Estimating parameters of interframe geometric transformation of an image sequence by the fixed point method

Krasheninnikov, V. R.; Потапов, М. А.

doi:10.1134/s1054661810030077

Cited by 2 publications

(7 citation statements)

References 1 publication

Supporting

Mentioning

Contrasting

Order By: Relevance

“…4a. Now, in (18)- (19), we will change the angle of rota tion α 1 , which leaves the other parameters constant. Then the trajectory of the corresponding FP will be elliptical ( Fig.…”

Section: Two Dimensional Images At High Rotation Angles and Scalingmentioning

confidence: 98%

“…The following table shows the estimates , , and of the unknown parameters a, b, and s, obtained by numerical solution of equations of the kind (19) for each of the eight lines. As can be seen from Table 1, the line No.…”

Section: Reduction Of Enumeration In Finding Trajectoriesmentioning

confidence: 99%

“…GT (18) is a parallel shift by the vector (a 1 , b 1 ), rotation by angle α 1 around the point (a 1 , b 1 ), and scaling by coef ficient s 1 . For different values of these parameters, there is a corresponding GT, the coordinates (u, v) of which satisfy the system of equations (19) where (20) Let the parameters (a 1 , b 1 ) in (18) be changed; then, the corresponding FP will move. If the points (a 1 , b 1 ) move in a straight line (SL) as follows: then FP will move along SL as shown below:…”

Section: Two Dimensional Images At High Rotation Angles and Scalingmentioning

confidence: 99%

“…For estimation of unknown parameters a, b, s, and α of GT (12), one must take two points on RT of FP, substitute their coordinates (u, v) in Eq. (19) and solve the resulting system of four equations for the unknowns a, b, s, and α. System (19) must be solved by numerical methods, since the analytical solution can not be found.…”

Section: Two Dimensional Images At High Rotation Angles and Scalingmentioning

confidence: 99%

“…After determining the parameters of this trajectory and solving the system of equations (19), we obtained the following estimates of the GT parameters:…”

Section: Spectrozonal Imagesmentioning

confidence: 99%

See 4 more Smart Citations

Estimation of parameters of geometric transformation of images by fixed-point method

Krasheninnikov

Потапов

2012

Pattern Recognit. Image Anal.

Self Cite

View full text Add to dashboard Cite

In solving many practical problems, the problem of combining images and the closely related problem of binding a portion of the image often arises. If the model of mutual interframe geometric transfor mation (GT) of two images is given, then the problem of combination is reduced to estimating the parameters of this model. There is a large number of works dedicated to the development of algorithms for estimating the parameters of GT; in particular, pseudo gradient algorithms have been well developed [3]. However, for the success of these (and many other) algorithms, a fairly accurate initial approximation of parameters of GT is required. Otherwise, one must take a set of initial approximations. Thus, it becomes necessary to make a rough estimate of the parameters of GT, which can be taken as an initial approximation for more accurate algorithms. In this paper, the method of rough estimation of the parameters of GT, which uses the notion of the fixed point, is proposed.

show abstract

“…4a. Now, in (18)- (19), we will change the angle of rota tion α 1 , which leaves the other parameters constant. Then the trajectory of the corresponding FP will be elliptical ( Fig.…”

Section: Two Dimensional Images At High Rotation Angles and Scalingmentioning

confidence: 98%

Section: Reduction Of Enumeration In Finding Trajectoriesmentioning

confidence: 99%

Section: Two Dimensional Images At High Rotation Angles and Scalingmentioning

confidence: 99%

Section: Two Dimensional Images At High Rotation Angles and Scalingmentioning

confidence: 99%

“…After determining the parameters of this trajectory and solving the system of equations (19), we obtained the following estimates of the GT parameters:…”

Section: Spectrozonal Imagesmentioning

confidence: 99%

See 3 more Smart Citations