Image Feature Information Extraction for Interest Point Detection: A Comprehensive Review

Ji, Jing; Gao, Tengfei; Zhang, Weichuan; Gao, Yongsheng; Sun, Changming

doi:10.1109/tpami.2022.3201185

Cited by 33 publications

(8 citation statements)

References 198 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The core of the Harris corner point detection algorithm is the autocorrelation of the image gray level in the window. The Harris operator realizes corner detection by a small movement of the window [37, 38]. Taking the pixel as the center and shifting to the x and y directions.…”

Section: Methodsmentioning

confidence: 99%

Fast autofocusing of recorded planes by salient feature region for coherent diffraction imaging

Zhou,

Li,

Qiao

et al. 2023

Journal of Biophotonics

View full text Add to dashboard Cite

In multi‐distance coherent diffraction imaging (MDCDI), the task of distance calculation for multi‐diffraction images is cumbersome. The information features are hard‐to‐extract and the region of interest extraction algorithms are difficult to be adopted. A universal salient feature region selection algorithm by using the area with the highest density of corners is proposed to extract the most representative feature region. In addition, equally spaced recording modes and mismatched diffraction distances will result in system noise and destroy image quality. The polydirectional maximum gradient is offered as a sharpness criterion to weigh a quantitative feature for the final pattern. A fast, sensitive, and high‐accuracy autofocusing and sample reconstruction can be achieved using only a small number of images while ensuring that morphological properties and quantification of the reconstructions are not compromised. The proposed method is promising for biological and medical dynamic observations for computational imaging systems.This article is protected by copyright. All rights reserved.

show abstract

Section: Methodsmentioning

confidence: 99%

Fast autofocusing of recorded planes by salient feature region for coherent diffraction imaging

Zhou,

Li,

Qiao

et al. 2023

Journal of Biophotonics

View full text Add to dashboard Cite

show abstract

“…In this paper, the first-and second-order directional derivative [25][26][27][28][29][30][31][32][33][34][35] of image local structural features are utilized to investigate the properties of the features which also enable us to study the existing LSI extraction, image data augmentation, and description of local structure feature techniques. Our research indicates that the existing image data augmentation techniques (e.g., lighting changes 36 , colorizing image 20 , and image affine transformations 23 ) have a great impact on the performance of FGVC.…”

Section: Image Local Structure Information Learning For Fine-grained ...mentioning

confidence: 99%

Image local structure information learning for fine-grained visual classification

Lü

Zhang

Zhao

et al. 2022

Sci Rep

Self Cite

View full text Add to dashboard Cite

Learning discriminative visual patterns from image local salient regions is widely used for fine-grained visual classification (FGVC) tasks such as plant or animal species classification. A large number of complex networks have been designed for learning discriminative feature representations. In this paper, we propose a novel local structure information (LSI) learning method for FGVC. Firstly, we indicate that the existing FGVC methods have not properly considered how to extract LSI from an input image for FGVC. Then an LSI extraction technique is introduced which has the ability to properly depict the properties of different local structure features in images. Secondly, a novel LSI learning module is proposed to be added into a given backbone network for enhancing the ability of the network to find salient regions. Thirdly, extensive experiments show that our proposed method achieves better performance on six image datasets. Particularly, the proposed method performs far better on datasets with a limited number of images.

show abstract

“…Within the spherical intensity images, the keypoints are automatically detected using the Speeded Up Robust Features (SURF) algorithm (Bay et al, 2008) and described using the Binary Robust Invariant Scalable Keypoints (BRISK) algorithm (Leutenegger et al, 2011). The choice is made based on our simplified analysis of different feature detectors and descriptors (Jing et al, 2021), where the selected algorithms outperformed the rest (the differences were in some cases marginal).…”

Section: Implemented Algorithmmentioning

confidence: 99%

“…points that can be well located in the scan and whose neighborhood can be unambiguously described. The corresponding points are found automatically in the scene using the existing feature description and matching algorithms for 2D local image keypoints (Jing et al, 2021). To achieve this, we exploit the regular and nearly continuous TLS scanning pattern to map the 3D TLS measurements onto 2D intensity images.…”

Section: Introductionmentioning

confidence: 99%

Keypoint-based deformation monitoring using a terrestrial laser scanner from a single station: Case study of a bridge pier

Medić

Ruttner

Holst

2022

Proceedings of the 5th Joint International Symposium on Deformation Monitoring - JISDM 2022

View full text Add to dashboard Cite

Terrestrial laser scanners (TLSs) offer a possibility for more automated and efficient deformation monitoring of civil engineering structures with higher spatial resolution than standard methods, as well as without the necessity of permanently installing the monitoring equipment. In such applications, scanners are usually placed so that the lines of sight are roughly aligned with the main directions of the expected deformations, and the deformations are estimated from point cloud differences between multiple epochs. This allows high sensitivity in the direction of the surface normal, but deformations along the surface are often undetected or hard to precisely quantify. In this work, we propose an algorithm based on the detection and matching of keypoints identified within TLS intensity images. This enables precise quantification of deformations along the scanned surfaces. We also present the application of the algorithm for monitoring a bridge pier of the Hochmoselbrücke in Germany, as a case study. Deformations up to about 4 cm due to thermal expansion and bending of the pier were successfully detected from scans taken throughout the day from a single location, up to 180 m from the monitored surfaces. The results agreed within a few millimeters to independent monitoring using state-of-the-art processing of TLS point clouds obtained from a different location and using a different type/brand of instrument. The newly proposed algorithm can either be used to complement existing TLS-based deformation analysis methods by adding sensitivity in certain directions, or it can be valuable as a standalone solution.

show abstract

Image Feature Information Extraction for Interest Point Detection: A Comprehensive Review

Cited by 33 publications

References 198 publications

Fast autofocusing of recorded planes by salient feature region for coherent diffraction imaging

Fast autofocusing of recorded planes by salient feature region for coherent diffraction imaging

Image local structure information learning for fine-grained visual classification

Keypoint-based deformation monitoring using a terrestrial laser scanner from a single station: Case study of a bridge pier

Contact Info

Product

Resources

About