
Hyperspectral Images are worthwhile data for many processing algorithms (e.g. Dimensionality Reduction, Target Detection, Change Detection, Classification and Unmixing). Classification is a key issue in processing hyperspectral images. Spectral-identification-based algorithms are sensitive to spectral variability and noise in acquisition. There are many algorithms for classification. This paper describes a new framework for classification of hyperspectral images, based on both spectral and spatial information. The spatial information is obtained by an enhanced Marker-based Hierarchical Segmentation (MHS) algorithm. The hyperspectral data is first fed into the Multi-Layer Perceptron (MLP) neural network classification algorithm. Then, the MHS algorithm is applied in order to increase the accuracy of less-accurately classified land-cover types. In the proposed approach, the markers are extracted from the classification maps obtained by MLP and Support Vector Machines (SVM) classifiers. Experimental results on Quebec City hyperspectral dataset, demonstrate that the proposed approach achieves approximately 9% and 5% better overall accuracy than the MLP and the original MHS algorithms respectively.
The invention of Ground Penetrating Radar (GPR) technology has facilitated the possibility of detecting buried utilities and has been used primarily in civil engineering for detecting structural defects, such as voids and cavities in road pavements, slabs and bridge decks, but has not been used to assess the condition of buried pipes. Pipe deterioration can be defined as pipes where, for example, cracking, differential deflection, missing bricks, collapses, holes, fractures and corrosion exists. Assessing the deterioration of underground pipes is important for service efficiency and asset management. This paper describes a research project that focused on the use of GPR for assessing the condition of buried pipes. The research involved the construction of a suitable GPR test facility in the laboratory to conduct controlled testing in a dry sand. Plastic pipes were chosen for the experiments. A series of laboratory experiments were conducted to determine the validity and effectiveness of standard commercially available GPR technology in assessing the condition of buried utilities with common types of damage. Several types of damage to the plastic pipe were investigated with respect to different GPR antenna frequencies. The GPR surveys were carried out in order to obtain signal signatures from damaged and undamaged pipes buried at 0.5m depth. These surveys were organised on a grid pattern across the surface of the sand in the test facility. The results presented in this paper show that GPR can identify certain types of damage associated with a buried pipe under these controlled laboratory conditions.
Sensor calibration, image orientation, object extraction and scene understanding from images and image sequences are important research topics in Photogrammetry, Remote Sensing, Computer Vision and Geoinformation Science. Within these areas, both geometry and semantics play an important role, and high quality results require appropriate handling of all these aspects. While individual algorithms differ according to the imaging geometry and the employed sensors and platforms, all mentioned aspects need to be integrated in a suitable workflow to solve most real-world problems.
The efficiency and high mobility of Unmanned Aerial Vehicle (UAV) made them essential to aerial photography assisted survey and mapping. Especially for urban land use and land cover, that they often changes, and need UAVs to obtain new terrain data and the new changes of land use. This study aims to collect image data and three dimensional ground control points in Taichung city area with Unmanned Aerial Vehicle (UAV), general camera and Real-Time Kinematic with positioning accuracy down to centimetre. The study area is an ecological park that has a low topography which support the city as a detention basin. A digital surface model was also built with Agisoft PhotoScan, and there will also be a high resolution orthophotos. There will be two conditions for this study, with or without ground control points and both were discussed and compared for the accuracy level of each of the digital surface models. According to check point deviation estimate, the model without ground control points has an average two-dimension error up to 40 centimeter, altitude error within one meter. The GCP-free RTK-airborne approach produces centimeter-level accuracy with excellent to low risk to the UAS operators. As in the case of the model with ground control points, the accuracy of x, y, z coordinates has gone up 54.62%, 49.07%, and 87.74%, and the accuracy of altitude has improved the most.
Mobile mapping often occurs in environments with poor or no GNSS signal reception (tunnels, urban canons, canopy …). Most mobile scanning systems are equipped with tactical grade inertial systems, GNSS outages longer than 30-60 seconds may lead to a rapid decrease in absolute positioning accuracy that might be below the client's expectations. In order to guarantee sufficient positioning accuracy even in such environments the data must be corrected. This can either happen by readjusting the final point cloud using control points (3D translations) or by directly correcting the trajectory by adding external position updates. The approach presented in this paper consists in detecting targets (either existing or placed before measurement) that can easily be identified in the point cloud and that have been measured independently. By identifying the 3D point closest to the target's center, computing the coordinate difference to the corresponding GCP and retrieving its GNSS-timestamp and the internal scanner coordinates, an external position update at a given time can be computed. This procedure has proven its efficiency in several projects including scanning tunnels up to 5km length, where the positional error (in 3D) of the resulting point cloud could be reduced from 5m (in the middle of the tunnel) down to better than 5cm using GCPS only every 400m.