Evaluating the performance of PC-ANN for the estimation of rice nitrogen concentration from canopy hyperspectral reflectance

In this study, a wide range of leaf nitrogen concentration levels was established in field-grown rice with the application of three fertilizer levels. Hyperspectral reflectance data of the rice canopy through rice whole growth stages were acquired over the 350 nm to 2500 nm range. Comparisons of prediction power of two statistical methods (linear regression technique (LR) and artificial neural network (ANN)), for rice N estimation (nitrogen concentration, mg nitrogen g^?1 leaf dry weight) were performed using two different input variables (nitrogen sensitive hyperspectral reflectance and principal component scores). The results indicted very good agreement between the observed and the predicted N with all model methods, which was especially true for the PC-ANN model (artificial neural network based on principal component scores), with an RMSE?=?0.347 and REP?=?13.14%. Compared to the LR algorithm, the ANN increased accuracy by lowering the RMSE by 17.6% and 25.8% for models based on spectral reflectance and PCs, respectively.     

Comparison between back propagation neural network and regression models for the estimation of pigment content in rice leaves and panicles using hyperspectral data

Two nitrogen experiments on rice were conducted in 2002, and the reflectances (350 to 2500 nm) and pigment contents (chlorophylls a and b, total chlorophylls and carotenoids) for leaf and panicle samples at different growth stages were measured in the laboratory. After performing an outlier analysis, the number of samples were 843 for leaves and 188 for panicles. Absorption features at 430, 460, 470, 640 and 660 nm for different pigments, and the relative reflectance of the green peak around 550 nm calculated by the continuum‐removed method, as well as the red edge position (REP) of rice leaves and panicles were selected as the independent variables, and measured pigment contents were selected as the dependent variables. Then, back propagation neural network (BPN) models, a kind of artificial neuron network (ANN), and multivariate linear regression models (MLR) were trained and tested. The main objective of this study was to compare the predictive ability of the ANN models to that of the MLR models in estimating the content of pigments in rice leaves and panicles. Results showed that all BPN models gave higher coefficients of determination (R²) and lower absolute errors (ABSEs) and root mean squared errors (RMSEs) than the corresponding MLR models, in both calibration and validation tests. Further significance tests by paired t tests and bootstrapping algorithms indicated that most of the BPN models outperformed the MLR models. When trained by combination data that did not meet the assumption of normal distribution, the BPN models appeared to not only have a better learning ability, but also had a more accurate predictive power than the MLR models. The estimation of leaf pigments was more accurate than that of panicle pigments, independent of which model was used.     

Estimation and distribution of near-surface meteorological elements over complex terrains: a case study in the Tibetan areas of West Sichuan Province,China

ABSTRACT

Meteorological elements are important for various fields related to human activities, including scientific research. Using the Tibetan Areas of West Sichuan Province (TAOWS) as an example, this study examined the estimation methods for near-surface air temperature (T_a), vapour pressure deficit (VPD), and atmospheric pressure (P) and their distribution characteristics in areas with complex terrains and sparse stations. An improved satellite-based approach, combining an artificial neural network and inverse distance weighting (ANN-IDW), is proposed for estimating T_a and VPD with high-accuracy under all weather conditions from Moderate Resolution Imaging Spectroradiometer (MODIS) data. The data of 41 meteorological stations in TAOWS and its adjacent areas were used for the training and validation of the ANN-IDW. For T_a and VPD, the mean absolute errors (MAEs) of the ANN-IDW are 1.45°C to 2.15°C and 0.54 hPa to 0.87 hPa, respectively. Also, the detailed features of the distribution of the estimated T_a and VPD are prominent and closely related to the terrain. The accuracy of the method was also verified indirectly. In addition, the improved method based on the existing method was applied for estimating P. The results confirm that (1) the ANN-IDW is suitable for estimating T_a and VPD in areas with complex terrain and sparse stations under all weather conditions; (2) the improved method is more suitable for estimating P at high-elevation. Moreover, the distribution characteristics of meteorological elements in TAOWS were also analysed. These elements influence agricultural production and animal husbandry and have a high application value. The results further show that topography is the most important factor affecting the spatial distribution and complexity of meteorological elements over complex terrains, but the degree of influence of topography varies greatly across different seasons.     

一种基于FNN的高速网络拥塞控制策略

以ATM（asynchronous transfer mode)为研究对旬,同种基于模糊神经网络（fuzzy neural network,简称FNN）的流量预测和拥塞控制策略,拥塞控制是高速网络（如ATM）研究中的关键问题之一,传统的基于BP神经网络的流量预测方法因其收敛速度较慢且具有较大的误差,影响了拥塞控制效果,而模糊神经网络由于具有处理不确定性问题和很强的学习能力,很好地解决这一问题,最后通过仿真,比较和分析了基于BP神经网络和基于FNN方法和性能,证明此方法是有效的。     

An optimized instance based learning algorithm for estimation of compressive strength of concrete

This article proposes an optimized instance-based learning approach for prediction of the compressive strength of high performance concrete based on mix data, such as water to binder ratio, water content, super-plasticizer content, fly ash content, etc. The base algorithm used in this study is the k nearest neighbor algorithm, which is an instance-based machine leaning algorithm. Five different models were developed and analyzed to investigate the effects of the number of neighbors, the distance function and the attribute weights on the performance of the models. For each model a modified version of the differential evolution algorithm was used to find the optimal model parameters. Moreover, two different models based on generalized regression neural network and stepwise regressions were also developed. The performances of the models were evaluated using a set of high strength concrete mix data. The results of this study indicate that the optimized models outperform those derived from the standard k nearest neighbor algorithm, and that the proposed models have a better performance in comparison to generalized regression neural network, stepwise regression and modular neural networks models.     

Prediction of the strength of mineral admixture concrete using multivariable regression analysis and an artificial neural network

This study applies multiple regression analysis and an artificial neural network in estimating the compressive strength of concrete that contains various amounts of blast furnace slag and fly ash, based on the properties of the additives (blast furnace slag and fly ash in this case) and values obtained by non-destructive testing rebound number and ultrasonic pulse velocity for 28 different concrete mixtures (M_control and M₁–M₂₇) at different curing times (3, 7, 28, 90, and 180 days). The results obtained using the two methods are then compared and discussed. The results reveal that although multiple regression analysis was more accurate than artificial neural network in predicting the compressive strength using values obtained from non-destructive testing, the artificial neural network models performed better than did multiple regression analysis models. The application of an artificial neural network to the prediction of the compressive strength in admixture concrete of various curing times shows great potential in terms of inverse problems, and it is suitable for calculating nonlinear functional relationships, for which classical methods cannot be applied.     

Comparison of two inversion methods for leaf area index using HJ-1 satellite data in a temperate meadow steppe

Leaf area index (LAI) is one of the most important parameters for determining grassland canopy conditions. LAI controls numerous biological and physical processes in grassland ecosystems. Remote-sensing techniques are effective for estimating grassland LAI at a regional scale. Comparison of LAI inversion methods based on remote sensing is significant for accurate estimation of LAI in particular areas. In this study, we developed and compared two inversion models to estimate the LAI of a temperate meadow steppe in Hulunbuir, Inner Mongolia, China, based on HJ-1 satellite data and field-measured LAI data. LAI was measured from early June to late August in 2013, obtained from 326 sampling data. The back propagation (BP) neural network method proved better than the statistical regression model for estimating grassland LAI, the accuracy of the former being 82.8%. We then explored the spatio-temporal distribution in LAI of Stipa baicalensis Roshev. in the meadow steppe of Hulunbuir, including cut, grazed, and fenced plots. The LAI in the cut and grazed plots reflected the growth variations in S. baicalensis Roshev. However, because of the obvious litter layer, the LAI in the fenced plots was underestimated.     

Enhancing estimation accuracy of daily maximum,minimum, and mean air temperature using spatio-temporal ground-based and remote-sensing data in southern Iran

ABSTRACT

In this research, a model was proposed for improving the estimation of air temperature (T_a), which enhanced computation accuracy by combining remote sensing, station data, and spatio-temporal interpolation methods. Stepwise linear regression model was used to find the relationship between daily mean, maximum, and minimum air temperature (T_mean, T_max, and T_min, respectively) with daytime and night-time land surface temperature (LST), normalized difference vegetation index (NDVI) from the Moderate Resolution Imaging Spectroradiometer sensor, elevation, geometric temperature (T_geom, as a function of the day of the year and latitude), and solar radiation (R_s). Measured T_a from 75 stations in the Fars province in southern Iran were used during 2003–2012. The results of stepwise linear regression showed that among the considered variables daytime and night-time LST, NDVI, T_geom, R_s, and elevation were significant in the final model. For increasing the accuracy of estimation, four interpolation methods were considered and analysed for the residual errors of multiple linear regression model consisting regression kriging, spatio-temporal regression kriging (STRK), regression inverse distance weighting, and spatio-temporal regression inverse distance weighting (STRIDW). The result showed that the STRIDW method had the best accuracy among the considered methods and a significant improvement in the accuracy was achieved with this method comparing to the others. The accuracy of estimations was less than 2°C for T_max, T_min, and T_mean (root mean square error = 1.6°C, 1.84°C, and 1.43°C, respectively) for the validation year (2012). Finally, using the proposed models, it was possible to estimate daily air temperatures in the Fars province with 1 km resolution, which is higher than methods that used purely station-based or purely remote-sensing data.     

Estimating spatial variations in soil organic carbon using satellite hyperspectral data and map algebra

This study evaluated the effectiveness of using Hyperion hyperspectral data in improving existing remote-sensing methodologies for estimating soil organic carbon (SOC) content on farmland. The study area is Big Creek Watershed in Southern Illinois, USA. Several data-mining techniques were tested to calibrate and validate models that could be used for predicting SOC content using Hyperion bands as predictors. A combined model of stepwise regression followed by a five hidden nodes artificial neural network was selected as the best model, with a calibration coefficient of determination (R ²) of 78.9% and a root mean square error (RMSE) of 3.3 tonnes per hectare (t ha^?1). The validation RMSE, however, was found to be 11.3 t ha^?1. Map algebra was implemented to extrapolate this model and produce a SOC map for the watershed. Hyperspectral data improved marginally the predictability of SOC compared to multispectral data under natural field conditions. They could not capture small annual variations in SOC, but could measure decadal variations with moderate error. Satellite-based hyperspectral data combined with map algebra can measure total SOC pools in various ecosystem or soil types to within a few per cent error.     

Cascaded neural network based small array synthesis with robustness to noise

A cascaded neural network approach has been presented in this paper to estimate the excitation for the desired field distribution using a radial basis function neural network (RBFNN). The article has employed an electromagnetic design example consisting of 5 × 5 and 6 × 6 planar antenna array of isotropic sources with inter element‐distance of 0.5λ to show the adaptation of the neural network model in estimating the desired output. A neural network is trained using a dataset of suitable excitation voltages and its corresponding radiation patterns, which proves to be efficient in predicting the excitation voltages required to generate the desired pattern. A set of techniques based on a cascaded neural network is adopted for pattern synthesis using magnitude and phase, magnitude only, and template‐based input data. The robustness of the method has also been tested by considering noise with different SNR levels. The results found in each case have a close fit with the desired pattern.     

Exploring internal mechanism of warrant in financial market with a hybrid approach

In this research, we explore the internal mechanism of warrant in financial market with a hybrid approach integrating Black–Scholes pricing method and Grey theory into a genetic algorithm (GA) based back-propagation neural network (BPN). Black–Scholes pricing method can help make earnings with little risk. Grey theory can decrease the random and implicative noise of tempestuously undulant warrant prices. GA is used to find the best architecture for BPN to avoid local optimum.In experiment, we find that most of selected input variables for BPN include Black–Scholes pricing values and Grey index values. It shows that those two kinds of values are crucial factors. And the earnings rate of warrant outperforms that of the underlying asset. In addition, the proposed model is verified to outperform traditional BPN. However, the high risk of warrant is another subject to which we should pay attention.     

Hybrid high order neural networks

Neural networks (NNs) represent a familiar artificial intelligence approach widely applied in many fields and to a wide range of issues. The back propagation network (BPN) is one of the most well-known NNs, comprising multilayer perceptrons (MLPs) with an error back propagation learning algorithm. BPN typically employs associate multiplicative weightings for layer connections. For single connections, BPN combines neuron inputs linearly to neuron outputs. In this study, the author develops and embeds high order connections (exponent multipliers) into the BPN. The resultant proposed hybrid high order neural network (HHONN) is intended to be applicable to both linear and high order connections. HHONN allows an additional connection type for BPN, which permits BPN to adapt to different scenarios. In this paper, learning equations for both weighting and high order connections are introduced in their general forms. A feedforward neural network with a topology of two hidden layers and one high order connection was developed and studied to confirm the improved performance of developed HHONN models. Case studies, including two basic tests (a function approximation and the TC problem) and squat wall strength learning, were used to verify HHONN performance. Results showed that, when the high order connection was employed anywhere except the eventual connection, HHONN delivered better results than achievable using traditional BPN. Such results show that HHONN successfully introduces high order connections into BPN.     

A Neural Syntactic Language Model

This paper presents a study of using neural probabilistic models in a syntactic based language model. The neural probabilistic model makes use of a distributed representation of the items in the conditioning history, and is powerful in capturing long dependencies. Employing neural network based models in the syntactic based language model enables it to use efficiently the large amount of information available in a syntactic parse in estimating the next word in a string. Several scenarios of integrating neural networks in the syntactic based language model are presented, accompanied by the derivation of the training procedures involved. Experiments on the UPenn Treebank and the Wall Street Journal corpus show significant improvements in perplexity and word error rate over the baseline SLM. Furthermore, comparisons with the standard and neural net based N-gram models with arbitrarily long contexts show that the syntactic information is in fact very helpful in estimating the word string probability. Overall, our neural syntactic based model achieves the best published results in perplexity and WER for the given data sets.This work was supported by the National Science Foundation under grant No. IIS-0085940.Editors: Dan Roth and Pascale Fung     

Dealing with missing values in neural network-based diagnostic systems

Backpropagation neural networks have been applied to prediction and classification problems in many real world situations. However, a drawback of this type of neural network is that it requires a full set of input data, and real world data is seldom complete. We have investigated two ways of dealing with incomplete data — network reduction using multiple neural network classifiers, and value substitution using estimated values from predictor networks — and compared their performance with an induction method. On a thyroid disease database collected in a clinical situation, we found that the network reduction method was superior. We conclude that network reduction can be a useful method for dealing with missing values in diagnostic systems based on backpropagation neural networks.     

Static, Dynamic, and Hybrid Neural Networks in Forecasting Inflation

The back-propagation neural network (BPN) model has been the most popular form of artificial neural network model used for forecasting, particularly in economics and finance. It is a static (feed-forward) model which has a learning process in both hidden and output layers. In this paper we compare the performance of the BPN model with that of two other neural network models, viz., the radial basis function network (RBFN) model and the recurrent neural network (RNN) model, in the context of forecasting inflation. The RBFN model is a hybrid model with a learning process that is much faster than the BPN model and that is able to generate almost the same results as the BPN model. The RNN model is a dynamic model which allows feedback from other layers to the input layer, enabling it to capture the dynamic behavior of the series. The results of the ANN models are also compared with those of the econometric time series models.     

Comparing surface- and canopy-layer urban heat islands over Beijing using MODIS data

The canopy-layer urban heat island (CLHI) and the surface-layer urban heat island (SLHI) of Beijing, the capital city of China, were compared on the spatial scale of a city and the temporal scale of a year in this study. A differential temperature vegetation index (DTVX) method was improved by suggesting a new parameterization scheme for estimating daytime air temperature (T_a); a binary linear regression equation was developed for estimating night-time T_a from Moderate Resolution Imaging Spectroradiometer (MODIS) land-surface temperature (T_s) and vegetation indices data during 2009–2010. Validations using weather station observations show that the spatially distributed T_a can be obtained with an accuracy of approximately 2 K. Comparisons between the CLHI and the SLHI indicate that the CLHI agrees well with the SLHI during night-time, but they have a greater difference during daytime either in heat island intensity or in spatial distribution pattern. The SLHI?CLHI intensity difference during daytime has a noticeable seasonal variation, which is small and negative in cold seasons, but large and positive in warm seasons, whereas that at night-time has no significant seasonal variations. The difference in the evapotranspiration cooling effects between urban and rural areas may be the predominant factor that drives the SLHI?CLHI difference.     

Performance comparison of artificial neural network models for daily rainfall prediction

With an aim to predict rainfall one-day in advance, this paper adopted different neural network models such as feed forward back propagation neural network (BPN), cascade-forward back propagation neural network (CBPN), distributed time delay neural network (DTDNN) and nonlinear autoregressive exogenous network (NARX), and compared their forecasting capabilities. The study deals with two data sets, one containing daily rainfall, temperature and humidity data of Nilgiris and the other containing only daily rainfall data from 14 rain gauge stations located in and around Coonoor (a taluk of Nilgiris). Based on the performance analysis, NARX network outperformed all the other networks. Though there is no major difference in the performances of BPN, CBPN and DTDNN, yet BPN performed considerably well confirming its prediction capabilities. Levenberg Marquardt proved to be the most effective weight updating technique when compared to different gradient descent approaches. Sensitivity analysis was instrumental in identifying the key predictors.     

Integrating spectral indices with environmental parameters for estimating heavy metal concentrations in rice using a dynamic fuzzy neural-network model

A generalized dynamic fuzzy neural network (GDFNN) was created to estimate heavy metal concentrations in rice by integrating spectral indices and environmental parameters. Hyperspectral data, environmental parameters, and heavy metal content were collected from field experiments with different levels of heavy metal pollution (Cu and Cd). Input variables used in the GDFNN model were derived from 10 variables acquired by gray relational analysis. The assessment models for Cd and Cu concentration employed five and six input variables, respectively. The results showed that the GDFNN for estimating Cu and Cd concentrations in rice performed well at prediction with a compact network structure using the training, validation, and testing sets (for Cu, fuzzy rules=9, R² greater than 0.75, and RMSE less than 2.5; for Cd, fuzzy rules=9, R² greater than 0.75, and RMSE less than 1.0). The final GDFNN model was then compared with a back-propagation (BP) neural network model, adaptive-network-based fuzzy interference systems (ANFIS), and a regression model. The accuracies of GDFNN model prediction were usually slightly better than those of the other three models. This demonstrates that the GDFNN model is more suitable for predicting heavy metal concentrations in rice.     

计算机辅助材料设计的偏最小二乘法－人工神经网络研究

在噪声小的ＰＬＳ（偏最小二乘法）空间上，样本集的局部投影可被用作ＢＰＮ（反向传播网络）的输入元素以建立一种“平衡”的神经网络结构，这种结构在很大程度上克服了通常ＢＰＮ过拟合的缺点。在ＰＬＳ子空间优化区，利用非线性逆映照技术设计的基于期望目标值的样本可通过ＰＬＳ－ＰＮ方法预报和选取。本文还利用此方法设计了若干以初始容量为目标的Ｎｉ／ＭＨ电池阴极材料。