Jinsong Deng

Identification of Nitrogen, Phosphorus, and Potassium Deficiencies in Rice Based on Static Scanning Technology and Hierarchical Identification Method

Establishing an accurate, fast, and operable method for diagnosing crop nutrition is very important for crop nutrient management. In this study, static scanning technology was used to collect images of a rice samples fully expanded top three leaves and corresponding sheathes. From these images, 32 spectral and shape characteristic parameters were extracted using an RGB mean value function and using the Regionprops function in MATLAB. Hierarchical identification was used to identify NPK deficiencies. First, the normal samples and non-normal (NPK deficiencies) samples were identified. Then, N deficiency and PK deficiencies were identified. Finally, P deficiency and K deficiency were identified. In the identification of every hierarchy, SVFS was used to select the optimal characteristic set for different deficiencies in a targeted manner, and Fisher discriminant analysis was used to build the diagnosis model. In the first hierarchy, the selected characteristics were the leaf sheath R, leaf sheath G, leaf sheath B, leaf sheath length, leaf tip R, leaf tip G, leaf area and leaf G. In the second hierarchy, the selected characteristics were the leaf sheath G, leaf sheath B, white region of the leaf sheath, leaf B, and leaf G. In the third hierarchy the selected characteristics were the leaf G, leaf sheath length, leaf area/leaf length, leaf tip G, difference between the 2nd and 3rd leaf lengths, leaf sheath G, and leaf lightness. The results showed that the overall identification accuracies of NPK deficiencies were 86.15, 87.69, 90.00 and 89.23% for the four growth stages. Data from multiple years were used for validation, and the identification accuracies were 83.08, 83.08, 89.23 and 90.77%.

Quantifying Nitrogen Status of Rice Using Low Altitude UAV-Mounted System and Object-Oriented Segmentation Methodology

Nitrogen deficiency can seriously reduce yield, while over-fertilization can result problems such as excess nutrient runoff and groundwater pollution. Hence, efficient methods for assessing crop nitrogen status are needed to enable more optimal fertilizer management. The ability to quantify the different nitrogen application rates by crops using digital images taken from an unmanned aerial vehicle (UAV) was investigated in comparison with ground-based hyperspectral reflectance and chlorophyll content data from 140 plots on a managed field. This research utilized new UAV system, comprised of remote-controlled helicopter (Hercules II) and digital camera (EOS 30D), was used to characterize spatial and temporal variation in crop production. Digital information was extracted based on an object-oriented segmentation method, and the color parameter was reduced and represented using principal component analysis (PCA). An estimating model was established after analyzing the relationship between the optimal color parameter and ground-based measurements. Model testing demonstrated that unknown samples could be associated with the controlled nitrogen application rates (0, 60, 90, and 120 kg N·hm−2 ): 91.6% %; N1 (60 kg N·hm−2 ): 70.83%; N2 (90 kg N·hm−2 ): 86.7%; N3 (120 kg N·hm−2 ): 95%. Overall, this result proved to provide a cost-effective and accurate way and the UAV was an exploratory and predictive tool when applied to quantify different status of nitrogen. In addition, it indicated that the application of digital image from UAV to the problem of estimating different nitrogen rates is promising.Copyright

Comparison of machine-learning methods for above-ground biomass estimation based on Landsat imagery

Abstract. Biomass is one significant biophysical parameter of a forest ecosystem, and accurate biomass estimation on the regional scale provides important information for carbon-cycle investigation and sustainable forest management. In this study, Landsat satellite imagery data combined with field-based measurements were integrated through comparisons of five regression approaches [stepwise linear regression, K-nearest neighbor, support vector regression, random forest (RF), and stochastic gradient boosting] with two different candidate variable strategies to implement the optimal spatial above-ground biomass (AGB) estimation. The results suggested that RF algorithm exhibited the best performance by 10-fold cross-validation with respect to R2 (0.63) and root-mean-square error (26.44  ton/ha). Consequently, the map of estimated AGB was generated with a mean value of 89.34  ton/ha in northwestern Zhejiang Province, China, with a similar pattern to the distribution mode of local forest species. This research indicates that machine-learning approaches associated with Landsat imagery provide an economical way for biomass estimation. Moreover, ensemble methods using all candidate variables, especially for Landsat images, provide an alternative for regional biomass simulation.

Identification and Assessment of Potential Water Quality Impact Factors for Drinking-Water Reservoirs

Various reservoirs have been serving as the most important drinking water sources in Zhejiang Province, China, due to the uneven distribution of precipitation and severe river pollution. Unfortunately, rapid urbanization and industrialization have been continuously challenging the water quality of the drinking-water reservoirs. The identification and assessment of potential impacts is indispensable in water resource management and protection. This study investigates the drinking water reservoirs in Zhejiang Province to better understand the potential impact on water quality. Altogether seventy-three typical drinking reservoirs in Zhejiang Province encompassing various water storage levels were selected and evaluated. Using fifty-two reservoirs as training samples, the classification and regression tree (CART) method and sixteen comprehensive variables, including six sub-sets (land use, population, socio-economy, geographical features, inherent characteristics, and climate), were adopted to establish a decision-making model for identifying and assessing their potential impacts on drinking-water quality. The water quality class of the remaining twenty-one reservoirs was then predicted and tested based on the decision-making model, resulting in a water quality class attribution accuracy of 81.0%. Based on the decision rules and quantitative importance of the independent variables, industrial emissions was identified as the most important factor influencing the water quality of reservoirs; land use and human habitation also had a substantial impact on water quality. The results of this study provide insights into the factors impacting the water quality of reservoirs as well as basic information for protecting reservoir water resources.

Rural Settlement Subdivision by Using Landscape Metrics as Spatial Contextual Information

Multiple policy projects have changed land use and land cover (LULC) in China’s rural regions over the past years, resulting in two types of rural settlements: new-fashioned and old-fashioned. Precise extraction of and discrimination between these two settlement types are vital for sustainable land use development. It is difficult to identify these two types via remote sensing images due to their similarities in spectrum, texture, and geometry. This study attempts to discriminate different types of rural settlements by using a spatial contextual information extraction method based on Gaofen 2 (GF-2) images, which integrate hierarchical multi-scale segmentation and landscape analysis. A preliminary LULC map was derived by using only traditional spectral and geometrical features from a finer scale. Subsequently, a vertical connection was built between superobjects and subobjects, and landscape metrics were computed. The vertical connection was used for assigning landscape contextual information to subobjects. Finally, a classification phase was conducted, in which only multi-scale contextual information was adopted, to discriminate between new-fashioned and old-fashioned rural settlements. Compared with previous studies on multi-scale contextual information, this paper employs landscape metrics to quantify contextual characteristics, rather than traditional spectral, textural, and topological relationship information, from superobjects. Our findings indicate that this approach effectively identified and discriminated two types of rural settlements, with accuracies over 80% for both producers and users. A comparison with a conventional top-down hierarchical classification scheme showed that this novel approach improved accuracy, precision, and recall. Our results confirm that multi-scale contextual information with landscape metrics provides valuable spatial information for classification, and indicates the practicability, applicability, and effectiveness of this synthesized approach in distinguishing different types of rural settlements.

Analyzing the Impact of Highways Associated with Farmland Loss under Rapid Urbanization

Highway construction has accelerated urban growth and induced direct and indirect changes to land use. Although many studies have analyzed the relationship between highway construction and local development, relatively less attention has been paid to clarifying the various impacts of highways associated with farmland loss. This paper integrates GIS spatial analysis, remote sensing, buffer analysis and landscape metrics to analyze the landscape pattern change induced by direct and indirect highway impacts. This paper explores the interaction between the impact of highways and farmland loss, using the case of the highly urbanized traffic hubs in eastern China, Hang-Jia-Hu Plain. Our results demonstrate that the Hang-Jia-Hu Plain experienced extensive highway construction during 1990–2010, with a clear acceleration of expressway development since 2000. This unprecedented highway construction has directly fragmented the regional landscape and indirectly disturbed the regional landscape by attracting a large amount of built-up land transition from farmland during the last two decades. In the highway-effect zone, serious farmland loss initially occurred in the urban region and then spread to the rural region. Moreover, we found the discontinuous expansion of built-up land scattered the farmland in the rural region and expressway-effect zone. Furthermore, farmland protection policies in the 1990s had the effect of controlling the total area of farmland loss. However, the cohesive farmland structure was still fragmented by the direct and indirect impacts of highway construction. We suggest that an overall farmland protection system should be established to enhance spatial control and mitigate the adverse impacts caused by highway construction. This work improves the understanding of regional sustainable development, and provides a scientific basis for balanced urban development with farmland protection in decision-making processes.

Dynamics of Hierarchical Urban Green Space Patches and Implications for Management Policy

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.

Spatio-Temporal Trends and Identification of Correlated Variables with Water Quality for Drinking-Water Reservoirs.

It is widely accepted that characterizing the spatio-temporal trends of water quality parameters and identifying correlated variables with water quality are indispensable for the management and protection of water resources. In this study, cluster analysis was used to classify 56 typical drinking water reservoirs in Zhejiang Province into three groups representing different water quality levels, using data of four water quality parameters for the period 2006–2010. Then, the spatio-temporal trends in water quality were analyzed, assisted by geographic information systems (GIS) technology and statistical analysis. The results indicated that the water quality showed a trend of degradation from southwest to northeast, and the overall water quality level was exacerbated during the study period. Correlation analysis was used to evaluate the relationships between water quality parameters and ten independent variables grouped into four categories (land use, socio-economic factors, geographical features, and reservoir attributes). According to the correlation coefficients, land use and socio-economic indicators were identified as the most significant factors related to reservoir water quality. The results offer insights into the spatio-temporal variations of water quality parameters and factors impacting the water quality of drinking water reservoirs in Zhejiang Province, and they could assist managers in making effective strategies to better protect water resources.

Application of multi-sensor images for detecting land cover change and analysis of urban expansion

Zhejiang province is playing an increasingly vital role in Chinas overall economic growth. Concomitant with the dramatic economic development, this region has been undergoing tremendous urban growth on an unprecedented scale and rate. Many urbanization-related problems have been identified, including agricultural land and wetland loss, water pollution and soil erosion. There is an urgent need to detect and monitor the land cover change and analyze the magnitude and pattern, accurately and timely for planning and management. Remote sensing is a powerful tool for monitoring rapid change in the landscape resulting from urban development. However, change detection capabilities are intrinsically limited by the spatial resolution of the digital imagery in urban. The application of multi-sensor data provides the potential to more accurately detect land-cover changes through integration of different features of sensor data. Taking Hangzhou city as case study, this paper presents a method that combines principal component analysis (PCA) of multi-sensor data (SPOT-5 XS and ETM Pan data) and a hybrid classification involving unsupervised and supervised classier to detect and analysis land cover change. The study demonstrates that this method provides a very useful way in monitoring rapid land cover change in urban environment.