转载请注明:http://www.cppblog.com/polly-yang/
一.房屋检测小结
一开始,直接用LSD(Line Segment Detector)检测VHR(Very High Resolution)遥感卫星图像中的房屋,效果很屎。效果很屎的主要原因是因为存在各种干扰,概括下来,主要有:
1. 道路。道路干扰性强主要是因为道路呈现各种形态,弯曲,笔直,宽度不一。同时还有桥梁也影响检测率,桥梁附近呈现较好的阴影效果和Line Segment,会干扰几何检测方法。
2.森林或农田。利用LSD检测Line Segment的时候,由于LSD原理限制,在森林或农田区域,会形成较多的Line Segment,影响几何检测方法。
3.房屋本身。实际VHR遥感图像中,房屋形态较多,大小差异较大,且有的呈现不规则的形状,有的成像后,边缘模糊,不利于产生较好的LSD结果。
4.遮挡。主要是被森林遮挡,屋顶形状不完整,不利于设计算法判别。
6.光照不均,屋顶自身纹理。光照不均和屋顶自己纹理不均匀,都会加大VHR中物体的检测难度。
二.Gabor 滤波器简介(部分资料来自维基百科)
在图像处理领域,Gabor滤波器是一个用于边缘检测的线性滤波器。Gabor滤波器的频率和方向表示接近人类视觉系统对于频率和方向的表示,并且它们常备用于纹理表示和描述。在空域,一个2维的Gabor滤波器是一个正弦平面波和高斯核函数的乘积。Gabor滤波器是自相似的,也就是说,所有Gabor滤波器都可以从一个母小波经过膨胀和旋转产生。实际应用中,Gabor滤波器可以在频域的不同尺度,不同方向上提取相关特征。
三.Gabor滤波器公式化定义
公式中:
λ:正弦函数波长;
θ:Gabor核函数的方向
ψ:相位偏移
σ:高斯函数的标准差
γ: 空间的宽高比(这个没太理解)
四.Gabor 滤波器opencv实现代码
CGaborFilter::CGaborFilter(float dLambda, float dTheta, float dRatio_S2L, float dGamma, float dPhi)
{
Lambda = dLambda;
Theta = dTheta;
sigma = dLambda*dRatio_S2L;
Gamma = dGamma;
Phi = dPhi;
m_pGaborFilter = NULL;
bParam = 1;
}
CGaborFilter::~CGaborFilter(void)
{
cvReleaseMat(&m_pGaborFilter);
}
void CGaborFilter::Init()
{
float dtmp;
int itmp;
if(is_param() == 0)
{
printf("The parameters are not enough!");
}
else
{
dtmp = sqrt(48*pow(sigma,2)+1);
itmp = cvRound(dtmp);
if(itmp%2 == 0)
itmp ++;
GaborWindow.height = GaborWindow.width = 16;
bInit = 1;
create_kernel();
}
}
void CGaborFilter::Init(float dSigma,float dTheta,float dPhi)
{
float dtmp;
int itmp;
sigma = dSigma;
Theta = dTheta;
Phi = dPhi;
Gamma = GAMMA;
Lambda = sigma/RATIO_S2L;
bParam = 1;
dtmp = sqrt(24*pow(sigma,2));
itmp = cvRound(dtmp);
if(itmp%2 == 0)
itmp ++;
GaborWindow.height = GaborWindow.width = itmp;
bInit = 1;
create_kernel();
}
void CGaborFilter::Init(float dLambda,float dTheta, float dPhi,float dGamma)
{
float dtmp;
int itmp;
Lambda = dLambda;
Theta = dTheta;
Phi = dPhi;
Gamma = dGamma;
sigma = Lambda * RATIO_S2L;
bParam = 1;
dtmp = sqrt(24*pow(sigma,2));
itmp = cvRound(dtmp);
if(itmp%2 == 0)
itmp ++;
GaborWindow.height = GaborWindow.width = itmp;
bInit = 1;
create_kernel();
}
void CGaborFilter::create_kernel()
{
float tmp1,tmp2,xtmp,ytmp,re;
int i,j,x,y;
if(is_init() == 0)
printf("The parameters haven't been initialed!");
else{
m_pGaborFilter = cvCreateMat(GaborWindow.height,GaborWindow.width,CV_32FC1);
for(i= 0; i< GaborWindow.height; i++)
for(j = 0; j< GaborWindow.width; j++)
{
x = j - GaborWindow.width/2;
y = i - GaborWindow.height/2;
xtmp = (float)x*cos(Theta) - (float)y*sin(Theta);
ytmp = (float)x*sin(Theta) + (float)y*cos(Theta);
tmp1 = exp(-(pow(xtmp,2)+pow(ytmp*Gamma,2))/(2*pow(sigma,2)));
tmp2 = cos(2*PI*xtmp/Lambda + Phi);
// int p=sizeof(float);
re = tmp1*tmp2;
cvSetReal2D((CvMat*)m_pGaborFilter,i,j,re);
}
bKernel = 1;
}
}
IplImage * CGaborFilter::get_Image()
{
if(is_kernel() == 0)
{
printf("The filter hasn't bee created!");
}
else
{
IplImage *pImg = cvCreateImage(GaborWindow,IPL_DEPTH_32F,1);
IplImage *pImgU8 = cvCreateImage(GaborWindow,IPL_DEPTH_8U,1);
CvMat * pMat = cvCreateMat(GaborWindow.height,GaborWindow.width,CV_32FC1);
cvCopy(m_pGaborFilter,pImg);
//pImg->imageData = (char *)pMat->data;
cvNormalize((IplImage*)pImg, (IplImage*)pImg,0,255,CV_MINMAX,NULL);
cvConvertScaleAbs(pImg,pImgU8,1,0);
return pImgU8;
}
}
IplImage * CGaborFilter::do_filter(const IplImage * src)
{
if(is_kernel()==false)
{
printf("The Gabor Kernel has not been created!");
}
else{
IplImage *pDestImage = cvCreateImage(cvSize(src->width,src->height),IPL_DEPTH_8U,1);
// IplImage * pGaborImage = get_Imge();
// CvMat GaborKernel = cvMat(pGaborImage->height,pGaborImage->width,CV_8U,pGaborImage->imageData);
IplImage *tmpImg = cvCloneImage(src);
IplImage *tmpGrayImg = cvCreateImage(cvSize(src->width,src->height),IPL_DEPTH_8U,1);
if(tmpImg->nChannels != 1)
{
cvCvtColor(tmpImg,tmpGrayImg,CV_BGR2GRAY);
}
else
{
cvReleaseImage(&tmpGrayImg);
tmpGrayImg = tmpImg;
}
CvMat * pGaborKernel = get_Mat();
cvFilter2D(tmpGrayImg,pDestImage,pGaborKernel,cvPoint((GaborWindow.width-1)/2,(GaborWindow.height-1)/2));
cvReleaseImage(&tmpImg);
return pDestImage;
}
}