Accurate point matching is a critical and challenging process in feature-based image registration. In this paper, simple robust feature algorithm, called Restricted Spatial Order Constraints (RSOC), proposed to remove outliers for registering aerial images with monotonous backgrounds, similar patterns, low overlapping areas, large affine transformation. RSOC, both local structure global information are considered. Based on adjacent spatial order, an invariant descriptor defined, formulated...

The detection of targets with complex backgrounds in aerial images is a challenging task. In this letter, we propose robust insulator algorithm based on local features and spatial orders for images. First, detect introduce multiscale multifeature descriptor to represent the features. Then, get several by training these features, it improves robustness algorithm. Finally, through coarse-to-fine matching strategy, eliminate background noise determine region insulators. We test our method...

The objects in natural images are often texturally inhomogeneous and prone to be falsely segmented into different parts by conventional methods. To overcome the difficulties caused texture inhomogeneity, a new active contour model is proposed extract insulators from aerial images. First, semilocal operator employed features of under Beltrami framework. layer feature single, thus, it can avoid high dimensionality space. Then, convex energy functional defined taking Xie's nonconvex global...

In this paper, a novel texture segmentation algorithm is proposed to partition the complex aerial insulator images into sub-regions with closed smooth contours. Firstly, Gray Level Co-occurrence Matrix (GLCM) employed extract features of insulators and calculated by rapid Integrated Algorithm (GLCIA). We divide extracted two categories: one stronger discriminative ability other weaker ability. The second category optimized Principal Component Analysis (PCA) better distinguish different...

Oil slicks and lookalikes (e.g., plant oil emulsion) all appear as dark areas in polarimetric Synthetic Aperture Radar (SAR) images are highly heterogeneous, so it is very difficult to use a single feature that can allow classification of objects SAR or lookalikes. We established multi-feature fusion support the discrimination In paper, simple analysis used rationalize preferred features subset. The analyzed include entropy, alpha, Single-bounce Eigenvalue Relative Difference (SERD) C-band...

Synthetic aperture radar automatic target recognition (SAR ATR) is a key technique of remote-sensing image recognition, which has many potential applications in the fields military surveillance, national defense, civil application, and so on. With development science technology, deep convolutional neural network (DCNN) been widely applied for SAR ATR. However, it difficult to use learning train models with limited ray images. To resolve this problem, we proposed an effectively lightweight...

The phenomenon of spectral aliasing exists for coastal wetland object types, which leads to class mixing. This letter proposes a multiobject convolutional neural network (CNN) decision fusion classification method hyperspectral images wetlands. adopts based on fuzzy membership rules applied single-object CNN obtain higher accuracy. Experimental results demonstrate the effectiveness proposed six including water, tidal flat, reed, and other vegetation types. overall accuracy is 82.11%, 3.33%...

Edge detection is a crucial approach for the location and acreage calculation of oil slick when spills on sea. In this paper, in view intensity inhomogeneity, high noise, blurring infrared (IR) aerial images, novel algorithm proposed to detect edges IR images. algorithm, we define an energy function model combining region-scalable-fitting concept global minimization active contour (GMAC) model. The avoids existence local minima meanwhile deals with weak edge boundaries exiting spill process...

Damping Bragg scattering from the ocean surface is basic underlying principle of synthetic aperture radar (SAR) oil slick detection, and they produce dark spots on SAR images. Dark spot detection first step in spill which affects accuracy detection. However, some natural phenomena (such as waves, currents, low wind belts, well human factors) may change backscatter intensity sea, resulting uneven intensity, high noise, blurred boundaries slicks or lookalikes. In this paper, Segnet used a...

In this paper, a new method of insulator fault detection by texture feature sequence is proposed. Morphology, Hough transform line and statistic are applied to method. Because there some noises during image shooting, it necessary that the should be preprocessed corrected. The preprocessing includes grayness, enhancement morphological processing. tilt correction based on detection. Then, processed separated into ten parts. Each part parts featured seven values. matrix constructed columns...

In this paper, a method of tempered glass insulator recognition based on intensity in HSI modal from airborne image by helicopter power grid is proposed. Morphology, extraction connected components and threshold value segmentation mean are applied to method. The contains the important information image, which base segmentation. According feature it helpful that become binary image. Because there some noises necessary should be preprocessed reduce noise. erode process morphology used here....

In hope of developing a method for oil spill detection in laser remote sensing, series refined and crude samples were investigated using time-resolved fluorescence conjunction with parallel factors analysis (PARAFAC). The time resolved emission spectra those taken by sensing system on laboratory basis distance 5 m. Based the intensity-normalized spectra, both well classified without overlapping, approach PARAFAC four factors. Principle component (PCA) has also been operated as comparison. It...

Liu, R.-J.; Zhang, J.; Cui, B.-G.; Ma, Y.; Song, P.J., and An, J.-B., 2019. Red tide detection based on high spatial resolution broad band satellite data: A case study of GF-1. In: Jung, H.-S.; Lee, S.; Ryu, J.-H., T. (eds.), Advances in Remote Sensing Geoscience Information Systems Coastal Environments. Journal Research, Special Issue No. 90, pp. 120-128. Coconut Creek (Florida), ISSN 0749-0208.Traditional red remote sensing methods are ocean color data with spectral (spectral width < 20...

Accurate point matching is widely used, and it a critical challenging process in feature-based image registration. To improve feature accuracy on putative matches with heavy outliers similar local structures, an accurate robust algorithm based minimum relative motion entropy (MRME) proposed, which the between their K-nearest neighbors formulated. Based clustering result, defined to find coincident motions. According motions MRME, are removed two-stage match strategy. With quasi-linear time...

Image registration is a crucial progress in detecting oil spilled on the sea and also important for estimating volume of spill, especially when one image cannot cover entire polluted region. In this article, new algorithm proposed to register geometrically distorted aerial images spill accurately automatically. There are two stages algorithm: coarse fine registration. Invariants-based similarity relative space distance applied matching. Then improved iterative closest point (ICP) used...