Automatic Extraction of Control Points for the Registration of Optical Satellite and LiDAR Images

Palenichka, Roman M.; Zaremba, Marek B.

doi:10.1109/tgrs.2010.2043677

Cited by 58 publications

(36 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The general overview of a salientpoint detector in the context of image matching is given in [31,32]. Salient points in images refer to image locations with distinctive features providing discrimination between objects of interest and the background [33]. A salient-point detector is a visual operator used to find image locations or regions containing object of interests [34][35][36][37][38][39].…”

Section: Salient-point Based Registrationmentioning

confidence: 99%

“…A salient-point visual operator can detect various types of salient features such as corners, lines, junctions, intensity blobs, and homogenous regions [16,35,40,41]. Although, these techniques have proven their effectiveness for control-point extraction, there are, however, many weaknesses [33]. Feature extraction and salient point detection using these techniques gives poor results at a large scale during image analysis.…”

Section: Salient-point Based Registrationmentioning

confidence: 99%

“…To overcome the aforesaid shortcomings [33] proposeda new method of control point extraction using a salient visual operator. The purposed solution was designed with the intention to registeroptical images and LiDARdata automatically with high accuracy.The development of control point extraction schemes starts with an extraction of image fragments which satisfy given requirements for intensity and shape features(local features).…”

Section: Salient-point Based Registrationmentioning

confidence: 99%

“…The multiscale SID concept is anextension of a salient point method of feature extraction formultiscalehigh-contrast homogenous regions [34,35]. "A SID is defined [33] as a circular image fragment of a variable diameter, which is inscribed into a homogeneous region (i.e., is tangent to the background in at least two points) and has the local maximum of contrast-tohomogeneity ratio". Three SID image descriptors were introduced: 1) planar pose characteristics (e.g.…”

Section: Salient-point Based Registrationmentioning

confidence: 99%

See 3 more Smart Citations

A Review of Optical Imagery and Airborne LiDAR Data Registration Methods

Mishra

Zhang²

2012

TORMSJ

View full text Add to dashboard Cite

Representing a scene completely from remote sensing data requires both spectral and 3-D-surface information. Integration of spectral information from optical images and 3-D-surface information from LiDAR is important in a number of remote sensing applications such as feature extraction, image classification, image analysis, building extraction, 3-D city modelling, canopy modelling etc. Therefore, numerous methods have been developed in the last decade to align both data sets into a common reference frame to effectively utilize their complementary characteristics. However, due to the significantly different characteristics between optical image and LiDAR data, there are a number of technical challenges in the alignment of both data sets. Different research papers introduced different strategy or methodology to overcome the challenges, reaching different alignment/registration results. This paper presents a review of classical and up to date optical-LiDAR registration methods with the emphasis on control point detection and matching. The aim of this paper is to provide readers with an overview of existing methods, identify their advantages and limitations, and give readers the overall information on what will be useful for researchers and practitioners to realistically select proper method for their application.

show abstract

Section: Salient-point Based Registrationmentioning

confidence: 99%

Section: Salient-point Based Registrationmentioning

confidence: 99%

Section: Salient-point Based Registrationmentioning

confidence: 99%

Section: Salient-point Based Registrationmentioning

confidence: 99%

See 2 more Smart Citations

A Review of Optical Imagery and Airborne LiDAR Data Registration Methods

Mishra

Zhang²

2012

TORMSJ

View full text Add to dashboard Cite

show abstract

“…RANSAC is an iterative method to estimate parameters of a mathematical model from a set of observed data. Therefore, RANSAC has been widely used in remote sensing registration (Huo et al, 2012;Palenichka and Zaremba, 2010;Cao et al, 2013). For remote sensing image registration, a fundamental matrix is typically used to detect outliers, which are false matching points.…”

Section: Introductionmentioning

confidence: 99%

A Robust False Matching Points Detection Method for Remote Sensing Image Registration

Shan

Tang

2015

Int. Arch. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

Commission VI, WG VI/4 KEY WORDS: Image registration, False matching points detection, RANSAC, GTM, K-NN graph, SIFT ABSTRACT:Given the influences of illumination, imaging angle, and geometric distortion, among others, false matching points still occur in all image registration algorithms. Therefore, false matching points detection is an important step in remote sensing image registration. Random Sample Consensus (RANSAC) is typically used to detect false matching points. However, RANSAC method cannot detect all false matching points in some remote sensing images. Therefore, a robust false matching points detection method based on Knearest-neighbour (K-NN) graph (KGD) is proposed in this method to obtain robust and high accuracy result. The KGD method starts with the construction of the K-NN graph in one image. K-NN graph can be first generated for each matching points and its K nearest matching points. Local transformation model for each matching point is then obtained by using its K nearest matching points. The error of each matching point is computed by using its transformation model. Last, L matching points with largest error are identified false matching points and removed. This process is iterative until all errors are smaller than the given threshold. In addition, KGD method can be used in combination with other methods, such as RANSAC. Several remote sensing images with different resolutions and terrains are used in the experiment. We evaluate the performance of KGD method, RANSAC + KGD method, RANSAC, and Graph Transformation Matching (GTM). The experimental results demonstrate the superior performance of the KGD and RANSAC + KGD methods.

show abstract

A Regenerated Feature Extraction Method for Cross-modal Image Registration

Yang

Wang

2018

Advances in Brain Inspired Cognitive Systems

View full text Add to dashboard Cite

Automatic Extraction of Control Points for the Registration of Optical Satellite and LiDAR Images

Cited by 58 publications

References 31 publications

A Review of Optical Imagery and Airborne LiDAR Data Registration Methods

A Review of Optical Imagery and Airborne LiDAR Data Registration Methods

A Robust False Matching Points Detection Method for Remote Sensing Image Registration

A Regenerated Feature Extraction Method for Cross-modal Image Registration

Contact Info

Product

Resources

About