Apply a specific threshold on the Moravec corner detection algorithm.
[ct] = NARVAL_IP_MoravecT(c,t)
corner matrix.
threshold.
updated corner matrix.
NARVAL_IP_MoravecT applies the specific threshold t on the corner matrix c generated by the Moravec corner detection algorithm (WIKIPEDIA). The output is stored in ct.
This is one of the earliest corner detection algorithms and defines a corner to be a point with low self-similarity. The algorithm tests each pixel in the image to see if a corner is present, by considering how similar a patch centered on the pixel is to nearby, largely overlapping patches. The similarity is measured by taking the sum of squared differences (SSD) between the two patches. A lower number indicates more similarity.
If the pixel is in a region of uniform intensity, then the nearby patches will look similar. If the pixel is on an edge, then nearby patches in a direction perpendicular to the edge will look quite different, but nearby patches in a direction parallel to the edge will result only in a small change. If the pixel is on a feature with variation in all directions, then none of the nearby patches will look similar.
The corner strength is defined as the smallest SSD between the patch and its neighbors (horizontal, vertical and on the two diagonals). If this number is locally maximal, then a feature of interest is present.
As pointed out by Moravec, one of the main problems with this operator is that it is not isotropic: if an edge is present that is not in the direction of the neighbours, then it will not be detected as an interest point.
dt=getdate(); seed=dt(10); rand('seed',seed);//initialization of the random values generator no=4;//quantity of obstacles (rectangle) L=1000;//squared area side hm=100;//minimal height hM=250;//maximal height wm=100;//minimal width wM=250;//maximal width Al=[0 %pi/2 %pi -%pi/2];//available angles for obstacles [Xs,Ys,X,Y,H,W,A]=NARVAL_IP_RectObstacles(no,L,hm,hM,wm,wM);//generation of obstacles [P]=NARVAL_IP_PotentialRot(X,Y,H,W,A,L);//generation of obstacle matrix z=10;//zoom factor o=2;// 1=mean, 2=max, 3=min, 4=median Pz=NARVAL_IP_MRA(P,z,o);//scale modification [Pzx,Pzy]=size(Pz);//image size w1=1;//window index w2=2;//window index scf(w1); clf(w1); grayplot(1:Pzx,1:Pzy,Pz);//graph visualization xset("colormap",graycolormap(128)); scf(w2); clf(w2); [PEz]=NARVAL_IP_Erosion(Pz);//contour performance Cont=Pz-PEz;//contour 1 [Contx,Conty]=size(Cont);//image size grayplot(1:Contx,1:Conty,Cont);//graph visualization xset("colormap",graycolormap(128)); w3=3;//window index scf(w3); clf(w3); wm=3 [cornerz]=NARVAL_IP_Moravec(Pz,wm);//Moravec algorithm [cornerzx,cornerzy]=size(cornerz);//image size grayplot(1:cornerzx,1:cornerzy,cornerz);//graph visualization xset("colormap",graycolormap(128)); T=min(cornerz)+(max(cornerz)-min(cornerz))*3/4;//threshold [cornerT]=NARVAL_IP_MoravecT(cornerz,T);//application of NARVAL_IP_MoravecT [cornerTx,cornerTy]=size(cornerT);//image size w4=4;//window index scf(w4); clf(w4); grayplot(1:cornerTx,1:cornerTy,cornerT);//graph visualization xset("colormap",graycolormap(128)); | ![]() | ![]() |
Dr. Foued Melakessou
Research Associate
Interdisciplinary Centre for Security, Reliability and Trust
Room F106
University of Luxembourg
6, rue Coudenhove Kalergi
L-1359 Luxembourg-Kirchberg
E-mail: foued.melakessou@uni.lu
Tel: (+352) 46 66 44 5346