Informatics Publishing Limited

K. Meena ¹, M. Manimekalai ²

Affiliations
1 Bharathidasan University, Tiruchirappalli, Tamil Nadu, IN
2 Department of MCA, Shrimati Indira Gandhi College, Tiruchirappalli, Tamil Nadu, IN

Abstract

The Protein structure prediction has been an active research area for the last 40 years or so. The technical progress in computational Molecular Biology during the last decades has contributed significantly to the progress we see today. The major goal of predicting Protein structures underpins the correct assumption that three dimensional structures confer protein function. The linear Amino Acid sequences must transform to nonlinear Secondary Structures and then to Tertiary and Quaternary Structures that are responsible for biological functions. Biological functions may remain similar or change in the related organisms through the evolutionary process. By considering the importance of the prediction of secondary structure of protein a detailed literature study of the same using Radial Basis Function Neural Networks (RBFNN) AND Support Vector Machines (SVM) has been reviewed in this paper.

Keywords

Radial Basis Function Neural Networks, Support Vector Machines, Secondary Structures, Tertiary Structure And Quaternary Structures

Full Text

     

K. Meena ¹, M. Manimekalai ²

Affiliations
1 Bharathidasan University, Tiruchirappalli, Tamil Nadu, IN
2 Department of MCA, Shrimati Indira Gandhi College, Tiruchirappalli, Tamil Nadu, IN

Abstract

In this paper Prediction of structure of Protein of Soy Beans using Radial Basis Function Neural Networks for RS126 Data set and PDB Data set has been made and compared with other traditional methods namely Chou-Fasman, GOR, APSSP, PHD, Prospect and SSpro.The training and testing sets for both have been taken into consideration to train and test the networks respectively. The major parameter for finding the accuracy of the protein secondary structure prediction is the per-residue prediction accuracy, Q3, which gives the percentage of all correctly predicted residues within the three-state (H, E, C) classes, and has also been employed for assessment of prediction approaches. The performance of the RBFNN protein secondary structure prediction models is evaluated based on their prediction accuracy . The accuracy of the developed approach is compared with other traditional methods to explore the performance of the proposed approach. It is found that the proposed techniques provide a prediction accuracy of about 81% which is very significant. The accuracy for different width of sliding windows. It clearly shows that, with the increase in the sliding window width the accuracy also increases.

Keywords

RBFNN, Prediction Accuracy, Training Set, Test Set, Sliding Window

Full Text

     

M. Pratheepa ¹, K. Meena ², K. R. Subramaniam ³, R. Venugopalan ⁴, H. Bheemanna ⁵

Affiliations
1 National Bureau of Agriculturally Important Insects, Post Bag No. 2491, H. A. Farm Post, Hebbal, Bellary Road, Bangalore 560024, Karnataka, IN
2 Shrimathi Indira Gandhi College, Tiruchirappalli 620002, Tamil Nadu, IN
3 Department of M.C.A., Shrimathi Indira Gandhi College, Tiruchirappalli 620002, Tamil Nadu, IN
4 Indian Institute of Horticultural Research, Hessaraghatta Lake Post, Bangalore 560089, Karnataka, IN
5 Department of Entomology, Agricultural Research Station, Raichur 584101, Karnataka, IN

Abstract

An attempt was made to study the effect of abiotic and naturally occurring biotic factors on Helicoverpa armigera (Hubner) (Lepidoptera: Noctuidae) with cotton as a model crop system. The results revealed that Chrysoperla sp. (carnea-group) (r = 0.344) was positively correlated with pest incidence and the weather parameters like maximum temperature (r = -0.309) and rainfall (r = -0.288) were negatively correlated with pest incidence. It was observed that post-monsoon season was most favourable for pest occurrence and it was more when the crop was in flowering and boll formation stage. Spiders and Chrysoperla sp. (carnea-group) were positively correlated with pest incidence during winter.

Keywords

Helicoverpa armigera, Cotton, Season, Spiders, Chrysopids.

Full Text

   

M. Pratheepa ¹, K. Meena ², K. R. Subramaniam ², H. Bheemanna ³

Affiliations
1 Department of M.C.A., Shrimathi Indira Gandhi College, Trichirappalli 620 002, Tamil Nadu, IN
2 Bharathidasan University, Trichirappalli 620 024, Tamil Nadu, IN
3 Department of Entomology, University of Agricultural Sciences, Raichur 584 102, IN

Abstract

The mirid bug, Creontiodes biseratense (Distant) (Hemiptera: Miridae) is as a serious pest of cotton crop. Forecasting model by linking the pest incidence with season, crop phenology, biotic and abiotic factors enable to understand the dynamics of pest occurrence likely to occur. A data mining technique decision tree induction model is proposed for forecasting the pest incidence and study the population dynamics of mirid bug, C. biseratense in relation to its natural enemies viz., spider Lycosa sp. and coccinellid Cheilomenes sexmaculata Fabricius and abiotic factors. The results of the decision tree agreed well with statistical analysis.

Keywords

Creontiodes biseratense, Cotton, Spiders, Coccinellids, Decision Tree, Information Theory, Abiotic.

Full Text

   

M. Pratheepa ¹, K. Meena ², K . R. Subramaniam ¹, R. Venugopalan ³, H. Bheemanna ⁴

Affiliations
1 Dept. of Computer Science, Shrimathi Indira Gandhi College, Tiruchirappalli 620 002, Tamil Nadu., IN
2 Bharathidasan University, Tiruchirappalli 620 024, Tamil Nadu., IN
3 Department of Entomology, University of Agricultural Science, Raichur 584 102, Karnataka, IN
4 Indian Institute of Horticultural Research, Hessaraghatta Lake Post, Bangalore 560 089, Karnataka, IN

Abstract

The cotton bollworm, Helicoverpa armigera (Hiibner) is an important pest in India damaging cotton crop and resulting in economic loss. Accurate and timely prediction of the pest, considering biotic and abiotic factors is essential to reduce the crop loss. In this paper, we present a neural-network classifier for predicting the pest incidence on cotton by considering the season, crop phenology, biotic factors (spiders and Chrysoperla zastrowi sillemi) and abiotic factors such as maximum temperature, minimum temperature, rainfall and relative humidity. Single layer perceptron neural-network with back-propagation algorithm was utilized for the design of the presented intelligent system. Decision tree is presented from the proposed trained neural-network. The results showed that the supervised neural network system could classify or predict the pest incidence as either 'high' or 'low' based upon economic threshold level with high degree of accuracy. Extracting rules from the decision tree helps the user to understand the role of biotic and abiotic factors on H. armigera incidence.

Keywords

Back-Propagation Algorithm, Biotic And Abiotic Factors, Helicoverpa armigera, Knowledge Extraction, Neuralnetwork Classifier, Pest Prediction.

Full Text

   

K. P. Mohammed Basheer ¹, K. Meena ², K. R. Subramanian ³

Affiliations
1 Department of Computer Applications, Shrimati Indira Gandhi College, Tiruchirappalli, Tamilnadu, IN
2 Bharathidasan University & Research Guide, IN
3 Department of Computer Applications, Shrimati Indira Gandhi College, Tiruchirappalli, IN

Abstract

With the huge development of broadband network, distribution of multimedia by means of Internet is an uncomplicated method of communication and data exchange. Intellectual Property (IP) protection is a vital component in a multimedia broadcast system. Traditional IP protection methods can be classified into two major categories: encryption and watermarking. Content protection has turned out to be one of the most considerable and demanding problems of this field. This paper proposes a multimedia content protection framework that is dependent on biometric data of the users, a layered encryption/decryption scheme and watermarking. Scalable encryption algorithms result from a transaction between implementation cost and resulting performances. In addition, this approach generally aims to be exploited competently on a large range of platforms. The computational necessities and applicability of the proposed method are addressed. By utilizing the benefit of the nature of cryptographic schemes and digital watermarking, the copyright of multimedia contents can be protected. In this paper, the scalable transmission technique is utilized over the broadcasting environment for encryption. The embedded watermark can be thus extracted with high confidence.

Keywords

Multimedia, Security, Biometrics, Watermarking, Scalable Encryption.

Full Text

     

K. Meena , M. Manimekalai

Abstract

One of the most significant problems in biomedical research today is the prediction of protein structure from knowledge of the primary amino acid sequence. Secondary Structure Prediction (SSP) is a very typical problem in the field of bioinformatics.Prediction of secondary structure of Proteins can be done from the Protein sequence. In the Protein structure prediction, the Amino Acid sequence of a Protein, the so-called primary structure, can be easily determined from the sequence on the Gene that codes for it. This primary structure exclusively determines a structure in its native environment. Thus primary structure plays a key role in understanding the function of the Protein. Majority of the previous research have ignored the influence of residue conformational preference on structure prediction of proteins. The primary focus of this research is to investigate a variety of approaches for employing ANN and Machine Learning techniques in order to predict the secondary structure of proteins in soybeans.

Keywords

Protein Structure Prediction, RBFNN, MELM, SVM, Amino Acid, Soybeans.

Full Text

     

K. Meena ¹, M. Manimekalai ²

Affiliations
1 Shrimati Indira Gandhi College, Trichy, IN
2 Dept. of IT & Applications, Shrimati Indira Gandhi College, Trichy, IN

Abstract

Proteins are a definite kind of biological macromolecules that is present in all biological organisms. Amino acids are the building blocks of proteins. They are primary structure, secondary structure, tertiary structure and quaternary structure. Most of the existing algorithms for predicting the content of the protein secondary structure elements have been based on the conventional amino acid composition, where no sequence coupling effects are taken into consideration. Prediction of three dimensional structure, secondary structure, and functional sites of proteins from primary structure are the three major problems in structural bioinformatics. More than a few different approaches have been previously used in these kind of predictions among which, artificial neural networks have been of great interest due to their capability of learning from observations and prediction of the structures for non classified instances. This paper proposes a technique for prediction of protein structure in soybeans using neural networks. This paper uses RBFNN in order to predict the secondary structure. In our approach, genetic encoding scheme is used to generate the centers and widths of radial basis function. In our approach, genetic encoding scheme is used to generate the centers and widths of radial basis function. The neural network architecture used in our approach is a feed forward and fully connected neural network whose Gaussian centers are optimized by genetic algorithm. Experimental are carried on dataset obtained from Protein Data Bank (PDB) to predict the structure of the protein present in it.

Keywords

Amino Acids (AA), Bioinformatics, Protein Structure Prediction, Secondary Structure Content, Neural Networks, Radial Basis Function Neural Networks (RBFNN), Genetic Algorithm (GA), Protein Data Bank (PDB).

Full Text

     

R. Reena Rose ¹, A. Suruliandi ², K. Meena ³

Affiliations
1 Department of Computer Applications, St. Xavier’s Catholic College of Engineering, IN
2 Department of Computer Science and Engineering, Manonmaniam Sundaranar University, IN
3 Department of Electronics and Communication Engineering, J. P. College of Engineering, IN

Abstract

Texture descriptors have an important role in recognizing face images. However, almost all the existing local texture descriptors use nearest neighbors to encode a texture pattern around a pixel. But in face images, most of the pixels have similar characteristics with that of its nearest neighbors because the skin covers large area in a face and the skin tone at neighboring regions are same. Therefore this paper presents a general framework called Local Texture Description Framework that uses only eight pixels which are at certain distance apart either circular or elliptical from the referenced pixel. Local texture description can be done using the foundation of any existing local texture descriptors. In this paper, the performance of the proposed framework is verified with three existing local texture descriptors Local Binary Pattern (LBP), Local Texture Pattern (LTP) and Local Tetra Patterns (LTrPs) for the five issues viz. facial expression, partial occlusion, illumination variation, pose variation and general recognition. Five benchmark databases JAFFE, Essex, Indian faces, AT & T and Georgia Tech are used for the experiments. Experimental results demonstrate that even with less number of patterns, the proposed framework could achieve higher recognition accuracy than that of their base models.

Keywords

Face Recognition, Local Texture Description Framework, Nearest Neighborhood Classification, Chi-Square Distance Metric.

Full Text

     

A. Suruliandi ¹, R. Reena Rose ², K. Meena ³

Affiliations
1 Department of Computer Science and Engineering, Manonmaniam Sundaranar University, IN
2 Department of Computer Applications, St. Xavier’s Catholic College of Engineering, IN
3 Department of Computer Science and Engineering, Sardar Raja College of Engineering, IN

Abstract

Texture feature plays a predominant role in recognizing face images. However different persons can have similar texture features that may degrade the system performance. Hence in this paper, the problem of face similarity is addressed by proposing a solution which combines textural and geometrical features. An algorithm is proposed to combine these two features. Five texture descriptors and few geometrical features are considered to validate the proposed system. Performance evaluations of these features are carried out independently and jointly for three different issues such as expression variation, illumination variation and partial occlusion with objects. It is observed that the combination of textural and geometrical features enhance the accuracy of face recognition. Experimental results on Japanese Female Facial Expression (JAFFE) and ESSEX databases indicate that the texture descriptor Local Binary Pattern achieves better recognition accuracy for all the issues considered.

Keywords

Face Recognition, Texture Features, Geometric Features, Nearest Neighborhood Classification, Chi-Square Distance Metric.

Full Text

     

K. Meena ¹, A. Suruliandi ², R. Reena Rose ³

Affiliations
1 Department of Electronics and Communication Engineering, J. P. College of Engineering, IN
2 Department of Computer Science and Engineering, Manonmaniam Sundaranar University, IN
3 Department of Computer Applications, St. Xavier’s Catholic College of Engineering, IN

Abstract

Automatic face recognition remains an interesting but challenging computer vision open problem. Poor illumination is considered as one of the major issue, since illumination changes cause large variation in the facial features. To resolve this, illumination normalization preprocessing techniques are employed in this paper to enhance the face recognition rate. The methods such as Histogram Equalization (HE), Gamma Intensity Correction (GIC), Normalization chain and Modified Homomorphic Filtering (MHF) are used for preprocessing. Owing to great success, the texture features are commonly used for face recognition. But these features are severely affected by lighting changes. Hence texture based models Local Binary Pattern (LBP), Local Derivative Pattern (LDP), Local Texture Pattern (LTP) and Local Tetra Patterns (LTrPs) are experimented under different lighting conditions. In this paper, illumination invariant face recognition technique is developed based on the fusion of illumination preprocessing with local texture descriptors. The performance has been evaluated using YALE B and CMU-PIE databases containing more than 1500 images. The results demonstrate that MHF based normalization gives significant improvement in recognition rate for the face images with large illumination conditions.

Keywords

Face Recognition, Texture Analysis, Texture Features.

Full Text

     

T. Chakravarthy ¹, K. Meena ², D. Nathiya ¹

Affiliations
1 A. V. V. M. Sri Pushpam College, Bharathidasan University, Tamil Nadu, IN
2 Shrimati Indira Gandhi College, Bharathidasan University, Tamil Nadu, IN

Abstract

Fingerprint classification based on statistical and structural (RNN and SVM) approach. RNNs are trained on a structured representation of the fingerprint image. They are also used to extract a set of distributed features of the fingerprint which can be integrated in this support vector machine. SVMs are combined with a new error correcting codes scheme. This approach has two main advantages. (a) It can tolerate the presence of ambiguous fingerprint images in the training set and (b) It can effectively identify the most difficult fingerprint images in the test set. In this experiment on the fingerprint database NIST-4 (National Institute of Science and Technology), our best classification accuracy of 94.7% is obtained by training SVM on both fingerCode and RNN -extracted futures of segmentation algorithm which has used very sophisticated "region growing process".

Keywords

Support Vector Machine, Recursive Neural Network, Region Growing, Error Correction Code.

Full Text