Diabetic retinopathy (DR) is a primary cause of blindness in which damage occurs to the retina due to an accretion of sugar levels in the blood. Therefore, prior detection, classification, and diagnosis of DR can prevent vision loss in diabetic patients. We proposed a novel and hybrid approach for prior DR detection and classification. We combined distinctive models to make the DR detection process robust or less error-prone while determining the classification based on the majority voting method. The proposed work follows preprocessing feature extraction and classification steps. The preprocessing step enhances abnormality presence as well as segmentation; the extraction step acquires merely relevant features; and the classification step uses classifiers such as support vector machine (SVM), K-nearest neighbor (KNN), and binary trees (BT). To accomplish this work, multiple severities of disease grading databases were used and achieved an accuracy of 98.06%, sensitivity of 83.67%, and 100% specificity.

In this paper, we present surface urban heat island (SUHI) analysis of Shanghai (China) based on the change in land use and land cover using satellite Landsat images from 2002 to 2013. With the rapid development of urbanization, urban ecological and environmental issues have aroused widespread concern. The urban heat island (UHI) effect is a crucial problem, as its generation and evolution are closely related to social and economic activities. Land-use and land-cover change (LUCC) is the key in analyzing the UHI effect. Shanghai, one of China’s major economic, financial and commercial centers, has experienced high development density for several decades. A tremendous amount of farmland and vegetation coverage has been replaced by an urban impervious surface, leading to an intensive SUHI effect, especially in the city’s center. Luckily, the SUHI trend has slowed due to reasonable urban planning and relevant green policies since the 2010 Expo. Data analyses demonstrate that an impervious surface (IS) has a positive correlation with land surface temperature (LST) but a negative correlation with vegetation and water. Among the three factors, impervious surface is the most relevant. Therefore, the policy implications of land use and control of impervious surfaces should pay attention to the relief of the current SUHI effect in Shanghai.

In this study, we present the evaluation of urban land carrying capacity (ULCC) based on an ecological sensitivity analysis. Remote sensing data and geographic information system (GIS) technology are employed to analyze topographic conditions, land-use types, the intensity of urban development, and ecological environmental sensitivity to create reasonable evaluation indicators to analyze urban land carrying capacity based on ecological sensitivity in the rapidly developing megacity of Hangzhou, China. In the study, ecological sensitivity is grouped into four levels: non-sensitive, lightly sensitive, moderately sensitive, and highly sensitive. The results show that the ecological sensitivity increases progressively from the center to the periphery. The results also show that ULCC is determined by ecologically sensitive levels and that the ULCC is categorized into four levels. Even though it is limited by the four levels, the ULCC still has a large margin if compared with the current population numbers. The study suggests that the urban ecological environment will continue to sustain the current population size in the short-term future. However, it is necessary to focus on the protection of distinctive natural landscapes so that decision makers can adjust measures for ecological conditions to carry out the sustainable development of populations, natural resources, and the environment in megacities like Hangzhou, China.

Land use/land cover (LULC) classification using optical and synthetic aperture radar (SAR) remote sensing images is becoming increasingly important to produce more accurate LULC products. As an important step, feature normalization techniques have been studied by the areas of pattern recognition. Nevertheless, because of the totally different imaging mechanisms of optical and SAR sensors, most of the existing normalization approaches are not suitable for optical and SAR data fusion. Moreover, whether normalization is a significant step remains unclear regarding optical and SAR fusion. Taking the Satellite Pour l'Observation de la Terre (SPOT-5) and the Advanced Land Observing Satellite (ALOS)/Phased Array type L-band SAR (PALSAR) (HH and HV polarizations) as the optical and SAR data, this letter aims to evaluate the impact of feature normalization. Experimental results indicated that feature normalization is not necessarily significant depending on fusion methods. For instance, distribution-dependent classifiers (e.g., a maximum likelihood classifier) are independent of feature normalization; thus, it has no impact on the results when using these classifiers. Moreover, advanced classifiers (e.g., a support vector machine) with built-in normalization are also not influenced by feature normalization. In contrast, a minimum distance classifier and an artificial neural network (ANN) depend on the input values of optical and SAR features and thus can be influenced by feature normalization. However, our experiments showed a fluctuation in classification accuracy using an ANN with normalized features. Therefore, more experiments are required to investigate the optimal normalization approaches for the optical and SAR images when using an ANN as the fusion method.

Polarimetric synthetic aperture radar (SAR) remote sensing provides an outstanding tool in oil spill detection and classification, for its advantages in distinguishing mineral oil and biogenic lookalikes. Various features can be extracted from polarimetric SAR data. The large number and correlated nature of polarimetric SAR features make the selection and optimization of these features impact on the performance of oil spill classification algorithms. In this paper, deep learning algorithms such as the stacked autoencoder (SAE) and deep belief network (DBN) are applied to optimize the polarimetric feature sets and reduce the feature dimension through layer-wise unsupervised pre-training. An experiment was conducted on RADARSAT-2 quad-polarimetric SAR image acquired during the Norwegian oil-on-water exercise of 2011, in which verified mineral, emulsions, and biogenic slicks were analyzed. The results show that oil spill classification achieved by deep networks outperformed both support vector machine (SVM) and traditional artificial neural networks (ANN) with similar parameter settings, especially when the number of training data samples is limited.

In this study, we assess the urban heat island (UHI) effect using remote sensing data, a phenomenon emerging under the background of global warming and urbanization. With the rapid development of satellite technology, remote sensing images are widely applied to evaluate the UHI effect on rapidly-urbanized areas in recent years. In the study, we applied Landsat 8 data to estimate the land surface temperature (LST) in the case study of Shenzhen and Hong Kong. The methods of the mono-window algorithm (MWA), single-channel method (SCM), Qin’s split-window algorithm (SWA-Q) and Sobrino’s split-window algorithm (SWA-S) are used to calculate the LST from Landsat 8 data on 29 November 2013, 16 November 2014, 18 October 2015, and 7 February 2016. The results show that LST retrievals by SWA-Q are better than those of the other algorithms in the case study of Shenzhen and Hong Kong. From 2013 to 2016, the high-LST zones or UHIs in Shenzhen and Hong Kong are substantially identical. Although the LST is not obviously correlated with vegetation distribution, the growth condition of vegetation may impact the distribution of the UHI, and the high LST is slightly correlated to the high atmospheric particulate concentration. Additionally, in general, Shenzhen and Hong Kong are weak UHI regions and the UHI-affected area in Shenzhen is larger than that in Hong Kong from 2013 to 2016.