Shengcheng Cui

Robust Semi-supervised Kernel-FCM Algorithm Incorporating Local Spatial Information for Remote Sensing Image Classification

Fuzzy c-means (FCM) algorithm is a popular method in image segmentation and image classification. However, the traditional FCM algorithm cannot achieve satisfactory classification results because remote sensing image data are not subjected to Gaussian distribution, contain some types of noise, are nonlinear, and lack labeled data. This paper presents a robust semi-supervised kernel-FCM algorithm incorporating local spatial information (RSSKFCM_S) to solve the aforementioned problems. In the proposed algorithm, insensitivity to noise is enhanced by introducing contextual spatial information. The non-Euclidean structure and the problem in nonlinearity are resolved through kernel methods. Semi-supervised learning technique is utilized to supervise the iterative process to reduce step number and improve classification accuracy. Finally, the performance of the proposed RSSKFCM_S algorithm is tested and compared with several similar approaches. Experimental results for the multispectral remote sensing image show that the RSSKFCM_S algorithm is more effective and efficient.

Analyses of aerosol properties in Hefei based on polarization measurements of a sun photometer

Data collected with a CIMEL CE-318 sun-tracking photometer in Hefei between 2006 and 2008 were collected and inverted using an algorithm which takes polarization information into account. The aerosol optical depths (AOD) showed a pronounced temporal trend, with a maximum value of 0.55 at 440 nm in spring, and values around 0.45 in the remaining seasons. The Ångström parameter for all seasons exceeds 0.80 even in spring. Size distribution showed that both fine and coarse aerosol fractions change significantly and periodically in magnitude and shape during a year. Geometric mean radii for fine and coarse modes are 0.18 μm (standard deviation is 0.016) and 2.7 μm (standard deviation is 0.37), respectively. The spectral dependence of single scattering albedo and reflective index is distinct in the four seasons due to the different aerosol components. All these properties are reported, and it is expected that these aerosol characterizations will help refine aerosol models and clarify the mechanisms of aerosol radiative forcing.

Toward a new radiative-transfer-based model for remote sensing of terrestrial surface albedo

This Letter formulates a simple yet accurate radiative-transfer-based theoretical model to characterize the fraction of radiation reflected by terrestrial surfaces. Emphasis is placed on the concept of inhomogeneous distribution of the diffuse sky radiation function (DSRF) and multiple interaction effects (MIE). Neglecting DSRF and MIE produces a -1.55% mean relative bias in albedo estimates. The presented model can elucidate the impact of DSRF on the surface volume scattering and geometry-optical scattering components, respectively, especially for slant illuminations with solar zenith angles (SZA) larger than 50°. Particularly striking in the comparisons between our model and ground-based observations is the achievement of the agreement level, indicating that our model can effectively resolve the longstanding issue in accurately estimating albedo at extremely large SZAs and is promising for land-atmosphere interactions studies.

Adaptive contourlet-wavelet iterative shrinkage/thresholding for remote sensing image restoration

In this paper, we present an adaptive two-step contourlet-wavelet iterative shrinkage/thresholding (TcwIST) algorithm for remote sensing image restoration. This algorithm can be used to deal with various linear inverse problems (LIPs), including image deconvolution and reconstruction. This algorithm is a new version of the famous two-step iterative shrinkage/thresholding (TwIST) algorithm. First, we use the split Bregman Rudin-Osher-Fatemi (ROF) model, based on a sparse dictionary, to decompose the image into cartoon and texture parts, which are represented by wavelet and contourlet, respectively. Second, we use an adaptive method to estimate the regularization parameter and the shrinkage threshold. Finally, we use a linear search method to find a step length and a fast method to accelerate convergence. Results show that our method can achieve a signal-to-noise ratio improvement (ISNR) for image restoration and high convergence speed.

Assessment of Aerosol Modes Used in the MODIS Ocean Aerosol Retrieval

AbstractCoastal and island Aerosol Robotic Network (AERONET) sites are used to determine characteristic aerosol modes over marine environments. They are compared with the assumed modes used in the operational Moderate Resolution Imaging Spectroradiometer (MODIS) ocean aerosol algorithm, and the results show that 1) the standard deviation values of three fine aerosol modes (0.6) and one dustlike aerosol mode (0.8) are much higher than the corresponding statistical AERONET modal values (0.45 and 0.6, respectively). The values of three sea salt aerosol modes (0.6) are somewhat lower than the corresponding statistical AERONET modal value (0.675). 2) The number median radius of the current fine and dustlike aerosol modes cannot span the dynamic range of corresponding aerosol distribution properly. 3) AERONET products show that the standard deviation and the number median radius exhibit an obvious negative correlation, especially for sea salt and dustlike aerosol modes. According to this, a refinement of the cu...

A new method for improving the retrieved aerosol fine-mode fraction from MODIS over ocean

A new method for improving the retrieved aerosol fine-mode fraction (550) based on the current Moderate Resolution Imaging Spectroradiometer (MODIS) ocean algorithm is proposed. In the current MODIS ocean algorithm, the top of the atmosphere (TOA) apparent reflectance needs calculation from lookup tables (LUTs). The weighting parameters used in the calculation show an obvious spectral dependence, which is not taken into account in the current algorithm. The main measure taken in this study is to consider the spectral dependence of the weighting parameters. The MODIS aerosol products and the Aerosol Robotic Network (AERONET) data of Hong Kong Hok Tsui, Midway Island, Martha’s Vineyard Coastal Observatory (MVCO) and COVE, Virginia, where aerosols exhibit different loading and size distribution, are used to test the new method. The results show that the new method improves the retrieved fine-mode fraction, which is underestimated in anthropogenic-dominated aerosol conditions and overestimated in the sea salt-dominated aerosol conditions by the current algorithm. The correlation of the retrieved fine-mode fraction between the new method and AERONET is much higher (correlation coefficient, r = 0.92) than that between the current MODIS and AERONET (r = 0.80). The retrieved aerosol optical depth (AOD) is also improved. More AODs retrieved from the new method lie within the expected error bars.