Data Persistence with cv::FileStorage(数据存储)

主要函数

构造一个文件对象

FileStorage::FileStorage();
 
FileStorage::FileStorage( string fileName, int flag );

打开文件对象

FileStorage::open( string fileName, int flag );

读入文件数据

cv::Mat anArray;
myFileStorage["calibrationMatrix"] >> anArray;
 
int aNumber;
myFileStorage["someInteger"] >> aNumber;
 
int aNumber;
aNumber = (int)myFileStorage["someInteger"]; //mapping类型数据，KEY=someInteger
 
//或通过iterator
cout << (int)lbpVal[i] //sequence类型数据的的i个元素

关闭文件

cv::FileStorage::release() 

写出

将数据保存在“*.xml” , “*.yml”里

• 构建文件对象cv::FileStorage
• 打开文件cv::FileStorage::open FileStorage::WRITE
• 写数据
  • mapping 键值对 标志:
  • sequence 一系列未命名条目 [ ]
• 释放文件对象cv::FileStorage::release

读入

• 创建对象并打开 FileStorage::READ

• data can be read with
  |-([]操作符重载) cv::FileStorage::operator
  | |-类型        | []中的值
  | |-mapping | string KEY
  | |-sequence | integer
  | |
  | |-返回一个cv::FileNode类型的对象
  |-(文件节点(file node)的迭代器)cv::FileNode::Iterator

• 对于cv::FileNode类型的对象，如果为基本数据类型,可通过强转(或隐式)直接获取

• 通过 = 或 >> 保存在变量中

• release

source code

//vs2019,OpenCV4.5.1
#include <iostream>
using namespace std;
#include <opencv2/core.hpp>
using namespace cv;
#include <ctime>
 
int main() {
	
	/************* 写出 *************/
	FileStorage fs;
	fs.open("test.yml", FileStorage::WRITE);
	//FileStorage fs("test.yml", FileStorage::WRITE); // 构建+打开
 
	fs << "frameCount" << 5;
 
	time_t rowTime;
	time(&rowTime);
	fs << "calibrationDate" << asctime(localtime(&rowTime));
 
	Mat caneraMatrix = (Mat_<double>(3,3)<<1000,0,320,0,1000,240,0,0,1);
	Mat distCoffs = (Mat_<double>(5, 1) << 0.1, 0.01, -0.001, 0, 0);
 
	fs << "caneraMatrix" << caneraMatrix << "distCoffs" << distCoffs;
 
	fs << "features" << "["; 
	for (int i = 0; i < 3; i++) {
		int x = rand() % 640;
		int y = rand() % 480;
		uchar lbp = rand() % 256;
 
		fs << "{:" << "x" << x << "y" << y << "lbp" << "[:";
		for (int j = 0; j < 8; j++)
			fs << ((lbp >> j) & 1); // 疑问
		fs << "]" << "}";
	}
	fs << "]";
 
	fs.release();
 
	cout << "test.yml Done!" << endl;
	system("pause");
 
	/************* 读入 *************/
	fs.open("test.yml", FileStorage::READ);
 
	//读入法1, 等号
	int frameCount = (int)fs["frameCount"];
 
	//读入法2, >>
	String calibrationDate;
	fs["calibrationDate"] >> calibrationDate;
	fs["caneraMatrix"] >> caneraMatrix;
	fs["distCoffs"] >> distCoffs;
 
	cout << "frameCount = " << frameCount
		<< "\ncalibrationDate = " << calibrationDate
		<< "\ncaneraMatrix = \n" << caneraMatrix
		<< "\n\ndistCoffs = \n" << distCoffs << endl;
 
	//通过迭代器访问元素
	FileNode features = fs["features"];
	FileNodeIterator iterator = features.begin(), iteratorEnd = features.end();
	vector<unsigned char> lbpVal;
 
	for (; iterator != iteratorEnd; iterator++) {
		cout << "\nx = " << (int)(*iterator)["x"] << ", " << "y = " << (int)(*iterator)["y"]<<", "<<"lbp = [ ";
		(*iterator)["lbp"] >> lbpVal;
		for (int i{ 0 }; i < lbpVal.size(); i++) {
			cout << (int)lbpVal[i]<<" " ;
		}
		cout << " ]" << endl;
	}
	fs.release();
	return 0;
}

test.yml

%YAML:1.0
---
frameCount: 5
calibrationDate: "Mon Apr  4 16:11:42 2022\n"
caneraMatrix: !!opencv-matrix
   rows: 3
   cols: 3
   dt: d
   data: [ 1000., 0., 320., 0., 1000., 240., 0., 0., 1. ]
distCoffs: !!opencv-matrix
   rows: 5
   cols: 1
   dt: d
   data: [ 1.0000000000000001e-01, 1.0000000000000000e-02,
       -1.0000000000000000e-03, 0., 0. ]
features:
   - { x:41, y:227, lbp:[ 0, 1, 1, 1, 1, 1, 0, 1 ] }
   - { x:260, y:449, lbp:[ 0, 0, 1, 1, 0, 1, 1, 0 ] }
   - { x:598, y:78, lbp:[ 0, 1, 0, 0, 1, 0, 1, 0 ] }

result