Limits...
Building keypoint mappings on multispectral images by a cascade of classifiers with a resurrection mechanism.

Li Y, Jing J, Jin H, Qiao W - Sensors (Basel) (2015)

Bottom Line: Initial keypoint mappings are built with the descriptors associated with keypoints; then, at each step, only a small number of keypoint mappings of a high confidence are classified to be incorrect.Due to the drawback of a classification rule, some correct keypoint mappings may be misclassified as incorrect at a step.Experimental results show that the proposed cascade structure combined with the resurrection mechanism can effectively build more reliable keypoint mappings on multispectral images than existing methods.

View Article: PubMed Central - PubMed

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

ABSTRACT
Inspired by the boosting technique for detecting objects, this paper proposes a cascade structure with a resurrection mechanism to establish keypoint mappings on multispectral images. The cascade structure is composed of four steps by utilizing best bin first (BBF), color and intensity distribution of segment (CIDS), global information and the RANSAC process to remove outlier keypoint matchings. Initial keypoint mappings are built with the descriptors associated with keypoints; then, at each step, only a small number of keypoint mappings of a high confidence are classified to be incorrect. The unclassified keypoint mappings will be passed on to subsequent steps for determining whether they are correct. Due to the drawback of a classification rule, some correct keypoint mappings may be misclassified as incorrect at a step. Observing this, we design a resurrection mechanism, so that they will be reconsidered and evaluated by the rules utilized in subsequent steps. Experimental results show that the proposed cascade structure combined with the resurrection mechanism can effectively build more reliable keypoint mappings on multispectral images than existing methods.

No MeSH data available.


On each dataset, from left to right: the number of pending mappings, the number of resurrected mappings and the number of resurrected correct mappings.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4481929&req=5

f5-sensors-15-11769: On each dataset, from left to right: the number of pending mappings, the number of resurrected mappings and the number of resurrected correct mappings.

Mentions: Figure 5 shows the performance of the resurrection mechanism. From left to right are the total number of pending mappings in Steps 1 and 2 (blue bar), the total number of resurrected mappings in Steps 2 and 3 (red bar) and the number of resurrected correct mappings (green bar). From Figure 5, it can be seen that the resurrection mechanism successfully recovers some wrongly-discarded correct mappings (green bar) on all datasets. In particular, the dataset ‘EOIR’ is more challenging than other datasets, as the image pairs contain much fewer correct mappings, and hence, the number of recovered correct ones by the resurrection mechanism is much smaller, as well.


Building keypoint mappings on multispectral images by a cascade of classifiers with a resurrection mechanism.

Li Y, Jing J, Jin H, Qiao W - Sensors (Basel) (2015)

On each dataset, from left to right: the number of pending mappings, the number of resurrected mappings and the number of resurrected correct mappings.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4481929&req=5

f5-sensors-15-11769: On each dataset, from left to right: the number of pending mappings, the number of resurrected mappings and the number of resurrected correct mappings.
Mentions: Figure 5 shows the performance of the resurrection mechanism. From left to right are the total number of pending mappings in Steps 1 and 2 (blue bar), the total number of resurrected mappings in Steps 2 and 3 (red bar) and the number of resurrected correct mappings (green bar). From Figure 5, it can be seen that the resurrection mechanism successfully recovers some wrongly-discarded correct mappings (green bar) on all datasets. In particular, the dataset ‘EOIR’ is more challenging than other datasets, as the image pairs contain much fewer correct mappings, and hence, the number of recovered correct ones by the resurrection mechanism is much smaller, as well.

Bottom Line: Initial keypoint mappings are built with the descriptors associated with keypoints; then, at each step, only a small number of keypoint mappings of a high confidence are classified to be incorrect.Due to the drawback of a classification rule, some correct keypoint mappings may be misclassified as incorrect at a step.Experimental results show that the proposed cascade structure combined with the resurrection mechanism can effectively build more reliable keypoint mappings on multispectral images than existing methods.

View Article: PubMed Central - PubMed

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

ABSTRACT
Inspired by the boosting technique for detecting objects, this paper proposes a cascade structure with a resurrection mechanism to establish keypoint mappings on multispectral images. The cascade structure is composed of four steps by utilizing best bin first (BBF), color and intensity distribution of segment (CIDS), global information and the RANSAC process to remove outlier keypoint matchings. Initial keypoint mappings are built with the descriptors associated with keypoints; then, at each step, only a small number of keypoint mappings of a high confidence are classified to be incorrect. The unclassified keypoint mappings will be passed on to subsequent steps for determining whether they are correct. Due to the drawback of a classification rule, some correct keypoint mappings may be misclassified as incorrect at a step. Observing this, we design a resurrection mechanism, so that they will be reconsidered and evaluated by the rules utilized in subsequent steps. Experimental results show that the proposed cascade structure combined with the resurrection mechanism can effectively build more reliable keypoint mappings on multispectral images than existing methods.

No MeSH data available.