Jiuxiang Hu

Road Network Extraction and Intersection Detection From Aerial Images by Tracking Road Footprints

In this paper, a new two-step approach (detecting and pruning) for automatic extraction of road networks from aerial images is presented. The road detection step is based on shape classification of a local homogeneous region around a pixel. The local homogeneous region is enclosed by a polygon, called the footprint of the pixel. This step involves detecting road footprints, tracking roads, and growing a road tree. We use a spoke wheel operator to obtain the road footprint. We propose an automatic road seeding method based on rectangular approximations to road footprints and a toe-finding algorithm to classify footprints for growing a road tree. The road tree pruning step makes use of a Bayes decision model based on the area-to-perimeter ratio (the A/P ratio) of the footprint to prune the paths that leak into the surroundings. We introduce a lognormal distribution to characterize the conditional probability of A/P ratios of the footprints in the road tree and present an automatic method to estimate the parameters that are related to the Bayes decision model. Results are presented for various aerial images. Evaluation of the extracted road networks using representative aerial images shows that the completeness of our road tracker ranges from 84% to 94%, correctness is above 81%, and quality is from 82% to 92%.

Curve matching for open 2D curves

We present a curve matching framework for planar open curves under similarity transform based on a new scale invariant signature. The signature is derived from the concept of integral of unsigned curvatures. If one input curve as a whole can be aligned with some part in the second curve then the algorithm will find the requisite starting and end positions and will estimate the similarity transform in O(Nlog(N)) time. We extend our frame work to a more general case where some part of the first input curve can be aligned with some part of the second input curve. This is a more difficult problem that we solve in O(N^3) time. The contributions of the paper are the new signature as well as faster algorithms for matching open 2D curves. We present examples from diverse application set to show that our algorithm can work across several domains.

Interactive Hyperspectral Image Visualization Using Convex Optimization

In this paper, we propose a new framework to visualize hyperspectral images. We present three goals for such a visualization: 1) preservation of spectral distances; 2) discriminability of pixels with different spectral signatures; 3) and interactive visualization for analysis. The introduced method considers all three goals at the same time and produces higher quality output than existing methods. The technical contribution of our mapping is to derive a simplified convex optimization from a complex nonlinear optimization problem. During interactive visualization, we can map the spectral signature of pixels to red, green, and blue colors using a combination of principal component analysis and linear programming. In the results, we present a quantitative analysis to demonstrate the favorable attributes of our algorithm.

A stereo photogrammetric technique applied to orographic convection

Abstract This paper describes a technique for photogrammetric analysis of stereo pairs of images that is applied to the study of orographic convection. The technique is designed for use with digital images and assumes detailed knowledge of the camera properties (focal length and imaging chip) and that the position and orientation are known as a first guess. An iterative scheme using known landmarks on the frame is used to determine the camera orientation. The scheme is accurate to 10–100 m at a distance of 15 km from the camera pair. The transition from shallow to deep convection over the Santa Catalina Mountains in southern Arizona on 26 July 2005 is presented. The three-dimensional structure of the visible portion of the cloud is determined and compared with the composite reflectivity from the National Weather Service Weather Surveillance Radar-1988 Doppler radar and the tropopause height from the 1200 UTC sounding in Tucson, Arizona, providing additional validation of the scheme. The shallow to deep tr...

A new image registration scheme based on curvature scale space curve matching

We propose a new image registration scheme for remote sensing images. This scheme includes three steps in sequence. First, a segmentation process is performed on the input image pair. Then the boundaries of the segmented regions in two images are extracted and matched. These matched regions are called confidence regions. Finally, a non-linear optimization is performed in the matched regions only to obtain a global set of transform parameters. Experiments show that this scheme is more robust and converges faster than registration of the original image pair. We also develop a new curve-matching algorithm based on curvature scale space to facilitate the second step.

Evolution of the Vertical Thermodynamic Profile during the Transition from Shallow to Deep Convection during CuPIDO 2006

Abstract The evolution of the vertical thermodynamic profile associated with two cases of deep orographic convection were studied with data from an instrumented aircraft, mobile surface based radiosondes, and stereo photogrammetric analyses. The data were collected during a field experiment [i.e., the Cumulus Photogrammetric, In Situ, and Doppler Observations (CuPIDO) experiment in 2006] performed over the Santa Catalina Mountains in southern Arizona. In both cases the vertical thermodynamic profile was modified in a way that supported subsequent deep convection. In one case, a midtropospheric stable layer was eroded through low-level warming and cooling at the cloud-top level that was likely due to an adiabatic adjustment of the profile through the action of gravity waves. In the second case, dry air aloft was moistened through the action of the shallow convection thus preventing the erosion of the convective turrets through entrainment of dry air. These cases illustrate mechanisms for convective conditi...

Geometric Calibration of Digital Cameras for 3D Cumulus Cloud Measurements

Abstract A technique for calibrating digital cameras for stereo photogrammetry of cumulus clouds is presented. It has been applied to characterize the formation of summer thunderstorms observed during the Cumulus Photogrammetric, In Situ, and Doppler Observations (CuPIDO) project. Starting from gross measurements of locations, orientations of cameras, and landmark surveys, accurate locations and orientations of the cameras are obtained by minimizing a geometric error (GE). Once accurate camera parameters are obtained, 3D positions of cloud-feature points are computed by triangulation. The main contributions of this paper are as follows. First, it is proven that the GE has only one minimum in the neighborhood of the real parameters of a camera. In other words, searching the minimum of the GE enables the authors to find the right camera parameters even if there are significant differences between the initial measurements and their true values. Second, a new coarse-to-fine iterative algorithm is developed th...

Color-to-gray conversion using ISOMAP

In this paper we present a new algorithm to transform an RGB color image to a grayscale image. We propose using nonlinear dimension reduction techniques to map higher dimensional color vectors to lower dimensional ones. This approach generalizes the gradient domain manipulation for high dimensional images. Our experiments show that the proposed algorithm generates competitive results and reaches a good compromise between quality and speed.

Fourier Shape Descriptors of Pixel Footprints for Road Extraction from Satellite Images

In this paper, an automatic road tracking method is presented for detecting roads from satellite images. This method is based on shape classification of a local homogeneous region around a pixel. The local homogeneous region is enclosed by a polygon, called the pixel footprint. We introduce a spoke wheel operator to obtain the pixel footprint and propose a Fourier-based approach to classify footprints for automatic seeding and growing of the road tracker. We experimentally demonstrate that our proposed road tracker can extract the centerlines of roads with sharp turns and intersections effectively, and has relatively small amount of leakage.

An Algorithm to Calibrate Field Cameras for Stereo Clouds

This paper presents a robust extrinsic parameter estimation algorithm to calibrate field cameras which were used to observe the formation of clouds on a mountainous region. Generally, camera calibration needs accurate landmark survey and image feature identification. However, our observation area is a large scale scene in a physically inaccessible area, therefore the landmark surveys are not precise. Since clouds are distant to cameras, cloud features in the images are also difficult to accurately identify for stereo correspondences. The noise in landmark survey and cloud feature correspondence makes it challenging to obtain desired cloud observation accuracy by using traditional least squares based camera calibration approaches. Our camera calibration approach is based on a generalized total least square (GTLS) algorithm instead of a normal least square method. Experiments show that the GTLS-based camera calibration is more accurate and robust than LS-based methods for our application.