ros+kinetic实现人脸识别

目录

１.使用电脑摄像头进行人脸识别

1.1  安装usb_cam

1.2  安装robot_vision

2.使用Kinect V2实现人脸识别

2.1 安装Kinect 驱动

2.2 运行kinect驱动程序

2.3  修改程序

2.4 运行程序

---

１.使用电脑摄像头进行人脸识别

1.1  安装usb_cam

sudo apt-get install ros-kinetic-usb-cam

1.2  安装robot_vision

(1)下载robot_vision安装包

git clone https://github.com/1417265678/robot_vision

(2)将安装包解压后放进工作空间catkin_ws/src内，进行编译

(3)运行代码

$ roslaunch robot_vision usb_cam.launch
$ roslaunch robot_vision face_detector.launch
$ rqt_image_view

在rqt_image_view中选择/cv_bridge_image主题

注意：在运行的时候可能会出现以下的问题

ERROR: cannot launch node of type [robot_vision/face_detector.py]:
 can't locate node [face_detector.py] in package [robot_vision]

解决方法：

(1)source一下工作空间

(2)给python文件权限

sudo chmod +x robot_vision/face_detector.py

 

如果出现问题：

ERROR: Could not load classifier cascade
OpenCV Error: Assertion failed (!empty()) in detectMultiScale, file /tmp/binarydeb/ros-kinetic-opencv3-3.3.1/modules/objdetect/src/cascadedetect.cpp, line 1698
terminate called after throwing an instance of 'cv::Exception'
  what():  /tmp/binarydeb/ros-kinetic-opencv3-3.3.1/modules/objdetect/src/cascadedetect.cpp:1698: error: (-215) !empty() in function detectMultiScale

参考文章：https://blog.csdn.net/qq_41906592/article/details/114894380?spm=1001.2014.3001.5501

 

2.使用Kinect V2实现人脸识别

2.1 安装Kinect 驱动

按照我前面的一篇文章，安装KInect 的驱动。

https://blog.csdn.net/qq_41906592/article/details/112231524

2.2 运行kinect驱动程序

$ roslaunch kinect2_bridge kinect2_bridge.launch
$ rosrun kinect2_viewer kinect2_viewer

2.3  修改程序

打开robot_vision文件夹，位置：robot_vison/launch/face_detector.launch

<launch>
    <node pkg="robot_vision" name="face_detector" type="face_detector.py" output="screen">
        <remap from="input_rgb_image" to="/usb_cam/image_raw" />
        <rosparam>
            haar_scaleFactor: 1.2
            haar_minNeighbors: 2
            haar_minSize: 40
            haar_maxSize: 60
        </rosparam>
        <param name="cascade_1" value="$(find robot_vision)/data/haar_detectors/haarcascade_frontalface_alt.xml" />
        <param name="cascade_2" value="$(find robot_vision)/data/haar_detectors/haarcascade_profileface.xml" />
    </node>
</launch>

将第三行修改为：

<remap from="input_rgb_image" to="/kinect2/qhd/image_color" />

2.4 运行程序

$ roslaunch robot_vision face_detector.launch
$ rqt_image_view

在rqt_image_view中选择/cv_bridge_image主题。如果操作无误，可现实如下内容