基於ROS實現人臉跟隨（OpenCV，arduino）
摘要：
該測試欲達成目標是實現人臉跟隨！
1 硬體
攝像頭：1個;
USB資料線：1個；
舵機：1個；
Arduino控制板：1個。
此處應有圖：
1.1 準備
這裡需要使用usb-cam軟體包：
$ cd ~/catkin_ws/src...
該測試欲達成目標是實現人臉跟隨！
1 硬體
攝像頭：1個;
USB資料線：1個；
舵機：1個；
Arduino控制板：1個。
此處應有圖：
1.1 準備
這裡需要使用usb-cam軟體包：
$ cd ~/catkin_ws/src $ git clone https://github.com/bosch-ros-pkg/usb_cam.git $ cd ~/catkin_ws $ catkin_make
注意：下面的程式執行前，需要啟動usb_cam節點。
roslaunch usb_cam usb_cam-test.launch
執行上面的節點，這時該節點會不斷的釋出/usb_cam/image_raw話題。如下圖所示：
下面的程式將使用這個話題資料。
2 程式
2.1 人臉識別節點，釋出一個話題"chatter",為人臉在影象的位置。
#include <ros/ros.h>
#include <image_transport/image_transport.h>
#include <cv_bridge/cv_bridge.h>
#include <sensor_msgs/image_encodings.h>
#include <std_msgs/Int16.h>
#include <opencv2/objdetect.hpp>
#include <opencv2/imgproc/imgproc.hpp>
#include <opencv2/highgui/highgui.hpp>
#include <opencv2/core.hpp>
using namespace std;
using namespace cv;
CascadeClassifier face_cascade;
static const std::string OPENCV_WINDOW = "Raw Image window";
class Face_Detector
{
ros::NodeHandle nh_;
image_transport::ImageTransport it_;
image_transport::Subscriber image_sub_;
ros::Publisher chatter_pub;
public:
Face_Detector()
: it_(nh_)
{
// Subscribe to input video feed and publish output video feed
image_sub_ = it_.subscribe("/usb_cam/image_raw", 1,
&Face_Detector::imageCb, this);
chatter_pub = nh_.advertise<std_msgs::Int16>("chatter", 100);
cv::namedWindow(OPENCV_WINDOW);
}
~Face_Detector()
{
cv::destroyWindow(OPENCV_WINDOW);
}
void imageCb(const sensor_msgs::ImageConstPtr& msg)
{
cv_bridge::CvImagePtr cv_ptr;
namespace enc = sensor_msgs::image_encodings;
try
{
cv_ptr = cv_bridge::toCvCopy(msg, sensor_msgs::image_encodings::BGR8);
}
catch (cv_bridge::Exception& e)
{
ROS_ERROR("cv_bridge exception: %s", e.what());
return;
}
// Draw an example circle on the video stream
if (cv_ptr->image.rows > 400 && cv_ptr->image.cols > 600){
detect_faces(cv_ptr->image);
//chatter_pub.publish(position_x);
}
}
void detect_faces(cv::Mat img)
{
RNG rng( 0xFFFFFFFF );
std_msgs::Int16 position_x;
int data;
int lineType = 8;
Mat frame_gray;
cvtColor( img, frame_gray, COLOR_BGR2GRAY );
equalizeHist( frame_gray, frame_gray );
//-- Detect faces
std::vector<Rect> faces;
face_cascade.detectMultiScale( frame_gray, faces );
if (faces.size()>=1){
size_t i = 0;
Point center( faces[i].x + faces[i].width/2, faces[i].y + faces[i].height/2 );
ellipse( img, center, Size( faces[i].width/2, faces[i].height/2 ), 0, 0, 360, Scalar( 255, 0, 255 ), 4 );
data = cvRound(faces[i].x + faces[i].width/2);
position_x.data = data;
chatter_pub.publish(position_x);
}
imshow(OPENCV_WINDOW,img);
waitKey(3);
}
};
int main(int argc, char** argv)
{
//從命令列讀取必要的資訊,注意路徑
CommandLineParser parser(argc, argv,
"{help h||}"
"{face_cascade|/home/junjun/projects/main2/haarcascade_frontalface_alt.xml|Path to face cascade.}");
parser.about( "\nThis program demonstrates using the cv::CascadeClassifier class to detect objects (Face + eyes) in a video stream.\n"
"You can use Haar or LBP features.\n\n" );
parser.printMessage();
String face_cascade_name = parser.get<String>("face_cascade");
//-- 1\. Load the cascades
if( !face_cascade.load( face_cascade_name ) )
{
cout << "--(!)Error loading face cascade\n";
return -1;
};
ros::init(argc, argv, "Face_Detector");
Face_Detector ic;
ros::spin();
return 0;
}
2.1.1 CMakeLists.txt檔案
這個很重要，以後再寫程式可以直接參考這個檔案
cmake_minimum_required(VERSION 2.8.3)
project(cv_bridge_tutorial_pkg)
find_package(catkin REQUIRED COMPONENTS
cv_bridge
image_transport
roscpp
sensor_msgs
std_msgs
)
find_package( OpenCV REQUIRED )
catkin_package()
include_directories(
${catkin_INCLUDE_DIRS}
${OpenCV_INCLUDE_DIRS}
)
add_executable(sample_cv_bridge_node src/sample_cv_bridge_node.cpp)
target_link_libraries(sample_cv_bridge_node
${catkin_LIBRARIES}
${OpenCV_LIBRARIES}
)
include_directories(${catkin_INCLUDE_DIRS} ${OpenCV_INCLUDE_DIRS})
2.1.2 package.xml檔案
<?xml version="1.0"?> <package> <name>cv_bridge_tutorial_pkg</name> <version>0.0.0</version> <description>The cv_bridge_tutorial_pkg package</description> <maintainer email="[email protected]">lentin</maintainer> <license>TODO</license> <buildtool_depend>catkin</buildtool_depend> <build_depend>cv_bridge</build_depend> <build_depend>image_transport</build_depend> <build_depend>roscpp</build_depend> <build_depend>sensor_msgs</build_depend> <build_depend>std_msgs</build_depend> <run_depend>cv_bridge</run_depend> <run_depend>image_transport</run_depend> <run_depend>roscpp</run_depend> <run_depend>sensor_msgs</run_depend> <run_depend>std_msgs</run_depend> <export> </export> </package>
2.2 人臉跟隨的控制程式（arduino）
訂閱話題"chatter"
#include <ros.h>
#include <std_msgs/Int16.h>
#include <Servo.h>
ros::NodeHandle nh;
Servo myservo;
int servoPin=9;
int servo_step_distance = 5; //此外可以修改，太大容易超調，不容易穩定；太小，舵機不動；
int position_x = 90;
int screenmaxx = 640;
int center_offset = 80;
int center_right = (screenmaxx / 2) - center_offset;
int center_left = (screenmaxx / 2) + center_offset;
void messageCb(std_msgs::Int16 servo_msg){
digitalWrite(13,HIGH);
int position = servo_msg.data;
//當人臉位於左側，舵機向左轉。
if (position > center_left){
position_x -= servo_step_distance;
if (position_x >= 10 and position_x <=170)
myservo.write(position_x);
}
//當人臉位於右側，舵機向右轉。
else if (position < center_right){
position_x += servo_step_distance;
if (position_x >= 10 and position_x <=170)
myservo.write(position_x);
}
delay(1);
digitalWrite(13,LOW);
}
ros::Subscriber<std_msgs::Int16> sub("chatter", &messageCb );
void setup()
{
pinMode(13, OUTPUT);
myservo.attach(servoPin);
nh.initNode();
nh.subscribe(sub);
myservo.write(position_x);// tell servo to go to position in variable 'pos'
}
void loop()
{
nh.spinOnce();
delay(1);
}
把程式燒入Arduino板子中。
3 執行
首先執行：roscore
然後：roslaunch usb_cam usb_cam-test.launch
然後：rosrun rosserial_python serial_node.py /dev/ttyACM0
最後：rosrun cv_bridge_tutorial_pkg sample_cv_bridge_node
應該放一個視訊的，但貌似不支援，拍了一個小視訊放到了抖音上！最後截一張圖。
人臉跟隨，人體識別，身份證，車牌號識別等技術都歸納在NDK體系。這裡整理的了技能圖，以及視訊資料；
NDK開發
加群免費領取安卓進階視訊教程，原始碼，面試資料，群內有大牛一起交流討論技術；964557053。點選連結加入群聊
（包括自定義控制元件、NDK、架構設計、混合式開發工程師(React native，Weex)、效能優化、完整商業專案開發等）
騰訊T3 Android高階技術視訊教程.png