The spectral reflectance data and the atmospheric parameters released by Radiometric Calibration Network (RadCalNet) have been widely used in many works and application fields. Unfortunately, due to many unexpected factors, the spectral reflectance data are missing, which will seriously influence the application level and usage domain of RadCalNet data. Therefore, how to compensate the missing RadCalNet data is extremely urgent to be solved. Taking Railroad Valley Playa site as the investigated site, a top-of-atmosphere (TOA) spectral reflectance model is proposed to simulate the missing RadCalNet data with high accuracy. The preliminary simulated TOA spectral reflectance is derived and validated. Compared to the real TOA spectral reflectance of RadCalNet data, it is found that the preliminary simulation accuracy is lower, of which the root-mean-square errors (RMSEs) are up to 3.31%. After analyzing the consistency of the time-series TOA reflectance measured by Landsat8 Operational Land Imager (OLI) and the corresponding equivalent TOA reflectance of RadCalNet data in five OLI bands, it is encouraging that the RMSEs of relative differences are less than 1.82%. Thus, the scale factors of these two TOA reflectances in five OLI bands are interpolated to obtain the corresponding scale factors of 61 bands in the spectral range from 400 to 1000 nm with 10-nm intervals to correct the preliminary simulated TOA spectral reflectance. The validation results show that the accuracies of the corrected simulation results are significantly improved, of which the RMSEs are all less than 1.85%, which can be used to effectively compensate the missing RadCalNet data. In addition, the influences of the input parameters, interpolation method, and geopositioning error of OLI on the simulation accuracy are sufficiently discussed to further reveal the probable effect factors of the proposed method.

The Baotou sandy site in China is an important field site in the Committee on Earth Observation Satellites radiometric calibration network (RadCalNet), which freely releases the bottom-of-atmosphere (BOA) and top-of-atmosphere (TOA) spectral reflectance products in the spectral range from 400 nm to 1000 nm at 10 nm intervals with the aid of a set of automatic spectral and atmospheric parameter measurement devices. However, the released TOA and BOA spectral reflectance products are nadir-view spectral reflectance, which could produce errors if the Baotou sandy site is offset from the nadir point when they are utilized to calibrate one satellite sensor with the vicarious calibration method or to detect radiometric performance changes in the target sensor. Therefore, to address this problem and further broaden the applicable scopes of TOA spectral reflectance products, this article develops TOA and BOA bidirectional reflectance distribution function (BRDF) models using valid time-series TOA and BOA spectral reflectance products to fulfil BRDF correction. Then, the corrected TOA spectral reflectance is validated with the on-orbit measured TOA spectral reflectance of the Landsat8 Operational Land Imager (OLI) sensor with high and stable radiometric performance. After analysing the fitting accuracies of the BOA and TOA BRDF models, the influences of the filter criteria of valid time-series nadir-view BOA and TOA spectral reflectance, geo-positioning error of Landsat8/OLI, and aerosol type in the BOA BRDF correction process on the validation results are discussed. The results show that the BOA BRDF model obtains a better correction effect and is more stable than the TOA BRDF model.

The wide field of view (WFV) imaging system of the GaoFen-6 (GF-6) satellite processes four popular spectral bands [blue, green, red, and near-infrared (NIR)] and four new spectral bands (costal, yellow, and two red edges). However, the corresponding reference spectral bands for these eight spectral bands from one reference sensor are insufficient in the cross-calibration process of the WFV; therefore, current cross-calibration methods should be improved. To address this problem, the Moderate-Resolution Imaging Spectroradiometer (MODIS), which has high radiometric performance with the aid of an onboard calibration system, is used as a reference sensor, and cross-calibration methods using the top of atmosphere (TOA) and the bottom of atmosphere (BOA) bidirectional reflectance distribution function (BRDF) models are developed and compared. The results reveal that the cross-calibration results in eight spectral bands with the BOA BRDF model that can obtain higher consistency with the official calibration coefficients (OCCs) compared with the TOA BRDF model. In addition, after various influencing factors are comprehensively analyzed, the spectral band adjustment factor (SBAF) correction in the shortwave infrared (SWIR) spectral bands can be neglected; moreover, by using five spectral bands (blue, green, red, NIR, and SWIR) of the MODIS sensor, the BOA BRDF model and the cubic polynomial interpolation method provide optimal cross-calibration schemes for WFV sensor. The total radiometric cross-calibration uncertainties of the proposed cross-calibration methods with the BOA BRDF model and the TOA BRDF model are less than 5.73% and 6.32%, respectively.

As an important boundary-based clustering algorithm, support vector clustering (SVC) can benefit many real applications owing to its capability of handling arbitrary cluster shapes, especially those directly or indirectly related to pattern exploration and description. As the application deepens, the importance of performance (i.e. criterions of accuracy and efficiency) of SVC increases. To identify gaps in the current methods and propose novel research directions for SVC, we present a survey of the literature in this area. Our approach is to classify the most recent advances into either theory or application. For theoretical contributions, advances related to parameter selection and optimization, dual-problem solutions, and cluster labeling are introduced. We also simultaneously summarize the advantages and drawbacks of each study. With respect to applications, we clearly describe eight groups of schemes based on SVC, either as individual or hybrid methods. Finally, we identify the gaps in SVC research and suggest several future research issues and trends.