Ground Control Point Extraction Research Articles

A CNN-Based Layer-Adaptive GCPs Extraction Method for TIR Remote Sensing Images

Ground Control Points (GCPs) are of great significance for applications involving the registration and fusion of heterologous remote sensing images (RSIs). However, utilizing low-level information rather than deep features, traditional methods based on intensity and local image features turn out to be unsuitable for heterologous RSIs because of the large nonlinear radiation difference (NRD), inconsistent resolutions, and geometric distortions. Additionally, the limitations of current heterologous datasets and existing deep-learning-based methods make it difficult to obtain enough precision GCPs from different kinds of heterologous RSIs, especially for thermal infrared (TIR) images that present low spatial resolution and poor contrast. In this paper, to address the problems above, we propose a convolutional neural network-based (CNN-based) layer-adaptive GCPs extraction method for TIR RSIs. Particularly, the constructed feature extraction network is comprised of basic and layer-adaptive modules. The former is used to achieve the coarse extraction, and the latter is designed to obtain high-accuracy GCPs by adaptively updating the layers in the module to capture the fine communal homogenous features of the heterologous RSIs until the GCP precision meets the preset threshold. Experimental results evaluated on TIR images of SDGSAT-1 TIS and near infrared (NIR), short wave infrared (SWIR), and panchromatic (PAN) images of Landsat-8 OLI show that the matching root-mean-square error (RMSE) of TIS images with SWIR and NIR images could reach 0.8 pixels and an even much higher accuracy of 0.1 pixels could be reached between TIS and PAN images, which performs better than those of the traditional methods, such as SIFT, RIFT, and the CNN-based method like D2-Net.

Automatic Satellite Images Orthorectification Using K–Means Based Cascaded Meta-Heuristic Algorithm

Orthorectification of high-resolution satellite images using a terrain- dependent rational function model (RFM) is a difficult task requiring a well-distributed set of ground control points (GCPs), which is often time-consuming and costly operation. Further, RFM is sensitive to over-parameterization due to its many coefficients, which have no physical meaning. Optimization-based meta-heuristic algorithms ap- pear to be an efficient solution to overcome these limitations. This pa- per presents a complete automated RFM terrain-dependent orthorec- tification for satellite images. The proposed method has two parts; the first part suggests automating the GCP extraction by combing Scale- Invariant Feature Transform and Speeded Up Robust Features algo- rithms; and the second part introduces the cascaded meta-heuristic al- gorithm using genetic algorithms and particle swarm optimization. In this stage, a modified K-means clustering selection technique was used to support the proposed algorithm for finding the best combinations of GCPs and RFM coefficients. The obtained results are promising in terms of accuracy and stability compared to other literature methods.

TECHNIQUES FOR COMPARING MULTI-TEMPORAL ARCHIVE AERIAL IMAGERY FOR GLACIER MONITORING WITH POOR GROUND CONTROL

Abstract. The measurement of geometric changes in Alpine glaciers is an essential aspect to assess their reaction to climate change effects. Archive aerial images may integrate valuable information to this purpose at times when other types of remotely sensed data are not available. The application of photogrammetric techniques such as Structure-from-Motion (SfM) and Multi-View-Stereo matching allows the extraction of dense point clouds to model the glacier environment. The comparison of multiple datasets requires to setup a stable reference system, a task that in archive photos is generally carried out by means of ground control points (GCPs). This paper would like to propose and assess some techniques to cope with the lack of ground control. Multitemporal SfM (MSfM) is presented and tested on a dataset including six different aerial blocks collected by means of analogue (PAN/RGB) and digital airborne cameras from 1967 to 2006. These images have been downloaded from the IGNF online repository and cover the area of the Val Veny (Brenva and Miage glaciers) in the Mount Blanc massif, at the border between Italy and France. Coupled with other solutions (i.e., extraction of GCPs from maps and DTMs and ICP co-registration of point clouds), MSfM has revealed as a suitable technique for coregistration of multiple photogrammetric blocks of aerial photos with minimum ground control. Some tests carried out in the case study area demonstrated that the integration of MSfM and ICP coregistration for refinement may significantly improve the comparison between multiple point clouds, which is a fundamental pre-requisite for the analysis of glacier changes over time.

Robust Automatic Generation of True Orthoimages From Very High-Resolution Panchromatic Satellite Imagery Based on Image Incidence Angle for Occlusion Detection

Orthoimages have proven to provide useful reference data for a variety of geospatial applications. Their improved versions, true orthoimages, are even more valuable as they have been adjusted also for high objects. Even though true orthoimages have been-over the last decades-generated mainly from aerial images, we believe there is a need to produce true orthoimages also from satellite images. The aim of this article is to present the already developed automatic procedure for generating true orthoimages from very high-resolution optical satellite images. The automatic workflow consists of five modules, starting with the extraction of ground control points, through the geometric processing of image blocks, occlusion detection, orthorectification, and finally the generation of true orthoimages. The orthorectification module uses a high-resolution digital surface model derived from laser scanning data. Occlusion detection is performed using a ray-tracing method based on the incidence angle, which is suitable for images acquired by satellites and processed with rational function models. The usability of the workflow was demonstrated with the test of six Pléiades images that were acquired on different dates and with different viewing angles. The research presents the accuracy of the geometric correction and the quality of the true orthoimage generated with different numbers of images. The results show that satellite-based true orthoimages are suitable for the extraction of spatial information that can be used directly by the end-users.

Automatic Orthorectification of High-Resolution Optical Satellite Images Using Vector Roads

This paper presents a completely automatic processing chain for orthorectification of optical pushbroom sensors. The procedure is robust and works without manual intervention from raw satellite image to orthoimage. It is modularly divided in four main steps: metadata extraction, automatic ground control point (GCP) extraction, geometric modeling, and orthorectification. The GCP extraction step uses georeferenced vector roads as a reference and produces a file with a list of points and their accuracy estimation. The physical geometric model is based on collinearity equations and works with sensor-corrected (level 1) optical satellite images. It models the sensor position and attitude with second-order piecewise polynomials depending on the acquisition time. The exterior orientation parameters are estimated in a least squares adjustment, employing random sample consensus and robust estimation algorithms for the removal of erroneous points and fine-tuning of the results. The images are finally orthorectified using a digital elevation model and positioned in a national coordinate system. The usability of the method is presented by testing three RapidEye images of regions with different terrain configurations. Several tests were carried out to verify the efficiency of the procedure and to make it more robust. Using the geometric model, subpixel accuracy on independent check points was achieved, and positional accuracy of orthoimages was around one pixel. The proposed procedure is general and can be easily adapted to various sensors.

Particle filtering methods for georeferencing panoramic image sequence in complex urban scenes

Georeferencing image sequences is critical for mobile mapping systems. Traditional methods such as bundle adjustment need adequate and well-distributed ground control points (GCP) when accurate GPS data are not available in complex urban scenes. For applications of large areas, automatic extraction of GCPs by matching vehicle-born image sequences with geo-referenced ortho-images will be a better choice than intensive GCP collection with field surveying. However, such image matching generated GCP’s are highly noisy, especially in complex urban street environments due to shadows, occlusions and moving objects in the ortho images. This study presents a probabilistic solution that integrates matching and localization under one framework. First, a probabilistic and global localization model is formulated based on the Bayes’ rules and Markov chain. Unlike many conventional methods, our model can accommodate non-Gaussian observation. In the next step, a particle filtering method is applied to determine this model under highly noisy GCP’s. Owing to the multiple hypotheses tracking represented by diverse particles, the method can balance the strength of geometric and radiometric constraints, i.e., drifted motion models and noisy GCP’s, and guarantee an approximately optimal trajectory. Carried out tests are with thousands of mobile panoramic images and aerial ortho-images. Comparing with the conventional extended Kalman filtering and a global registration method, the proposed approach can succeed even under more than 80% gross errors in GCP’s and reach a good accuracy equivalent to the traditional bundle adjustment with dense and precise control.

Automatic GCP Extraction in Mountainous Areas Using DEM and PolSAR Data

Mountainous areas in synthetic aperture radar (SAR) images suffer severe geometric distortions caused by different look directions. Consequently, ground control point (GCP) extraction hardly obtains accurate results for aircraft positioning. Based on the digital elevation model (DEM) and polarimetric SAR (PolSAR) data, we propose a method for extracting GCPs in mountainous areas by introducing the polarization orientation angle shift (POAS) to minimize geometric distortions. In this method, DEM data are used as the reference map by providing POASs at arbitrary look directions to make up for the look-direction sensitivity of POASs transformed from PolSAR data. The geometric distortions between POASs transformed from DEM and PolSAR data are effectively reduced by calculating the POASs from DEM data at the same look direction and look angles of the PolSAR data. In contrast to the SAR data, which have a large dynamic range, the values of POAS are limited to a small interval. Therefore, the illumination distortions induced by visualization can be reduced. Finally, the GCP extraction between the POAS images is conducted by bilateral filter scale-invariant feature transform. Experiments using various data at different look directions demonstrate that the proposed method obtains better-quality GCPs but less invalid keypoints than the method using only intensity images for mountainous areas.

Automatic GCP Extraction of Fully Polarimetric SAR Images

This paper presents a method for automatic extraction of ground control points (GCPs) of fully polarimetric synthetic aperture radar (SAR) (PolSAR) images obtained from various satellites with different viewing angles. The scale-invariant feature transform (SIFT) algorithm is applied to extract candidate GCPs, where two-way keypoint matching eliminates improbable correspondence keypoints. Minimizing the root-mean-square error (rmse) also removes matching points with large rmse through a pseudoaffine transformation. In addition, information entropy and spatial dispersion quality constraints enable quantification of the spatial distribution of the GCPs. In accordance with full polarization, applying the SIFT-OCT algorithm (SIFT algorithm with the first scale-space octave skipped) to PolSAR data is examined. The total power (TP) image represents a combination of the characteristics of all four polarization images [horizontal transmitting and horizontal receiving (HH), horizontal transmitting and vertical receiving (HV), vertical transmitting and horizontal receiving (VH), and vertical transmitting and vertical receiving (VV)]. Therefore, GCP extraction using a TP image rather than each polarization image is proposed in order to maximize the accuracy of GCP extraction for all of the polarization data, as the TP image generates the highest signal-to-noise ratio (SNR) value. The SNR in conjunction with the matching correlation surface is used as an indicator of the reliability and accuracy of GCP extraction. After successfully applying the method to Advanced Land Observing Satellite/Phased Array type L-band Synthetic Aperture Radar and Japanese Earth Resources Satellite-1 SAR images, the GCP matching accuracy is further improved by using geometric calibration.

Extraction of ground control points (GCPs) from synthetic aperture radar images and SRTM DEM

Qualified ground control points (GCPs) are crucial in the geolocation of a remotely sensed image. If a region has no published map, the geographical coordinates of GCPs must be obtained indirectly. Although these can be re‐constructed from a synthetic aperture radar (SAR) image using sensor position and velocity, this conventional method does not provide accurate GCPs owing to the variable elevations of actual topography. We propose an inverse geolocation method for GCPs by SAR simulation. This improves the accuracy of extracted GCPs by accommodating topographic effects, and requires a high‐resolution digital elevation model (DEM) and SAR with precise orbit data. The errors in the northing derive mainly from the imaging pulse time, and were improved by applying an azimuth time correction. In this study ERS SAR data with precise Delft orbit and shuttle radar topography mission (SRTM) DEM were used. Mean values for the planimetric distance error in the GCPs were 4.1 and 5.4 m with standard deviations of 10.6 and 16.9 m in northing and easting, respectively. The obtained GCPs were applied to an IRS image for geo‐rectification, and the result was mean image positional errors of 3.6 and 2.7 m with standard deviations of 8.4 and 15.4 m in northing and easting, respectively. The root mean square errors are 9.0 and 15.3 m in northing and easting, respectively.