A Novel Change Detection Method for Natural Disaster Detection and Segmentation from Video Sequence.

Change detection (CD) is critical for natural disaster detection, monitoring and evaluation. Video satellites, new types of satellites being launched recently, are able to record the motion change during natural disasters. This raises a new problem for traditional CD methods, as they can only detect areas with highly changed radiometric and geometric information. Optical flow-based methods are able to detect the pixel-based motion tracking at fast speed; however, they are difficult to determine an optimal threshold for separating the changed from the unchanged part for CD problems. To overcome the above problems, this paper proposed a novel automatic change detection framework: OFATS (optical flow-based adaptive thresholding segmentation). Combining the characteristics of optical flow data, a new objective function based on the ratio of maximum between-class variance and minimum within-class variance has been constructed and two key steps are motion detection based on optical flow estimation using deep learning (DL) method and changed area segmentation based on an adaptive threshold selection. Experiments are carried out using two groups of video sequences, which demonstrated that the proposed method is able to achieve high accuracy with F1 value of 0.98 and 0.94, respectively.

Ground-level PM2.5 estimation over urban agglomerations in China with high spatiotemporal resolution based on Himawari-8.

High concentrations of particulate matter with diameter of <2.5 µm (PM2.5) demonstrate severe effects on human health, especially in the metropolitan agglomerations of China. Estimating PM2.5 based on satellite aerosol optical depth (AOD) is a widely used method. AOD data from Himawari-8, a geostationary satellite, enable improvement of the temporal resolution of PM2.5 estimates to the hourly level, thereby reflecting diurnal variations of pollutants compared with AOD products from polar orbit satellites, which only have one value per day. In this study, PM2.5 concentrations are estimated based on Himawari-8 AOD and other ancillary data by constructing spatiotemporal linear mixed effects model in Central China (CCH), Beijing-Tianjin-Henan (BTH), Yangtze River Delta (YRD) and Pearl River Delta (PRD) regions, respectively. The determination coefficient (R2) between the measurements and estimates of PM2.5 calculated with the tenfold cross-validation method are 0.82, 0.84, 0.80 and 0.74 in CCH, BTH, YRD and PRD, respectively. The spatial distributions of PM2.5 present large regional variation, which is highly correlated with land-use type. Heavily polluted zones are mainly located in urban or rural areas, which have dense population and high anthropogenic emissions. Comparisons among different seasons show that particle pollution during the cold seasons (autumn and winter) is relatively severe with an average PM2.5 of >60 µg/m3 in CCH, BTH and YRD, whereas the level does not greatly change throughout the year in the PRD region. During the daytime, particulate pollution levels are generally high in the morning.

Feature Selection for Object-Based Classification of High-Resolution Remote Sensing Images Based on the Combination of a Genetic Algorithm and Tabu Search.

In object-based image analysis of high-resolution images, the number of features can reach hundreds, so it is necessary to perform feature reduction prior to classification. In this paper, a feature selection method based on the combination of a genetic algorithm (GA) and tabu search (TS) is presented. The proposed GATS method aims to reduce the premature convergence of the GA by the use of TS. A prematurity index is first defined to judge the convergence situation during the search. When premature convergence does take place, an improved mutation operator is executed, in which TS is performed on individuals with higher fitness values. As for the other individuals with lower fitness values, mutation with a higher probability is carried out. Experiments using the proposed GATS feature selection method and three other methods, a standard GA, the multistart TS method, and ReliefF, were conducted on WorldView-2 and QuickBird images. The experimental results showed that the proposed method outperforms the other methods in terms of the final classification accuracy.

Entropy-Based Registration of Point Clouds Using Terrestrial Laser Scanning and Smartphone GPS.

Automatic registration of terrestrial laser scanning point clouds is a crucial but unresolved topic that is of great interest in many domains. This study combines terrestrial laser scanner with a smartphone for the coarse registration of leveled point clouds with small roll and pitch angles and height differences, which is a novel sensor combination mode for terrestrial laser scanning. The approximate distance between two neighboring scan positions is firstly calculated with smartphone GPS coordinates. Then, 2D distribution entropy is used to measure the distribution coherence between the two scans and search for the optimal initial transformation parameters. To this end, we propose a method called Iterative Minimum Entropy (IME) to correct initial transformation parameters based on two criteria: the difference between the average and minimum entropy and the deviation from the minimum entropy to the expected entropy. Finally, the presented method is evaluated using two data sets that contain tens of millions of points from panoramic and non-panoramic, vegetation-dominated and building-dominated cases and can achieve high accuracy and efficiency.

A "Tuned" Mask Learnt Approach Based on Gravitational Search Algorithm.

Texture image classification is an important topic in many applications in machine vision and image analysis. Texture feature extracted from the original texture image by using "Tuned" mask is one of the simplest and most effective methods. However, hill climbing based training methods could not acquire the satisfying mask at a time; on the other hand, some commonly used evolutionary algorithms like genetic algorithm (GA) and particle swarm optimization (PSO) easily fall into the local optimum. A novel approach for texture image classification exemplified with recognition of residential area is detailed in the paper. In the proposed approach, "Tuned" mask is viewed as a constrained optimization problem and the optimal "Tuned" mask is acquired by maximizing the texture energy via a newly proposed gravitational search algorithm (GSA). The optimal "Tuned" mask is achieved through the convergence of GSA. The proposed approach has been, respectively, tested on some public texture and remote sensing images. The results are then compared with that of GA, PSO, honey-bee mating optimization (HBMO), and artificial immune algorithm (AIA). Moreover, feature extracted by Gabor wavelet is also utilized to make a further comparison. Experimental results show that the proposed method is robust and adaptive and exhibits better performance than other methods involved in the paper in terms of fitness value and classification accuracy.

[Real-time automatic cloud detection during the process of taking aerial photographs].

The present paper adopted a method based on the spectrum signatures with thresholds to detect cloud. Through analyzing the characteristic in the aspect of spectrum signatures of cloud, two effective signatures were explored, one was brightness signature I and the other was normalized difference signature P. Combined with corresponding thresholds, each spectrum condition can detect some cloud pixels. By composing the union of two spectrum conditions together, cloud can be detected more completely. In addition, the threshold was also very important to the accuracy of the detection result. In order to detect cloud efficiently, correctly and automatically, this paper proposed a new strategy about the assignment of thresholds to acquire suitable thresholds. Firstly, the images should be classified into three kinds of types which were images with no cloud, with thin cloud and with thick cloud. Secondly, different assignment methods of automatic thresholds of signatures would be adopted according to different types of images. For images with thick cloud, they would be further classified into three kinds by another standard and assigned by different thresholds integrated by automatic thresholds from other spectrum signatures. The automatic thresholds were acquired by Otsu algorithm and an improved Otsu algorithm. For images with thin cloud, the cloud would be detected by score algorithm. Due to this flexible strategy, cloud in images can be detected rightly and if there isn't cloud in images the detection will be null to show that there is no cloud. Compared to the detection results of other different methods, the contrast results show that the efficiency of the detection method proposed in this paper is high and the accuracy satisfies the demand of real-time evaluation and the application range is wider.

[Automatic houses detection with color aerial images based on image segmentation].

In order to achieve housing automatic detection from high-resolution aerial imagery, the present paper utilized the color information and spectral characteristics of the roofing material, with the image segmentation theory, to study the housing automatic detection method. Firstly, This method proposed in this paper converts the RGB color space to HIS color space, uses the characteristics of each component of the HIS color space and the spectral characteristics of the roofing material for image segmentation to isolate red tiled roofs and gray cement roof areas, and gets the initial segmentation housing areas by using the marked watershed algorithm. Then, region growing is conducted in the hue component with the seed segment sample by calculating the average hue in the marked region. Finally through the elimination of small spots and rectangular fitting process to obtain a clear outline of the housing area. Compared with the traditional pixel-based region segmentation algorithm, the improved method proposed in this paper based on segment growing is in a one-dimensional color space to reduce the computation without human intervention, and can cater to the geometry information of the neighborhood pixels so that the speed and accuracy of the algorithm has been significantly improved. A case study was conducted to apply the method proposed in this paper to high resolution aerial images, and the experimental results demonstrate that this method has a high precision and rational robustness.