Local Feature Descriptor for Image Matching: A Survey

Leng, Chengcai; Zhang, Hai; Li, Bo; Cai, Guorong; Pei, Zhao; He, Li

doi:10.1109/access.2018.2888856

Cited by 94 publications

(51 citation statements)

References 153 publications

(151 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…O último é uma alternativa aberta suportada pela biblioteca OpenCV. Detalhes sobre esses e outros métodos podem ser encontrados no trabalho de Leng et al (2019). Neste texto, a título de ilustração, iremos abordar em um pouco mais de detalhes o método SIFT.…”

Section: Características Visuaisunclassified

Recuperação de Informação Multimídia em Big Data Utilizando OpenCV Python

Goularte¹,

Trojahn²,

Kishi³

2019

Minicursos Do XXV Simpósio Brasileiro De Sistemas Multimídia E Web

View full text Add to dashboard Cite

The popularization of systems, applications and devices to produce, view and share multimedia, saw the need to treat a large volume of data arise. In related areas (such as Multimedia Big Data, Data Science and Multimedia Information Retrieval) a key step is commonly referred as Multimedia Indexing or Multimedia Big Data Analysis, where the aim is to represent multimedia content into smaller, more manageable units, allowing the extraction of data features and information essential to the proper performance of the associated services. This mini-course discusses current tools and techniques for indexing, extracting and processing of multimodal multimedia content. The techniques are exemplified in Python OpenCV over different content (like images, audio, text and video), leading to the interest of services like Netflix, Google and YouTube on this subject, attracting the interest of researchers and developers. ResumoA popularização de sistemas, aplicativos e dispositivos para produzir, exibir e compartilhar conteúdo multimídia fez surgir a necessidade de tratar um grande volume de dados. Nas áreas relacionadas (como Multimedia Big Data, Ciência de Dados e Recuperação de Informação Multimídia) um pré-requisito chave é comumente conhecido como Indexação Multimídia (ou Análise Multimídia em Big Data), onde o objetivo é representar o conteúdo em unidades menores e mais gerenciáveis, permitindo a extração de features dos dados e informações essenciais para o bom funcionamento dos serviços associados. Este minicurso aborda ferramentas e técnicas atuais para indexação, extração e processamento de conteúdo multimídia multimodal. As técnicas XXV Simpósio Brasileiro de Sistemas Multimídia e Web: Minicursos Listagem 5.1. Exemplo de código Python utilizando OpenCV. Arquivo exemplo1.py XXV Simpósio Brasileiro de Sistemas Multimídia e Web: Minicursos

show abstract

Section: Características Visuaisunclassified

Recuperação de Informação Multimídia em Big Data Utilizando OpenCV Python

Goularte¹,

Trojahn²,

Kishi³

2019

Minicursos Do XXV Simpósio Brasileiro De Sistemas Multimídia E Web

View full text Add to dashboard Cite

show abstract

“…Traditional image registration algorithms are mainly classified into pixel-based algorithms and feature-based algorithms [5,6]. In pixel-based image registration algorithms, the original pixel values are directly used to estimate the transformation relationship between images [7,8].…”

Section: Introductionmentioning

confidence: 99%

Image Registration Algorithm Based on Convolutional Neural Network and Local Homography Transformation

Wang

Jiang

et al. 2020

Applied Sciences

View full text Add to dashboard Cite

In order to overcome the poor robustness of traditional image registration algorithms in illuminating and solving the problem of low accuracy of a learning-based image homography matrix estimation algorithm, an image registration algorithm based on convolutional neural network (CNN) and local homography transformation is proposed. Firstly, to ensure the diversity of samples, a sample and label generation method based on moving direct linear transformation (MDLT) is designed. The generated samples and labels can effectively reflect the local characteristics of images and are suitable for training the CNN model with which multiple pairs of local matching points between two images to be registered can be calculated. Then, the local homography matrices between the two images are estimated by using the MDLT and finally the image registration can be realized. The experimental results show that the proposed image registration algorithm achieves higher accuracy than other commonly used algorithms such as the SIFT, ORB, ECC, and APAP algorithms, as well as another two learning-based algorithms, and it has good robustness for different types of illumination imaging.

show abstract

“…High-quality local feature descriptors describe key points with uniqueness, repeatability, accuracy, compactness, and effective representation. These key points can keep robust and constant in terms of scaling, rotation, affine transformation, illumination, and occlusion [ 5 ]. Here we focus on the theoretical and mathematical descriptions of various local feature descriptors.…”

Section: Introductionmentioning

confidence: 99%

NCC Based Correspondence Problem for First- and Second-Order Graph Matching

Cui

Créput

2020

Sensors

View full text Add to dashboard Cite

Automatically finding correspondences between object features in images is of main interest for several applications, as object detection and tracking, identification, registration, and many derived tasks. In this paper, we address feature correspondence within the general framework of graph matching optimization and with the principal aim to contribute. We proposed two optimized algorithms: first-order and second-order for graph matching. On the one hand, a first-order normalized cross-correlation (NCC) based graph matching algorithm using entropy and response through Marr wavelets within the scale-interaction method is proposed. First, we proposed a new automatic feature detection processing by using Marr wavelets within the scale-interaction method. Second, feature extraction is executed under the mesh division strategy and entropy algorithm, accompanied by the assessment of the distribution criterion. Image matching is achieved by the nearest neighbor search with normalized cross-correlation similarity measurement to perform coarse matching on feature points set. As to the matching points filtering part, the Random Sample Consensus Algorithm (RANSAC) removes outliers correspondences. One the other hand, a second-order NCC based graph matching algorithm is presented. This algorithm is an integer quadratic programming (IQP) graph matching problem, which is implemented in Matlab. It allows developing and comparing many algorithms based on a common evaluation platform, sharing input data, and a customizable affinity matrix and matching list of candidate solution pairs as input data. Experimental results demonstrate the improvements of these algorithms concerning matching recall and accuracy compared with other algorithms.

show abstract

Local Feature Descriptor for Image Matching: A Survey

Cited by 94 publications

References 153 publications

Recuperação de Informação Multimídia em Big Data Utilizando OpenCV Python

Recuperação de Informação Multimídia em Big Data Utilizando OpenCV Python

Image Registration Algorithm Based on Convolutional Neural Network and Local Homography Transformation

NCC Based Correspondence Problem for First- and Second-Order Graph Matching

Contact Info

Product

Resources

About