Robomaster-CV教程（1）-官方视觉代码的分析

前言

在robomaster比赛中，自瞄占据了视觉组工作的关键部分，而在自瞄中又包含了装甲板的识别以及追踪等等方面，笔者作为视觉组的一位萌新，将记录下来自己的学习历程以及心得体会，希望可以不断进步

官方例程的分析

#include "stdafx.h"
#include "cv.h"
#include "highgui.h"
#include "cxcore.h"
#include "omp.h"

using namespace cv;
using namespace std;

#define T_ANGLE_THRE 10
#define T_SIZE_THRE 5


void BrightAdjust(IplImage* src, IplImage* dst,
	double dContrast, double dBright)
{
	int nVal;

	unsigned char* SrcData = (unsigned char*)src->imageData; //存储原始图片数据像素信息的地址
	unsigned char* DstData = (unsigned char*)dst->imageData; //存储目标图片地址信息的地址
	int step = src->widthStep / sizeof(unsigned char) / 3;//每一行有多少个像素 作为步长
	//widthstep 每一行存储像素的总共字节数 对于RGB图像来说的话 因为channel = 3 所以像素的数量 = 字节数量/3
	omp_set_num_threads(8);
#pragma omp parallel for

	for (int nI = 0; nI<src->height; nI++)
	{
		for (int nJ = 0; nJ <src->width; nJ++)
		{
			for (int nK = 0; nK < 3; nK++)
			{
				//nVal = (int)(dContrast * SrcData[(nI*step + nJ) * 3 + nK] + dBright);
				nVal = (int)(dContrast * SrcData[(nI*step + nJ) * 3 + nK] + dBright);
				if (nVal < 0)
					nVal = 0;
				if (nVal > 255)
					nVal = 255;
				DstData[(nI*step + nJ) * 3 + nK] = nVal;
			}
		}
	}
}
void GetDiffImage(IplImage* src1, IplImage* src2, IplImage* dst, int nThre)
{	//如果R通道-G通道的像素值大于25，那么目标像素值就是255，否则就是0
	unsigned char* SrcData1 = (unsigned char*)src1->imageData;
	unsigned char* SrcData2 = (unsigned char*)src2->imageData;
	unsigned char* DstData = (unsigned char*)dst->imageData;
	int step = src1->widthStep / sizeof(unsigned char);//计算出每一行像素的步长

	omp_set_num_threads(8);
#pragma omp parallel for

	for (int nI = 0; nI<src1->height; nI++)
	{
		for (int nJ = 0; nJ <src1->width; nJ++)
		{
			if (SrcData1[nI*step + nJ] - SrcData2[nI*step + nJ]> nThre)
			{
				DstData[nI*step + nJ] = 255;
			}
			else
			{
				DstData[nI*step + nJ] = 0;
			}
		}
	}
}

vector<CvBox2D> ArmorDetect(vector<CvBox2D> vEllipse)
{
	vector<CvBox2D> vRlt;
	CvBox2D Armor;
	int nL, nW;
	double dAngle;
	vRlt.clear();
	if (vEllipse.size() < 2)
		return vRlt;
	for (unsigned int nI = 0; nI < vEllipse.size() - 1; nI++)
	{
		for (unsigned int nJ = nI + 1; nJ < vEllipse.size(); nJ++)
		{
			dAngle = abs(vEllipse[nI].angle - vEllipse[nJ].angle);
			while (dAngle > 180)
				dAngle -= 180;
			if ((dAngle < T_ANGLE_THRE || 180 - dAngle < T_ANGLE_THRE) && abs(vEllipse[nI].size.height - vEllipse[nJ].size.height) < (vEllipse[nI].size.height + vEllipse[nJ].size.height) / T_SIZE_THRE && abs(vEllipse[nI].size.width - vEllipse[nJ].size.width) < (vEllipse[nI].size.width + vEllipse[nJ].size.width) / T_SIZE_THRE)
			{//上述条件语句实现的功能是判断LED是否属于同一个装甲板的两个灯条
				Armor.center.x = (vEllipse[nI].center.x + vEllipse[nJ].center.x) / 2;
				Armor.center.y = (vEllipse[nI].center.y + vEllipse[nJ].center.y) / 2;
				Armor.angle = (vEllipse[nI].angle + vEllipse[nJ].angle) / 2; //所在矩形的旋转角
				if (180 - dAngle < T_ANGLE_THRE)
					Armor.angle += 90;
				nL = (vEllipse[nI].size.height + vEllipse[nJ].size.height) / 2;
				nW = sqrt((vEllipse[nI].center.x - vEllipse[nJ].center.x) * (vEllipse[nI].center.x - vEllipse[nJ].center.x) + (vEllipse[nI].center.y - vEllipse[nJ].center.y) * (vEllipse[nI].center.y - vEllipse[nJ].center.y));
				//装甲板的宽度等于匹配的两个旋转矩形的中心坐标的距离
				if (nL < nW)
				{
					Armor.size.height = nL;
					Armor.size.width = nW;
				}
				else
				{
					Armor.size.height = nW;
					Armor.size.width = nL;
				}
				vRlt.push_back(Armor);//将找出的装甲板的旋转矩形保存下来
			}
		}
	}
	return vRlt;
}

void DrawBox(CvBox2D box, IplImage* img)
{
	CvPoint2D32f point[4];
	int i;
	for (i = 0; i<4; i++)
	{
		point[i].x = 0;
		point[i].y = 0;
	}
	cvBoxPoints(box, point); //计算二维盒子顶点 
	CvPoint pt[4];
	for (i = 0; i<4; i++)
	{
		pt[i].x = (int)point[i].x;
		pt[i].y = (int)point[i].y;
	}
	cvLine(img, pt[0], pt[1], CV_RGB(0, 0, 255), 2, 8, 0); //这四条语句的作用是将四个顶点连接起来
	cvLine(img, pt[1], pt[2], CV_RGB(0, 0, 255), 2, 8, 0);
	cvLine(img, pt[2], pt[3], CV_RGB(0, 0, 255), 2, 8, 0);
	cvLine(img, pt[3], pt[0], CV_RGB(0, 0, 255), 2, 8, 0);
}

int main()
{
	CvCapture* pCapture0 = cvCreateFileCapture("RawImage\\RedCar.avi"); //对视频进行逐帧提取
	//CvCapture* pCapture0 = cvCreateCameraCapture(0);
	IplImage* pFrame0 = NULL;

	CvSize pImgSize;
	CvBox2D s;
	vector<CvBox2D> vEllipse;
	vector<CvBox2D> vRlt;
	vector<CvBox2D> vArmor;
	CvScalar sl;
	bool bFlag = false;
	CvSeq *pContour = NULL;
	CvMemStorage *pStorage = cvCreateMemStorage(0);


	pFrame0 = cvQueryFrame(pCapture0);//从视频中捕获并解压一帧后返回 用户不要释放或者重定义该返回帧

	pImgSize = cvGetSize(pFrame0);

	IplImage *pRawImg = cvCreateImage(pImgSize, IPL_DEPTH_8U, 3);

	IplImage* pGrayImage = cvCreateImage(pImgSize, IPL_DEPTH_8U, 1);//创建灰度图像
	IplImage* pRImage = cvCreateImage(pImgSize, IPL_DEPTH_8U, 1);
	IplImage* pGImage = cvCreateImage(pImgSize, IPL_DEPTH_8U, 1);
	IplImage *pBImage = cvCreateImage(pImgSize, IPL_DEPTH_8U, 1);
	IplImage *pBinary = cvCreateImage(pImgSize, IPL_DEPTH_8U, 1);
	IplImage *pRlt = cvCreateImage(pImgSize, IPL_DEPTH_8U, 1);

	CvSeq* lines = NULL;
	CvMemStorage* storage = cvCreateMemStorage(0);
	while (1)
	{
		if (pFrame0) //当读取到视频中的每一帧时
		{
			BrightAdjust(pFrame0, pRawImg, 1, -120);
			cvSplit(pRawImg, pBImage, pGImage, pRImage, 0);//将三个通道的像素值分离

			GetDiffImage(pRImage, pGImage, pBinary, 25);
			cvDilate(pBinary, pGrayImage, NULL, 3); //对图像进行膨胀处理 细小物质被填充
			cvErode(pGrayImage, pRlt, NULL, 1);//对图像进行腐蚀处理 消除了噪点和干扰点
			cvFindContours(pRlt, pStorage, &pContour, sizeof(CvContour), CV_RETR_CCOMP, CV_CHAIN_APPROX_SIMPLE);
			for (; pContour != NULL; pContour = pContour->h_next)
			{
				if (pContour->total > 10)
				{
					bFlag = true;
					s = cvFitEllipse2(pContour);//获取轮廓的椭圆边界框
					for (int nI = 0; nI < 5; nI++) //遍历以旋转矩形中心点为中心的5*5区域的像素块
					{
						for (int nJ = 0; nJ < 5; nJ++)
						{
							if (s.center.y - 2 + nJ > 0 && s.center.y - 2 + nJ < 480 && s.center.x - 2 + nI > 0 && s.center.x - 2 + nI <  640)
							{//判断该中心点及其附近的区域是否在图像边框的范围内
								sl = cvGet2D(pFrame0, (int)(s.center.y - 2 + nJ), (int)(s.center.x - 2 + nI));//获取每一个点的RGB像素值
								if (sl.val[0] < 200 || sl.val[1] < 200 || sl.val[2] < 200)
									bFlag = false;
							}
						}
					}
					if (bFlag)
					{
						vEllipse.push_back(s); //将图片中的椭圆边界框填充到vEllipse当中
						//cvEllipseBox(pFrame0, s, CV_RGB(255, 0, 0), 2, 8, 0);
					}
				}
				
			}
			vRlt = ArmorDetect(vEllipse);

			for (unsigned int nI = 0; nI < vRlt.size(); nI++)
				DrawBox(vRlt[nI], pFrame0);
			//将每一个装甲板对应的顶点都画在相应的图像上面

			cvShowImage("Raw", pFrame0);
			cvWaitKey(0);
			vEllipse.clear();
			vRlt.clear();
			vArmor.clear();
		}
		pFrame0 = cvQueryFrame(pCapture0); //重新对视频进行逐帧抓取
	}
	cvReleaseCapture(&pCapture0);//释放videocapture结构体指针
	return 0;
}