realsense+dlib目标检测

最新推荐文章于 2024-06-17 00:13:16 发布

蒝莱茹茈

最新推荐文章于 2024-06-17 00:13:16 发布

阅读量962

点赞数 1

分类专栏：机器学习文章标签： ros realsense dlib 目标检测

本文链接：https://blog.csdn.net/qq_40077167/article/details/105256160

版权

本文档介绍了在win7+ubuntu18.04双系统环境下，利用ROS Melodic、librealsense、opencv和dlib进行目标检测的步骤。首先，详细讲解了ROS Melodic的安装，接着阐述了librealsense、opencv和dlib的安装过程。然后，指导如何使用imglab创建数据集，并用dlib训练及测试模型。最后，提供了现有模型的资源链接。

摘要由CSDN通过智能技术生成

realsense+dlib目标检测

1.环境

双系统 win7 + ubuntu18.04
ROS Melodic安装：http://wiki.ros.org/cn/melodic/Installation/Ubuntu
注：Melodic可以安装在 Ubuntu Artful (17.10)，Bionic (18.04 LTS)，Debian Stretch 和其他一些平台上。
Kinetic在Ubuntu Wily(15.10)和Ubuntu Xenial (16.04 LTS)可用。
Indigo在Ubuntu Trusty (14.04 LTS) 和其他平台可用。
libreasense（Intel® RealSense™ SDK 2.0）安装： https://github.com/IntelRealSense/librealsense/blob/master/doc/distribution_linux.md
opencv：https://docs.opencv.org/master/d2/de6/tutorial_py_setup_in_ubuntu.html
dlib：https://github.com/davisking/dlib

2.训练自己的模型

imglab制作数据集：https://github.com/davisking/dlib/tree/master/tools/imglab
训练及测试：https://github.com/davisking/dlib/blob/master/python_examples/train_object_detector.py
现有模型：https://github.com/davisking/dlib-models

3.代码

#!/usr/bin/python
import pyrealsense2 as rs
import numpy as np
import cv2
import time
import darknet as dn
import dlib
import rospy
from cv_bridge import CvBridge
import sensor_msgs
from object_msgs.msg import ObjectsInBoxes,ObjectInBox
import geometry_msgs
 
# Declare pointcloud object, for calculating pointclouds and texture mappings
pc = rs.pointcloud()
# We want the points object to be persistent so we can display the last cloud when a frame drops
points = rs.points()

 
pipeline = rs.pipeline()
config = rs.config()
config.enable_stream(rs.stream.depth, 640, 480, rs.format.z16, 15)
config.enable_stream(rs.stream.color, 640, 480, rs.format.bgr8, 15)
 
# Start streaming
pipe_profile = pipeline.start(config)
 
# Create an align object
# rs.align allows us to perform alignment of depth frames to others frames
# The "align_to" is the stream type to which we plan to align depth frames.
align_to = rs.stream.color
align = rs