2.rectangle(image, (x1, y1), (x2, y2), (0, 255, 0), 2)
        cv2.putText(image, f'Fire: {conf:.2f}', (x1, y1 - 10), cv2.FONT_HERSHEY_SIMPLEX, 0.9, (0, 255, 0), 2)

    cv2.imshow('Fire Detection', image)
    cv2.waitKey(0)
    cv2.destroyAllWindows()

if __name__ == '__main__':
    image_path = 'path/to/image.jpg'
    detect_fire(image_path)

5. PyQt5 GUI

创建GUI界面

创建一个PyQt5界面用于选择图片、视频或摄像头进行实时检测：

import sys
from PyQt5.QtWidgets import QApplication, QWidget, QVBoxLayout, QPushButton, QLabel, QFileDialog, QComboBox, QLineEdit
from PyQt5.QtGui import QImage, QPixmap
from PyQt5.QtCore import QTimer
import cv2
import numpy as np

class FireDetectionApp(QWidget):
    def __init__(self):
        super().__init__()
        self.initUI()

    def initUI(self):
        self.setWindowTitle('Forest Fire Detection')
        layout = QVBoxLayout()

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

        self.source_combo = QComboBox(self)
        self.source_combo.addItem('Image')
        self.source_combo.addItem('Video')
        self.source_combo.addItem('Camera')
        layout.addWidget(self.source_combo)

        self.browse_button = QPushButton('Browse', self)
        self.browse_button.clicked.connect(self.browse_file)
        layout.addWidget(self.browse_button)

        self.detect_button = QPushButton('Detect', self)
        self.detect_button.clicked.connect(self.detect_fire)
        layout.addWidget(self.detect_button)

        self.setLayout(layout)

    def browse_file(self):
        file_dialog = QFileDialog()
        file_dialog.setNameFilter("Images (*.jpg *.jpeg *.png)")
        if self.source_combo.currentText() == 'Video':
            file_dialog.setNameFilter("Videos (*.mp4 *.avi *.mov)")
        if file_dialog.exec_():
            self.file_path = file_dialog.selectedFiles()[0]

    def detect_fire(self):
        if self.source_combo.currentText() == 'Image':
            self.detect_image()
        elif self.source_combo.currentText() == 'Video':
            self.detect_video()
        elif self.source_combo.currentText() == 'Camera':
            self.detect_camera()

    def detect_image(self):
        image = cv2.imread(self.file_path)
        image_rgb = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
        qimage = QImage(image_rgb.data, image_rgb.shape[1], image_rgb.shape[0], QImage.Format_RGB888)
        pixmap = QPixmap.fromImage(qimage)
        self.image_label.setPixmap(pixmap)

    def detect_video(self):
        cap = cv2.VideoCapture(self.file_path)
        while True:
            ret, frame = cap.read()
            if not ret:
                break
            frame_rgb = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
            qimage = QImage(frame_rgb.data, frame_rgb.shape[1], frame_rgb.shape[0], QImage.Format_RGB888)
            pixmap = QPixmap.fromImage(qimage)
            self.image_label.setPixmap(pixmap)
            self.update()

    def detect_camera(self):
        cap = cv2.VideoCapture(0)
        while True:
            ret, frame = cap.read()
            if not ret:
                break
            frame_rgb = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
            qimage = QImage(frame_rgb.data, frame_rgb.shape[1], frame_rgb.shape[0], QImage.Format_RGB888)
            pixmap = QPixmap.fromImage(qimage)
            self.image_label.setPixmap(pixmap)
            self.update()

if __name__ == '__main__':
    app = QApplication(sys.argv)
    ex = FireDetectionApp()
    ex.show()
    sys.exit(app.exec_())

6. 性能评估

使用标准的性能评估指标如准确率、召回率、F1分数等进行模型评估。

from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score

def evaluate_model(y_true, y_pred):
    accuracy = accuracy_score(y_true, y_pred)
    precision = precision_score(y_true, y_pred)
    recall = recall_score(y_true, y_pred)
    f1 = f1_score(y_true, y_pred)
    print(f'Accuracy: {accuracy:.2f}')
    print(f'Precision: {precision:.2f}')
    print(f'Recall: {recall:.2f}')
    print(f'F1 Score: {f1:.2f}')

# Example usage
y_true = [0, 1, 0, 1, 1, 0, 1, 0, 1, 1]
y_pred = [0, 1, 0, 1, 1, 0, 1, 0, 1, 1]
evaluate_model(y_true, y_pred)

仅供参考，学习