J. Pal Choudhury

Image recognition and processing using Artificial Neural Network

There are several techniques for image recognition. Among those methods, application of soft computing models on digital image has been considered to be an approach for a better result. The main objective of the present work is to provide a new approach for image recognition using Artificial Neural Networks. Initially an original gray scale intensity image has been taken for transformation. The Input image has been added with Salt and Peeper noise. Adaptive median Filter has been applied on noisy image such that the noise can be removed and the output image would be considered as filtered Image. The estimated Error and average error of the values stored in filtered image matrix have been calculated with reference to the values stored in original data matrix for the purpose of checking of proper noise removal. Now each pixel data has been converted into binary number (8 bit) from decimal values. A set of four pixels has been taken together to form a new binary number with 32 bits and it has been converted into a decimal. This process continues to produce new data matrix with new different set of values. This data matrix has been taken as original data matrix and saved in data bank. Now for recognition, a new test image has been taken and the same steps as salt & pepper noise insertion, removal of noise using adaptive median filter as mentioned earlier have been applied to get a new test matrix. Now the average error of the second image with respect to original image has been calculated based on the both generated matrices. If the average error is more than 45% then a conclusion can be made that the images are different and cannot be matched. But if the value of average error has been found to be less than or equal to 45%, an effort has been made to use the artificial neural network on test data matrix with reference to original data matrix thereby producing a new matrix of the second image(test image). The total average error has been calculated on generated data matrix produced after the application of artificial neural networks on test data matrix to check whether proper identification can be made or not. It has been observed that the value of average error is less than that of test image without application of artificial neural network. Further it has been observed that the test image is matching and recognized with respect to original image.

A framework for the Recognition of Human Emotion using Soft Computing models

Human-computer intelligent interaction (HCII) is an emerging field of science. The interaction between human beings and computers will be more natural if computers are able to perceive and respond to human non-verbal communication such as emotions. The most expressive way humans display emotions is through facial expressions. In this paper a method for emotion recognition from facial images has been proposed. The system consists of three steps. At the very outset some pre-processing has been applied on the input image and face features have been extracted from face images before applying the emotion recognition technique. A comparison between two edge detection techniques-Sobel edge detection and Fuzzy logic based edge detection has been shown. Observation of various emotions characterizes that eye exhibits ellipses of different parameters for different types of emotions. Genetic Algorithm has been applied to optimize the ellipse characteristics of the eye feature. Finally a classification has been carried out by using Back-propagation Neural Network (BPNN). The proposed approaches are tested on a number of face images. General Terms Emotion Recognition, Image Processing.

Prediction of productivity of mustard plant at maturity using harmony search

The production of a plant can be measured by various parameters like shoot length, number of leaves, root length, number of roots etc. During the growth of a plant, new leaves may appear and some leaves may fall. As a result, it is very difficult to predict the growth of the plant based on the leaf numbers. It is also difficult to measure the plant growth on the basis of root numbers as it grows underground. So, it is very convenient to measure the plant growth with respect to shoot length. In this paper, an effort has been made to predict the shoot length of a mustard plant by harmony search. The average error has been calculated based on the actual shoot length and predicted shoot length by harmony search. The result obtained in this paper has been compared with other soft computing and statistical methods which have been applied to the same problem. It has been observed that harmony search has given the minimum error compared to that of the others. The least square method has been applied on the predicted shoot length to find the shoot length at maturity. Finally, the pod yield has been predicted based on the shoot length at maturity.

Application of Neuro-Statistic Model on a Time Series Data

The output of some physical problem is dependent on huge number of time dependent parameters. But many of them are not significant or they are highly correlated with others parameters. Some parameters which are play significant role in the problem and give the information which is mandatory and not correlated with the others. So, same result can be produced by fewer parameters instead of considering all parameters. In this paper, an effect has been made to find the significant environmental parameters in production of mustard plant using principal component and factor analysis. Finally, artificial neural network has been applied on highly significant parameters to predict the production of mustard plant at maturity.

Prediction of Productivity of Mustard Plant Using Variable Reduction and Artificial Neural Network Model

The productivity of mustard plant is dependent on huge number of time dependent parameters. But many of them are not significant or they are highly correlated with others parameters. Some parameters playing significant role in growth of the plant and give the information which is mandatory but not correlated with the others. So, same result can be produced by fewer parameters instead of considering all parameters. In this paper, an effect has been made to reduce the significant environmental parameters which have dominated the growth of mustard plant using principal component and factor analysis. These two methods have been used as variable reduction model. The artificial neural network has been applied on highly significant parameters to predict the shoot length of mustard plant. The linear equation has been used to find the shoot length at maturity. Finally, the productivity of the plant has been predicted based on shoot length of the plant at maturity.