Recognize Facial Impression using Artificial Neural Network

This paper aims to design an artificial neural network to discover the impression by recognizing the expression of the human face. To achieve this goal, the artificial neural network was analyzed and to create patterns of the database containing a set of images with different expressions. The learning process of the network was also conducted through patterns training. The extent to which patterns of online training were recognized was compared to the true values of expressions. The grid was trained in 200 patterns and the anomalies were removed. Then re-learned the network again and analyzed the network performance by comparing the real expression with the expected expression and outputting the error for the network appearing. Impression recognition in the grid applied a three-layer back propagation model, with an average error of 0.321. The performance of the artificial neural network in the recognition of impressions was 80%

Simulating Manual Signature using Elman Back Propagation Model to Create Pseudo Digital Signature

Manual Signatures are used in authentication worldwide. But they are still not used in VANETs and in ad hoc networks for security. In our research we try to use manual signature in place of Digital Signatures for the security of message and stimulated the same by pattern recalling mechanism of Artificial Neural Network using Elman Back Propagation Algorithm to create pseudo digital signature. These pseudo digital signatures are now used as the identity of message sender in communication. We also maintained the speed of manual signature recognition and verification to stop the delay in identification of the sender.

Unbalance Rotor Parameters Detection Based on Artificial Neural Network

Unbalance is an important fault that can damage or shut down vital rotary systems such as the gas turbine, compressors, and others, so to avoid this trouble, the balancing process is very crucial, even though it is time-consuming and costly. Thus, having a technique which can predict the unbalance location and its parameters will be valuable and practical. The current study represents a model that can identify the unbalance’s mass, radius, and location of the eccentric mass based on the artificial neural network (ANN) model. The inputs of the proposed ANN, which is based on a feed forward with back propagation model, is the bearing acceleration signal in the frequency domain. It has 10 hidden layers with 10 neurons through each layer. The accuracy in prediction was acquired at 96%, 96%, and 94% for the disc number (plane), the eccentric radius, and eccentric mass values, respectively.

Forecasting monthly rainfall using autoregressive integrated moving average model (ARIMA) and artificial neural network (ANN) model: A case study of Junagadh, Gujarat, India

The onset, withdrawal and quantity of rainfall greatly influence the agricultural yield, economy, water resources, power generation and ecosystem. Time series modelling has been extensively used in stochastic hydrology for predicting various hydrological processes. The principles of stochastic processes have been increasingly and successfully applied in the past three decades to model many of the hydrological processes which are stochastic in nature. Time lagged models extract maximum possible information from the available record for forecasting. Artificial neural network has been found to be effective in modelling hydrological processes which are stochastic in nature. The ARIMA model was used to simulate and forecast rainfall using its linear approach and the performance of the model was compared with ANN. The computational approach of ANN is inspired from nervous system of living beings and the neurons possess the parallel distribution processing nature. ANN has proven to be a reliable tool for modelling compared to conventional methods like ARIMA and therefore ANN has been used in this study to estimate rainfall. In this study, rainfall estimation of Junagadh has been attempted using monthly rainfall training data of 32 years (1980-2011) and testing data of 5 years (2012-2016). A number of ANN model structures were tested, and the appropriate ANN model was selected based on its performance measures like root mean square error and correlation coefficient. The correlation coefficient Seasonal ARIMA (1,0,0)(3,1,1)12 and ANN back-propagation model (5-12-1) on the testing data was found to be 0.75 and 0.79 respectively. Seasonal ARIMA (1,0,0)(3,1,1)12 and ANN back-propagation model (5-12-1) were used for forecasting rainfall of 5 years (2017-2021).