yolov5 ros 行人检测

### 使用YOLOv5在ROS中实现行人检测 #### 安装依赖项 为了在ROS环境中部署YOLOv5，需安装必要的Python库和其他工具。确保已安装`ros-noetic-desktop-full`或其他对应版本的ROS环境。 ```bash pip install torch torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/cu113 pip install rospkg catkin_pkg rospy message_generation std_msgs sensor_msgs cv_bridge opencv-python ``` #### 下载并配置YOLOv5仓库 克隆官方YOLOv5仓库，并设置好模型权重文件路径： ```bash git clone https://github.com/ultralytics/yolov5.git cd yolov5 pip install -r requirements.txt wget https://github.com/ultralytics/yolov5/releases/download/v5.0/yolov5s.pt ``` #### 创建ROS节点 创建一个新的ROS包来集成YOLOv5与ROS通信功能: ```bash catkin_create_pkg yolo_v5_node rospy std_msgs sensor_msgs cv_bridge ``` 编辑`CMakeLists.txt`和`package.xml`以包含适当依赖关系。 #### 编写ROS节点代码 编写一个简单的ROS Python脚本作为YOLOv5推理引擎接口[^1]: ```python #!/usr/bin/env python import rospy from sensor_msgs.msg import Image as RosImage from cv_bridge import CvBridge, CvBridgeError import cv2 import numpy as np import torch import time class YOLOV5Node(object): def __init__(self): self.bridge = CvBridge() self.image_subscriber = rospy.Subscriber("/camera/image_raw", RosImage, self.callback) self.model = torch.hub.load('ultralytics/yolov5', 'yolov5s') # 加载预训练模型 def callback(self, ros_image): try: frame = self.bridge.imgmsg_to_cv2(ros_image, "bgr8") except CvBridgeError as e: print(e) results = self.model(frame) # 执行推断操作 labels, confidences = [], [] for detection in results.xyxy[0]: label = int(detection[-1]) confidence = float(detection[-2]) if label == 0 and confidence >= 0.7: # 假设标签索引0代表'person' x_min, y_min, x_max, y_max = map(int, detection[:4].tolist()) # 绘制矩形框标记行人位置 cv2.rectangle(frame, (x_min,y_min), (x_max,y_max),(0,255,0), thickness=2) resized_frame = cv2.resize(frame,(int(frame.shape[1]/2), int(frame.shape[0]/2))) cv2.imshow("Detected Pedestrians",resized_frame) key=cv2.waitKey(1)&0xFF if(key==ord('q')): exit() if __name__=="__main__": rospy.init_node('yolov5_pedestrian_detector') node = YOLOV5Node() while not rospy.is_shutdown(): rospy.spin() ``` 此段程序订阅了摄像头话题获取原始图像流，利用加载好的YOLOv5模型处理每一帧图片，当发现置信度超过设定阈值的人类对象时，则绘制边框标注出来。最后显示带有标注的结果窗口给用户查看。