Fuzzy Neural Network Models for Supervised Classification: Multispectral Image Analysis     

Arun D. Kulkarni  Kamlesh Lulla 《国际地球制图》2013,28(4):42-51

Abstract

It has been well established that neural networks provide a reasonable and powerful alternative to conventional classifiers. During the past few years there has been a large and energetic upswing in research efforts aimed at synthesizing fuzzy logic with neural networks. This combination of fuzzy logic and neural networks seems natural because two approaches generally attack the design of “intelligent” systems from quite different angles. Neural networks provide algorithms for learning, classification, and optimization whereas fuzzy logic deals with issues such as reasoning on a higher (semantic or linguistic) level. Consequently the two technologies complement each other. In this paper we propose two novel fuzzy‐neural network models for supervised learning. The first model consists of three layers, and the second model consists of four layers. In both models, the first two layers implement fuzzy membership functions and the remaining layers implement the inference engine. Both models use the gradient decent technique for learning. As an illustration, we have analyzed two Thematic mapper images using these models. Results are presented in the paper.  相似文献   

Observed and Forecasted Intraseasonal Activity of Southwest Monsoon Rainfall over India During 2010, 2011 and 2012     

D. R. Pattanaik  L. S. Rathore  Arun Kumar 《Pure and Applied Geophysics》2013,170(12):2305-2328

The monsoon seasons of 2010 and 2011, with almost identical seasonal total rainfall over India from June to September, are associated with slightly different patterns of intraseasonal rainfall fluctuations. Similarly, the year 2012, with relatively less rainfall compared to 2010 and 2011, also witnessed different intraseasonal rainfall fluctuations, leading to drought-like situations over some parts of the country. The present article discusses the forecasting aspect of monsoon activity over India during these 3 years on an extended range time scale (up to 3 weeks) by using the multimodel ensemble (MME), based on operational coupled model outputs from the ECMWF monthly forecasting system and the NCEP’s Climate Forecast System (CFS). The average correlation coefficient (CC) of weekly observed all-India rainfall (AIR) and the corresponding MME forecast AIR is found to be significant, above the 98 % level up to 2 weeks (up to 18 days) with a slight positive CC for the week 3 (days 19–25) forecast. However, like the variation of observed intraseasonal rainfall fluctuations during 2010, 2011 and 2012 monsoon seasons, the MME forecast skills of weekly AIR are also found to be different from one another, with the 2012 monsoon season indicating significant CC (above 99 % level) up to week 2 (12–18 days), and also a comparatively higher CC (0.45) during the week 3 forecast (days 19–25). The average CC between observed and forecasted weekly AIR rainfall over four homogeneous regions of India is found to be the lowest over the southern peninsula of India (SPI), and northeast India (NEI) is found to be significant only for the week 1 (days 5–11) forecast. However, the CC is found to be significant over northwest India (NWI) and central India (CEI), at least above the 90 % level up to 18 days, with NWI having slightly better skill compared to the CEI. For the individual monsoon seasons of 2010, 2011 and 2012, there is some variation in CC and other skill scores over the four homogeneous regions. Thus the slight variations in the characteristics of intraseasonal monsoon rainfall over India is associated with variations in predictive skill of the coupled models and the MME-based predictions of intraseasonal monsoon fluctuations for 2–3 weeks, providing encouraging results. The MME forecast in 2010 is also able to provide useful guidance, well in advance, about an active September associated with a delayed withdrawal of the monsoon and also the heavy rainfall over north Pakistan.  相似文献   

Enigmatic aspects of entropy inside the black hole: what do falling comoving observers see?     

C. Sivaram  Kenath Arun 《Astrophysics and Space Science》2012,337(1):169-172

In a recent paper it was suggested that inclusion of mutual gravitational interactions can give a possible scenario for reversing gravitation collapse and averting a singular phase. We extend this idea to the still unsolved problem of matter collapsing beyond black hole event horizons. For a comoving observer there is no change in entropy as he goes through the horizon. Matter collapses to a minimum radius, and then can re-expand with the same entropy. It is shown that phase space inside a collapsing black hole is also invariant.  相似文献   

Identification and characterization of size-segregated bioaerosols at Jawaharlal Nehru University, New Delhi   总被引：1，自引：1，他引：0  

Arun Srivastava  Manish Singh  V. K. Jain 《Natural Hazards》2012,60(2):485-499

Ambient levels of viable bioaerosol were measured at four different sites on the campus of Jawaharlal Nehru University (JNU), New Delhi, India. Most of the viable bioaerosol was associated with the fungal fraction with the remainder due to bacteria. The highest concentrations of fungi were found at a health center (1,293 cfu/m³). The highest concentrations of both gram-positive (338.8 cfu/m²) and gram-negative (614 cfu/m²) bacteria were found at a garbage dump site. Gram-negative bacteria were found in larger numbers than gram-positive, possibly due to the higher ambient temperatures during their sampling. Most of the fungal bioaerosol identified is associated with immunotoxic diseases such as sick building syndrome and allergic diseases and was found in respirable fractions. Since the bacterial fractions were identified on the basis of morphology, i.e. only to the level of genus, their relationship to health effects could not be established. Although most of the sampling occurred indoors in naturally ventilated buildings, external temperature and humidity did not vary significantly during the study. Thus, it is more likely that the types of organisms and levels observed had more to do with the sources (decomposing garbage, spores carried from wooded areas, etc.) than the ambient or indoor environmental conditions.  相似文献   

Prediction of algal blooms using genetic programming   总被引：2，自引：0，他引：2  

C. Sivapragasam  Nitin Muttil  V.M. Arun 《Marine pollution bulletin》2010,60(10):1849-1855

In this study, an attempt was made to mathematically model and predict algal blooms in Tolo Harbor (Hong Kong) using genetic programming (GP). Chlorophyll plays a vital role in blooms and was used in this model as a measure of algal bloom biomass, and eight other variables were used as input for its prediction. It has been observed that GP evolves multiple models with almost the same values of errors-of-measure. Previous studies on GP modeling have primarily focused on comparing GP results with actual values. In contrast, in this study, the main aim was to propose a systematic procedure for identifying the most appropriate GP model from a list of feasible models (with similar error-of-measure) using a physical understanding of the process aided by data interpretation. Evaluation of the GP-evolved equations shows that they correctly identify the ecologically significant variables. Analysis of the final GP-evolved mathematical model indicates that, of the eight variables assumed to affect algal blooms, the most significant effects are due to chlorophyll, total inorganic nitrogen and dissolved oxygen for a 1-week prediction. For longer lead predictions (biweekly), secchi-disc depth and temperature appear to be significant variables, in addition to chlorophyll.  相似文献   

Adaptive multi‐rate interface: development and experimental verification for real‐time hybrid simulation          下载免费PDF全文

Amin Maghareh  Jacob P. Waldbjørn  Shirley J. Dyke  Arun Prakash  Ali I. Ozdagli 《地震工程与结构动力学》2016,45(9):1411-1425

Real‐time hybrid simulation (RTHS) is a powerful cyber‐physical technique that is a relatively cost‐effective method to perform global/local system evaluation of structural systems. A major factor that determines the ability of an RTHS to represent true system‐level behavior is the fidelity of the numerical substructure. While the use of higher‐order models increases fidelity of the simulation, it also increases the demand for computational resources. Because RTHS is executed at real‐time, in a conventional RTHS configuration, this increase in computational resources may limit the achievable sampling frequencies and/or introduce delays that can degrade its stability and performance. In this study, the Adaptive Multi‐rate Interface rate‐transitioning and compensation technique is developed to enable the use of more complex numerical models. Such a multi‐rate RTHS is strictly executed at real‐time, although it employs different time steps in the numerical and the physical substructures while including rate‐transitioning to link the components appropriately. Typically, a higher‐order numerical substructure model is solved at larger time intervals, and is coupled with a physical substructure that is driven at smaller time intervals for actuator control purposes. Through a series of simulations, the performance of the AMRI and several existing approaches for multi‐rate RTHS is compared. It is noted that compared with existing methods, AMRI leads to a smaller error, especially at higher ratios of sampling frequency between the numerical and physical substructures and for input signals with high‐frequency content. Further, it does not induce signal chattering at the coupling frequency. The effectiveness of AMRI is also verified experimentally. Copyright © 2016 John Wiley & Sons, Ltd.  相似文献   

Attenuation of Coda Waves in the Saurashtra Region, Gujarat (India)     

Babita Sharma  Dinesh Kumar  S. S. Teotia  B. K. Rastogi  Arun K. Gupta  Srichand Prajapati 《Pure and Applied Geophysics》2012,169(1-2):89-100

The attenuation characteristics based on coda waves of two areas—Jamnagar and Junagarh of Saurashtra, Gujarat (India)—have been investigated in the present study. The frequency dependent relationships have been developed for both the areas using single back scattering model. The broadband waveforms of the vertical components of 33 earthquakes (Mw 1.5–3.5) recorded at six stations of the Jamnagar area, and broadband waveforms of 68 earthquakes (Mw 1.6–5) recorded at five stations of the Junagarh area have been used for the analysis. The estimated relations for the Junagarh area are: Q _c?=?(158?±?5)f^(0.99±0.04) (lapse time : 20?s), Q _c?=?(170?±?4.4)f^(0.97±0.02) (lapse time : 30?s) and Q _c?=?(229?±?6.6)f^(0.94±0.03) (lapse time : 40?s) and for the Jamnagar area are: Q _c?=?(178?±?3)f^(0.95±0.05) (lapse time : 20?s), Q _c?=?(224?±?6)f^(0.98±0.06) (lapse time : 30?s) and Q _c?=?(282?±?7)f^(0.91±0.03) (lapse time : 40?s). These are the first estimates for the areas under consideration. The Junagarh area appears to be more attenuative as compared to the Jamnagar area. The increase in Q _c values with lapse time found here for both the areas show the depth dependence of Q _c as longer lapse time windows will sample larger area. The rate of decay of attenuation (Q ^?1) with frequency for the relations obtained here is found to be comparable with those of other regions of the world though the absolute values differ. A comparison of the coda-Q estimated for the Saurashtra region with those of the nearby Kachchh region shows that the Saurashtra region is less heterogeneous. The obtained relations are expected to be useful for the estimation of source parameters of the earthquakes in the Saurashtra region of Gujarat where no such relations were available earlier. These relations are also important for the simulation of earthquake strong ground motions in the region.  相似文献   

Variation in the intrinsic shapes of elliptical galaxies     

D. K. Chakraborty  Arun Kumar Singh  Firdous Gaffar 《Monthly notices of the Royal Astronomical Society》2008,383(4):1477-1484

  相似文献   

Assessment of genotoxic and humoral immune system alterations in silica exposed workers from pottery industries in South India     

Balasubramanian Balamuralikrishnan  Vellingiri Balachandar  Mohana Devi Subramaniam  Karthick Kumar Alagumuthu  Shanmugam Sureshkumar  Meyyazhagan Arun  Sundaramoorthy Arun  Krishnan Padmavathi  Abdul Hakkim Razeena  Mohan Gomathi  Palanivel Velmurugan  Shahnaz N. Dharwadkar  Keshavarao Sasikala 《Stochastic Environmental Research and Risk Assessment (SERRA)》2014,28(7):1801-1814

Chemicals, especially silica, have been suspected to cause genetic alterations in pottery industry workers. The present study aims to analyze the frequency of chromosomal aberrations (CA), micronucleus (MN) and DNA damage (comet assay) in the peripheral blood lymphocytes and the immunological alterations workers chronically exposed to silica and in control subjects. In our study, 50 silica exposed workers and 35 control subjects were recruited and silica level was measured by respirable dust and respirable quartz concentrations of badge dosimeter. The serum immunoglobulins (IgM, IgG, and IgA) of all the subjects were measured by using ELISA method. The individuals exposed to silica have a significant increase in the frequency of CA, MN and the total DNA damage (p < 0.05). Immunoglobulin elevation on silica exposed workers was statistically significant (p < 0.05) on comparison with their respective controls. Investigation of the smoking and alcohol habitats coupled with silica exposure in exposed and control subjects represents alcohol consumption and smoking as additional risk factors and must be avoided. Multiple linear regression analysis obtained for CA, MN and comet assay confirm these tests as biomarkers for silica exposed pottery workers. Some confounding factors also showed significant influence on exposed subjects. These results indicate the mutagenic risk in the working environment has a high probability of association with the silica dust exposure in pottery industries. Nevertheless, the present study will create awareness and public concern not only among the silica exposed workers but also to the welfare of their progeny.  相似文献   

Predictable patterns of the Asian and Indo-Pacific summer precipitation in the NCEP CFS     

Jianyin Liang  Song Yang  Zeng-Zhen Hu  Bohua Huang  Arun Kumar  Zuqiang Zhang 《Climate Dynamics》2009,32(7-8):989-1001

The predictable patterns of the Asian and Indo-Pacific summer precipitation in the NCEP climate forecast system (CFS) are depicted by applying a maximized signal-to-noise empirical orthogonal function analysis. The CFS captures the two most dominant modes of observed climate patterns. The first most dominant mode is characterized by the climate features of the onset years of El Niño-Southern Oscillation (ENSO), with strong precipitation signals over the tropical eastern Indian and western Pacific oceans, Southeast Asia, and tropical Asian monsoon regions including the Bay of Bengal and the South China Sea. The second most dominant mode is characterized by the climate features of the decay years of ENSO, with weakening signals over the western-central Pacific and strengthening signals over the Indian Ocean. The CFS is capable of predicting the most dominant modes several months in advance. It is also highly skillful in capturing the air–sea interaction processes associated with the precipitation features, as demonstrated in sea surface temperature and wind patterns.  相似文献