Limits...
Assigning Main Orientation to an EOH Descriptor on Multispectral Images.

Li Y, Shi X, Wei L, Zou J, Chen F - Sensors (Basel) (2015)

Bottom Line: EOH has a better matching ability than SIFT (scale-invariant feature transform) on multispectral images, but does not assign a main orientation to keypoints.Then, EOH is computed for every keypoint with respect to its main orientation.In addition, an implementation variant is proposed for fast computation of the EOH descriptor.

View Article: PubMed Central - PubMed

Affiliation: Beijing University of Posts and Teles., School of Electronic Engineering, Rd. Xitucheng 10#, Beijing 100876, China. yli@bupt.edu.cn.

ABSTRACT
This paper proposes an approach to compute an EOH (edge-oriented histogram) descriptor with main orientation. EOH has a better matching ability than SIFT (scale-invariant feature transform) on multispectral images, but does not assign a main orientation to keypoints. Alternatively, it tends to assign the same main orientation to every keypoint, e.g., zero degrees. This limits EOH to matching keypoints between images of translation misalignment only. Observing this limitation, we propose assigning to keypoints the main orientation that is computed with PIIFD (partial intensity invariant feature descriptor). In the proposed method, SIFT keypoints are detected from images as the extrema of difference of Gaussians, and every keypoint is assigned to the main orientation computed with PIIFD. Then, EOH is computed for every keypoint with respect to its main orientation. In addition, an implementation variant is proposed for fast computation of the EOH descriptor. Experimental results show that the proposed approach performs more robustly than the original EOH on image pairs that have a rotation misalignment.

No MeSH data available.


Related in: MedlinePlus

The five filters used in [2,22]. The filters compute directional derivatives in 0°, 45°, 90°, 135° and the non-direction showed in (a–d) and (e) respectively. (a–d) are the direction filters and (e) is the non-direction filter.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4541846&req=5

f1-sensors-15-15595: The five filters used in [2,22]. The filters compute directional derivatives in 0°, 45°, 90°, 135° and the non-direction showed in (a–d) and (e) respectively. (a–d) are the direction filters and (e) is the non-direction filter.

Mentions: The EOH computes the gradient orientation at each edge pixel with the following five filters. These filters correspond to the 0°, 45°, 90°, 135° and non-direction, as shown in Figure 1. The filters shown in Figure 1a–d, are called direction filters, while the one shown in Figure 1e is called the non-direction filter. For a pixel, the filter giving the maximum response is defined to be the direction at the pixel. Formally, let fk(x, y),k = 0, 1, 2, 3,4, denote the mathematical representation of the five filters shown in Figure 1, then an edge pixel at (x, y) will contribute one to the bin defined by:(6)binEOH(x,y)=argmaxk/fk(x,y)•I(x,y)/where ‘•’ is the correlation between image and filter.


Assigning Main Orientation to an EOH Descriptor on Multispectral Images.

Li Y, Shi X, Wei L, Zou J, Chen F - Sensors (Basel) (2015)

The five filters used in [2,22]. The filters compute directional derivatives in 0°, 45°, 90°, 135° and the non-direction showed in (a–d) and (e) respectively. (a–d) are the direction filters and (e) is the non-direction filter.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4541846&req=5

f1-sensors-15-15595: The five filters used in [2,22]. The filters compute directional derivatives in 0°, 45°, 90°, 135° and the non-direction showed in (a–d) and (e) respectively. (a–d) are the direction filters and (e) is the non-direction filter.
Mentions: The EOH computes the gradient orientation at each edge pixel with the following five filters. These filters correspond to the 0°, 45°, 90°, 135° and non-direction, as shown in Figure 1. The filters shown in Figure 1a–d, are called direction filters, while the one shown in Figure 1e is called the non-direction filter. For a pixel, the filter giving the maximum response is defined to be the direction at the pixel. Formally, let fk(x, y),k = 0, 1, 2, 3,4, denote the mathematical representation of the five filters shown in Figure 1, then an edge pixel at (x, y) will contribute one to the bin defined by:(6)binEOH(x,y)=argmaxk/fk(x,y)•I(x,y)/where ‘•’ is the correlation between image and filter.

Bottom Line: EOH has a better matching ability than SIFT (scale-invariant feature transform) on multispectral images, but does not assign a main orientation to keypoints.Then, EOH is computed for every keypoint with respect to its main orientation.In addition, an implementation variant is proposed for fast computation of the EOH descriptor.

View Article: PubMed Central - PubMed

Affiliation: Beijing University of Posts and Teles., School of Electronic Engineering, Rd. Xitucheng 10#, Beijing 100876, China. yli@bupt.edu.cn.

ABSTRACT
This paper proposes an approach to compute an EOH (edge-oriented histogram) descriptor with main orientation. EOH has a better matching ability than SIFT (scale-invariant feature transform) on multispectral images, but does not assign a main orientation to keypoints. Alternatively, it tends to assign the same main orientation to every keypoint, e.g., zero degrees. This limits EOH to matching keypoints between images of translation misalignment only. Observing this limitation, we propose assigning to keypoints the main orientation that is computed with PIIFD (partial intensity invariant feature descriptor). In the proposed method, SIFT keypoints are detected from images as the extrema of difference of Gaussians, and every keypoint is assigned to the main orientation computed with PIIFD. Then, EOH is computed for every keypoint with respect to its main orientation. In addition, an implementation variant is proposed for fast computation of the EOH descriptor. Experimental results show that the proposed approach performs more robustly than the original EOH on image pairs that have a rotation misalignment.

No MeSH data available.


Related in: MedlinePlus