Search

The dynamic ensemble selection of classifiers is an effective approach for processing label-imbalanced data classifications. However, such a technique is prone to overfitting, owing to the lack of regularization methods and the dependence on local geometry of data. In this study, focusing on binary imbalanced data classification, a novel dynamic ensemble method, namely adaptive ensemble of classifiers with regularization (AER), is proposed, to overcome the stated limitations. The method solves the overfitting problem through a new perspective of implicit regularization. Specifically, it leverages the properties of stochastic gradient descent to obtain the solution with the minimum norm, thereby achieving regularization; furthermore, it interpolates the ensemble weights by exploiting the global geometry of data to further prevent overfitting. According to our theoretical proofs, the seemingly complicated AER paradigm, in addition to its regularization capabilities, can actually reduce the asymptotic time and memory complexities of several other algorithms. We evaluate the proposed AER method on seven benchmark imbalanced datasets from the UCI machine learning repository and one artificially generated GMM-based dataset with five variations. The results show that the proposed algorithm outperforms the major existing algorithms based on multiple metrics in most cases, and two hypothesis tests (McNemar’s and Wilcoxon tests) verify the statistical significance further. In addition, the proposed method has other preferred properties such as special advantages in dealing with highly imbalanced data, and it pioneers the researches on regularization for dynamic ensemble methods.

Due to the growing demand for land resources, coastal reclamation has become a global land use practice. Although the reclamation of coastal wetlands brings substantial economic benefits, it can lead to a series of adverse ecological impacts. Considerably, economic and ecological trade-offs of reclamation activities must be analyzed to make targeted land use policy decisions. This study focuses on land reclamation on the east coast of China, taking Hangzhou Bay as a case study. Remote sensing techniques were used to detect the coastline and landscape changes in the reclaimed areas from 1985 to 2015. Subsequently, field surveys and empirical parameters were employed to evaluate the impact on coastal ecosystem services. Finally, the monetary analysis was used to evaluate the trade-offs between ecological loss and economic gain due to land reclamation. The results showed that a total of 75134.3 ha of coastal wetlands (accounting for 8.58% of the total land area) were reclaimed from 1985 to 2015 in the Hangzhou Bay. A substantial decline in ecosystem service value (ESV) was observed with a drop from 866,400 Yuan ha−1 in 1985 to 285,800 Yuan ha−1 in 2015 due to the conversion of natural to artificial landscape. The total ESV loss exceeded the economic benefits throughout the study period, thus the economic benefits failed to compensate for the ESV loss. This trade-off relationship resulted from reclamation expansion on short-term economic benefits at the expense of long-term ecological value. According to these findings, we propose three main ways to achieve a “win–win” situation between gains and loss of reclamation for sustainability.

Arable land in China is undergoing significant changes, with massive losses of arable land due to rapid urbanization and the reclamation of arable land from other lands to compensate for these losses. Many studies have analyzed arable land loss, but less attention has been paid to land reclamation, and the utilization of reclaimed land remains unclear. The goal of our study was to characterize the patterns and efficiency of the utilization of reclaimed land and to identify the factors influencing the land utilization process in Wenzhou using remote sensing, geographic information systems and logistic regression. Our results showed that only 37% of the total reclaimed land area was under cultivation, and other lands were still bare or had been covered by trees and grasses. The likelihood that reclaimed land was used for cultivation was highly correlated with the land use type of its neighboring or adjacent parcels. Reclaimed land utilization was also limited at high elevations in lands with poor soil fertility and in lands at a great distance from rural residential areas. In addition, parcels located in the ecological protection zone were less likely to be cultivated. Therefore, we suggest that the important determinants should be considered when identifying the most suitable land reclamation areas.

Accurately quantifying the variation of urban green space is the prerequisite for fully understanding its ecosystem services. However, knowledge about the spatiotemporal dynamics of urban green space is still insufficient due to multiple challenges that remain in mapping green spaces within heterogeneous urban environments. This paper uses the city of Hangzhou to demonstrate an analysis methodology that integrates sub-pixel mapping technology and landscape analysis to fully investigate the spatiotemporal pattern and variation of hierarchical urban green space patches. Firstly, multiple endmember spectral mixture analysis was applied to time series Landsat data to derive green space coverage at the sub-pixel level. Landscape metric analysis was then employed to characterize the variation pattern of urban green space patches. Results indicate that Hangzhou has experienced a significant loss of urban greenness, producing a more fragmented and isolated vegetation landscape. Additionally, a remarkable amelioration of urban greenness occurred in the city core from 2002 to 2013, characterized by the significant increase of small-sized green space patches. The green space network has been formed as a consequence of new urban greening strategies in Hangzhou. These strategies have greatly fragmented the built-up areas and enriched the diversity of the urban landscape. Gradient analysis further revealed a distinct pattern of urban green space landscape variation in the process of urbanization. By integrating both sub-pixel mapping technology and landscape analysis, our approach revealed the subtle variation of urban green space patches which are otherwise easy to overlook. Findings from this study will help us to refine our understanding of the evolution of heterogeneous urban environments.

In this study, three-phase satellite images were used to define rules for the allocation of time and space in construction land resources based on a complex adaptive system and game theory. The decision behavior and rules of government agent, enterprise agent and resident agent in construction land growth were explored. A distinctive and dynamic simulation model of construction land growth was built, which integrated multi-agent, GIS technology and RS data and described the interaction among influencing agents. Taking Fuyang City in the Changjiang River Delta as an example, an assessment process for the remote sensing data in construction land and scenario planning was constructed. Repast and ArcGIS were used as simulation platforms. A simulation of the spatial pattern in land-use planning and the setting of scenario planning were conducted by using the incomplete active game, which was based on different natural, social and economic levels. Through this model, a simulation of urban planning space and decision-making for Fuyang City was created. Relevant non-structured problems arising from urban planning management could be identified, and the process and logic of urban planning spatial decision-making could thus be improved. Cel1-by-cel1 comparison showed that the simulation accuracy was over 72%. This model has great potential for use by government and town planners in decision support and technique support in the policy-making process.

Administering an urban boundary (UB) is increasingly important for curbing disorderly urban land expansion. The traditionally manual digitalization is time-consuming, and it is difficult to connect UB in the urban fringe due to the fragmented urban pattern in daytime data. Nighttime light (NTL) data is a powerful tool used to map the urban extent, but both the blooming effect and the coarse spatial resolution make the urban product unable to meet the requirements of high-precision urban study. In this study, precise UB is extracted by a practical and effective method using NTL data and Landsat 8 data. Hangzhou, a megacity experiencing rapid urban sprawl, was selected to test the proposed method. Firstly, the rough UB was identified by the search mode of the concentric zones model (CZM) and the variance-based approach. Secondly, a buffer area was constructed to encompass the precise UB that is near the rough UB within a certain distance. Finally, the edge detection method was adopted to obtain the precise UB with a spatial resolution of 30 m. The experimental results show that a good performance was achieved and that it solved the largest disadvantage of the NTL data-blooming effect. The findings indicated that cities with a similar level of socio-economic status can be processed together when applied to larger-scale applications.

There is an increasing need to understand pattern and growth of impervious surfaces in rural regions. However, studies using remote sensing of impervious surfaces have often focused on mapping impervious surfaces in urban regions with less emphasis placed on the rural impervious surfaces. In this paper, we proposed a new index, Rural Impervious Surface Index (RISI) by taking advantage of narrow spectral bands of Landsat 8 OLI for estimating impervious surfaces within rural land covers. This index is based on the combination of Normalized Difference Built-up Index (NDBI), Soil Adjusted Vegetation Index (SAVI) and Soil Index (SI). Respectively, these represent the three major rural land covers components: impervious surfaces, vegetation, and soil. The index was further used for estimating fraction of impervious surfaces using fuzzy KNN classifier. The performance of this technique was also compared with Linear Spectral Mixture Analysis (LSMA). Our results showed that RISI could accurately detect spatial pattern of rural impervious surfaces due to the suppressing background noise and minimizing spectral confusion. Accuracy assessment revealed that incorporation of RISI with fuzzy KNN classification generates higher correlation coefficient, lower root mean square and systematic error compared to the LSMA technique.

Although China has promoted the construction of Chinese Sustainable Ground Transportation (CSGT) to guide sustainable development, it may create substantial challenges, such as rapid urban growth and land limitations. This research assessed the effects of the Hangzhou Bay Bridge on impervious surface growth in Cixi County, Ningbo, Zhejiang Province, China. Changes in impervious surfaces were mapped based on Landsat images from 1995, 2002, and 2009 using a combination of multiple endmember spectral mixture analysis (MESMA) and landscape metrics. The results indicated that the area and density of impervious surfaces increased significantly during construction of the Hangzhou Bay Bridge (2002–2009). Additionally, the bridge and connected road networks promoted urban development along major roads, resulting in compact growth patterns of impervious surfaces in urbanized regions. Moreover, the Hangzhou Bay Bridge promoted the expansion and densification of impervious surfaces in Hangzhou Bay District, which surrounds the bridge. The bridge also accelerated socioeconomic growth in the area, promoting rapid urban growth in Cixi County between 2002 and 2009. Overall, the Hangzhou Bay Bridge is an important driver of urban growth in Cixi County, and policy suggestions for sustainable urban growth should be adopted in the future.