In this study, a deep learning (DL)-based approach is proposed for the detection and reconstruction of buildings from a single aerial image. The pre-required knowledge to reconstruct the 3D shapes of buildings, including the height data as well as the linear elements of individual roofs, is derived from the RGB image using an optimized multi-scale convolutional–deconvolutional network (MSCDN). The proposed network is composed of two feature extraction levels to first predict the coarse features, and then automatically refine them. The predicted features include the normalized digital surface models (nDSMs) and linear elements of roofs in three classes of eave, ridge, and hip lines. Then, the prismatic models of buildings are generated by analyzing the eave lines. The parametric models of individual roofs are also reconstructed using the predicted ridge and hip lines. The experiments show that, even in the presence of noises in height values, the proposed method performs well on 3D reconstruction of buildings with different shapes and complexities. The average root mean square error (RMSE) and normalized median absolute deviation (NMAD) metrics are about 3.43 m and 1.13 m, respectively for the predicted nDSM. Moreover, the quality of the extracted linear elements is about 91.31% and 83.69% for the Potsdam and Zeebrugge test data, respectively. Unlike the state-of-the-art methods, the proposed approach does not need any additional or auxiliary data and employs a single image to reconstruct the 3D models of buildings with the competitive precision of about 1.2 m and 0.8 m for the horizontal and vertical RMSEs over the Potsdam data and about 3.9 m and 2.4 m over the Zeebrugge test data.
13 November 2017 in ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Nowadays, Unmanned Aerial System (UAS)-based photogrammetry offers an affordable, fast and effective approach to real-time acquisition of high resolution geospatial information and automatic 3D modelling of objects for numerous applications such as topography mapping, 3D city modelling, orthophoto generation, and cultural heritages preservation. In this paper, the capability of four different state-of-the-art software packages as 3DSurvey, Agisoft Photoscan, Pix4Dmapper Pro and SURE is examined to generate high density point cloud as well as a Digital Surface Model (DSM) over a historical site. The main steps of this study are including: image acquisition, point cloud generation, and accuracy assessment. The overlapping images are first captured using a quadcopter and next are processed by different software to generate point clouds and DSMs. In order to evaluate the accuracy and quality of point clouds and DSMs, both visual and geometric assessments are carry out and the comparison results are reported.
18 October 2019 in The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Forest inventory provides comprehensive information about the geometric and biometric state of forests as well as vegetated areas. In this study, a point-based 3D method is presented for tree detection as well as measuring of structural properties of forests such as the tree height, tree position and canopy area using high resolution point cloud which is provided by an Unmanned Aerial Vehicle (UAV)-based LiDAR sensor. The proposed method is based on the density of point cloud and 2D and 3D distance measurements. It includes three main steps as pre-processing, tree detection, and extraction of tree structural attributes. After generating a canopy height model, an image is created based on the density of point cloud. Next, points are classified based on 2D and 3D distance measurements, sequentially, from the highest to the lowest. According to the results, the rate of tree detection is about 95% and the main structural parameters of a tree such as the position, height, area and length of the canopy are estimated with the accuracy of 1.97 m, 0.36 m, 12.78 m2 and 0.79 m, respectively.
18 October 2019 in The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Automatic detection and extraction of buildings from aerial images are considerable challenges in many applications, including disaster management, navigation, urbanization monitoring, emergency responses, 3D city mapping and reconstruction. However, the most important problem is to precisely localize buildings from single aerial images where there is no additional information such as LiDAR point cloud data or high resolution Digital Surface Models (DSMs). In this paper, a Deep Learning (DL)-based approach is proposed to localize buildings, estimate the relative height information, and extract the buildings’ boundaries using a single aerial image. In order to detect buildings and extract the bounding boxes, a Fully Connected Convolutional Neural Network (FC-CNN) is trained to classify building and non-building objects. We also introduced a novel Multi-Scale Convolutional-Deconvolutional Network (MS-CDN) including skip connection layers to predict normalized DSMs (nDSMs) from a single image. The extracted bounding boxes as well as predicted nDSMs are then employed by an Active Contour Model (ACM) to provide precise boundaries of buildings. The experiments show that, even having noises in the predicted nDSMs, the proposed method performs well on single aerial images with different building shapes. The quality rate for building detection is about 86% and the RMSE for nDSM prediction is about 4 m. Also, the accuracy of boundary extraction is about 68%. Since the proposed framework is based on a single image, it could be employed for real time applications.
24 September 2013 in The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
This paper presents a satellite tracking method based on optical celestial images. The proposed method is composed of two acquiring modes that are called Sidereal Stare Mode (SSM) and Track Rate Mode (TRM). To track the unknown satellites, first of all the entire of the sky is observed and the primary knowledge of the satellite location is predicted in SSM. Then, the measured pointing data is calculated and used by the telescope to acquire the satellite tracking images at the appropriate time. The framework of SSM contains main steps such as the image correction due to the noise and the image background by applying the iterative filtering methods, the star detection and removal using a double gate filter and PSF concepts, the star identification for image calibration based on a star catalogue, the streak detection of the satellite using the matched filter and finally, the extraction of the celestial coordinate of the satellite as a predicted position. In TRM, the group of star streaks should be identified in the free background image. Then, Fast Fourier Transformation (FFT) is applied and the characteristic of star streaks are extracted. Finally, the centroid of the satellite is estimated precisely by applying a binary mask and the right ascension and declination of it is calculated using astrometric transformation parameters. In this study, the searching and tracking algorithms is applied on the simulation images to detect and identify the satellite position in the sky. Therefore, a telescope along with a CCD camera is simulated to generate sequences images using the well-known star catalogues, the optical and orbital parameters. The precise celestial coordinate of a real satellite and two fictitious satellites are obtained from simulated images. The study results show that the accuracy of the satellite estimated position after star calibration is better than 5 arc seconds and the satellites tracking accuracy is around 1–3 arc seconds.