Volume 7 Number 3 December 2017


Face Recognition Based on Hybridization of PPCA and SIFT Algorithm
S. Anjela Mary, I. Laurence Aroquiaraj

Abstract- In recent years, Face recognition has received substantial attention from both research communities and the market, but still remaining as a challenging in applications. Many face recognition algorithms, along with their medications, have been developed during the past decades. Their performances also suffer a severe degradation under variations in expressions or poses, especially when there is one gallery per subject only. The high resolution (HR) face images nowadays, some HR face database has currently been developed. In this work, a pose invariant face recognition method is presented for high resolution face verification test. A new key point descriptor, namely Pore-PCASIFT is used for extraction features from HR images. According to experimental analysis the proposed method is which is robust to alignment errors, using the HR information based on pore-scale facial features are the better then results.


Mosquito Abundance Forecast
R. Archana

Abstract- The mosquito species is one of most threatening insect vectors of several diseases, namely, malaria, filariasis, chikungunya, dengue and so on. In recent years, as the number of people who involve in outdoor activities continues to increase, the infection caused by these vectors also increases. Mosquito-borne diseases can make people ill and, in severe cases, can cause death. Furthermore, mosquito activity prediction is crucial for managing the safety and the health of humans. This model has been implemented as an effective solution against the spread of fatal diseases.


Performance Analysis of Routing in Wireless Sensor Network Using Optimization Techniques
S. Selvaraj, R. Rathipriya

Abstract- This paper describes the concept of optimization techniques in Wireless Sensor Networks. Wireless Sensor Network consist of many sensor nodes in which each sensor node collects the data from sensing environment and transmit to the base station. Optimization techniques used in wireless sensor networks for minimizing energy consumption generally and for solve routing problems. For improving network lifetime and energy consumption various optimization techniques have been proposed. The paper gives overview of most successful classes of swarm intelligence (SI) based algorithm for solving energy based lifetime optimization problem.


Protein Sequence Segment Selection Using Fuzzy Entropy
R. Keerthiga, K.Thangavel, K. Sasirekha

Abstract- Bioinformatics is the presentation of information technology to the management of molecular genetic data. A protein sequence concept is important role in bioinformatics. In this work, 3000 protein sequences are taken for implementation from PISCES: a protein sequence culling server. The sliding window method is used to produce protein sequence segment which generates 6, 60, 364 sequence segments for 3000 protein sequence. In particular, each sliding window denotes one sequence segments. Here, the produced sequence segments do not have classes or labels. Hence, the generative segment selection technique has to be used to select significant segments. In this work, Fuzzy entropy is proposed to select the optimal segments for further processing. Subsequently, the selected segments are clustered into strong structural similarity clusters and weak structural similarity clusters with the exploit of K-Means clustering algorithm. The Davies-Bouldins Index measure is used to measure the structural similarity for each cluster


MRI Brain Image Segmentation using Soft Computing Techniques
N. Sivaranjani, I. Laurence Aroquiaraj

Abstract- Image processing is a dynamic research zone in which restorative image processing is a very difficult field. Medical imaging methods are utilized to image the internal segments of the human body for restorative diagnosis. Cerebrum tumor is a serious life changing disease condition. Image segmentation assumes a huge part in image processing as it helps in the extraction of suspicious areas from the restorative images. In this dissertation we have proposed segmentation of cerebrum MRI images utilizing Rough-fuzzy clustering algorithm took after by morphological separating which keeps away from the misclustered areas that can unavoidably be framed after segmentation of the cerebrum MRI picture for identification of tumor location.