A Review of Keypoints’ Detection and Feature Description in Image Registration

Liu, Cuiyin; Xu, Jishang; Wang, Feng

doi:10.1155/2021/8509164

Cited by 22 publications

(22 citation statements)

References 90 publications

(222 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although image alignment can be done using feature detectors-such as MSER, FAST, SIFT, and others (Işık, 2015;Liu et al, 2021)-spinal cord imaging movies contain a variety of features across multiple layers that cause these methods to identify non-relevant features. Here, we tracked consistent vasculature, as a proxy for spinal cord motion, which improved registration (Extended Data Fig.…”

Section: Deep Learning Feature Identification and Control Point Regis...mentioning

confidence: 99%

Long-term optical imaging of the spinal cord in awake, behaving animals

Ahanonu

Crowther

Kania

et al. 2023

Preprint

View full text Add to dashboard Cite

Advances in optical imaging approaches and fluorescent biosensors have enabled an understanding of the spatiotemporal and long-term neural dynamics in the brain of awake animals. However, methodological difficulties and the persistence of post-laminectomy fibrosis have greatly limited similar advances in the spinal cord. To overcome these technical obstacles, we combinedin vivoapplication of fluoropolymer membranes that inhibit fibrosis; a redesigned, cost-effective implantable spinal imaging chamber; and improved motion correction methods that together permit imaging of the spinal cord in awake, behaving mice, for months to over a year. We also demonstrate a robust ability to monitor axons, identify a spinal cord somatotopic map, conduct Ca2+imaging of neural dynamics in behaving animals responding to pain-provoking stimuli, and observe persistent microglial changes after nerve injury. The ability to couple neural activity and behavior at the spinal cord level will drive insights not previously possible at a key location for somatosensory transmission to the brain.

show abstract

Section: Deep Learning Feature Identification and Control Point Regis...mentioning

confidence: 99%

Long-term optical imaging of the spinal cord in awake, behaving animals

Ahanonu

Crowther

Kania

et al. 2023

Preprint

View full text Add to dashboard Cite

show abstract

“…The major characteristics of these models are summarized in Ajayi (2019) and Ajayi (2020). Also, Cuiyin et al (2021) gave a systematic review of frequently used keypoints' detector and feature descriptors in image registration, including their characteristics, corresponding principles and analysis. On the other hand, Guorong and Shuangming (2020) proposed a new feature descriptor for multimodal image registration by taking advantage of the illumination and contrast invariant properties of phase congruency to provide solution to the problem of nonlinear changes between image pairs.…”

Section: Introductionmentioning

confidence: 99%

Accuracy Assessment of the Effect of Different Feature Descriptors on the Automatic Co-Registration of Overlapping Images

Ajayi,

Nwadialor

2024

Geodesy and Cartography

View full text Add to dashboard Cite

This research seeks to assess the effect of different selected feature descriptors on the accuracy of an automatic image registration scheme. Three different feature descriptors were selected based on their peculiar characteristics, and implemented in the process of developing the image registration scheme. These feature descriptors (Modified Harris and Stephens corner detector (MHCD), the Scale Invariant Feature Transform (SIFT) and the Speeded Up Robust Feature (SURF)) were used to automatically extract the conjugate points common to the overlapping image pairs used for the registration. Random Sampling Consensus (RANSAC) algorithm was used to exclude outliers and to fit the matched correspondences, Sum of Absolute Differences (SAD) which is a correlation-based feature matching metric was used for the feature match, while projective transformation function was used for the computation of the transformation matrix (T). The obtained overall result proved that the SURF algorithm outperforms the other two feature descriptors with an accuracy measure of -0.0009 pixels, while SIFT with a cumulative signed distance of 0.0328 pixels also proved to be more accurate than MHCD with a cumulative signed distance of 0.0457 pixels. The findings affirmed the importance of choosing the right feature descriptor in the overall accuracy of an automatic image registration scheme.

show abstract

“…David Lowe devised the scale-invariant feature transform (SIFT) algorithm in 1999, which was employed in computer vision to detect, describe, and match features in images [ 15 , 16 ]. One of its applications includes aligning two or more images through the identification and matching of corresponding features or keypoints in the images [ 17 , 18 ]. The accuracy of SIFT has been established to be one of the highest among feature detector descriptor algorithms for scale and rotation variations.…”

Section: Introductionmentioning

confidence: 99%

Comprehensive Evaluation of Multispectral Image Registration Strategies in Heterogenous Agriculture Environment

Rana,

Gerbino,

Crimaldi

et al. 2024

J. Imaging

View full text Add to dashboard Cite

This article is focused on the comprehensive evaluation of alleyways to scale-invariant feature transform (SIFT) and random sample consensus (RANSAC) based multispectral (MS) image registration. In this paper, the idea is to extensively evaluate three such SIFT- and RANSAC-based registration approaches over a heterogenous mix containing Triticum aestivum crop and Raphanus raphanistrum weed. The first method is based on the application of a homography matrix, derived during the registration of MS images on spatial coordinates of individual annotations to achieve spatial realignment. The second method is based on the registration of binary masks derived from the ground truth of individual spectral channels. The third method is based on the registration of only the masked pixels of interest across the respective spectral channels. It was found that the MS image registration technique based on the registration of binary masks derived from the manually segmented images exhibited the highest accuracy, followed by the technique involving registration of masked pixels, and lastly, registration based on the spatial realignment of annotations. Among automatically segmented images, the technique based on the registration of automatically predicted mask instances exhibited higher accuracy than the technique based on the registration of masked pixels. In the ground truth images, the annotations performed through the near-infrared channel were found to have a higher accuracy, followed by green, blue, and red spectral channels. Among the automatically segmented images, the accuracy of the blue channel was observed to exhibit a higher accuracy, followed by the green, near-infrared, and red channels. At the individual instance level, the registration based on binary masks depicted the highest accuracy in the green channel, followed by the method based on the registration of masked pixels in the red channel, and lastly, the method based on the spatial realignment of annotations in the green channel. The instance detection of wild radish with YOLOv8l-seg was observed at a mAP@0.5 of 92.11% and a segmentation accuracy of 98% towards segmenting its binary mask instances.

show abstract

A Review of Keypoints’ Detection and Feature Description in Image Registration

Cited by 22 publications

References 90 publications

Long-term optical imaging of the spinal cord in awake, behaving animals

Long-term optical imaging of the spinal cord in awake, behaving animals

Accuracy Assessment of the Effect of Different Feature Descriptors on the Automatic Co-Registration of Overlapping Images

Comprehensive Evaluation of Multispectral Image Registration Strategies in Heterogenous Agriculture Environment

Contact Info

Product

Resources

About