Informatics Publishing Limited

T. Sridevi ¹, A. Murugan ²

Affiliations
1 Mother Teresa Women’s University, Kodaikanal, Tamil Nadu, IN
2 Department of Computer Science, Dr. Ambedkar Govt. Arts College, Chennai, Tamil Nadu, IN

Abstract

Data mining (DM) techniques are used to determine interesting patterns from different domains according to the need of applications and the analyst. Medical field is one among the major user of the mining technology for diagnosing the attributes for the medical issues. Breast cancer is one of the most important medical problems. The modern researchers and technological advancements attempted to determine the cause and prevention in an effective manner with lesser number of attributes. But the diagnosis is lengthy process with multiple and multilevel attribute analysis in certain cases. In order to improve the accuracy of diagnosis with limited attributes, in this paper rough set based relative reduct algorithm is used to reduce the number of attributes using equivalence relation. The effectiveness of proposed Rough Set Reduction algorithm is analyzed on Wisconsin Breast Cancer Dataset (WBCD) and presented as a part of the paper. The experimental results show that the relative reduct performs better attribute reduction.

Keywords

Data mining, Data Preprocessing, Rough Set, Data reduction, Breast Cancer Diagnosis

Full Text

   

T. Sridevi ¹, Habibulla Khan ¹, Ch. Sindhu ¹, P. Naga Praveen Kumar ², P. Sagar ¹

Affiliations
1 Department of ECE, KL University, Vijayawada, A.P, IN
2 Department of ECE, KL University, Vijayawada, A.P, IN

Abstract

Number Plate Recognition systems are used to track and monitor the moving vehicles by automatically extracting the number plates. The objective of this system is to recognize vehicles based on license plate information. Number plate recognition is part of vehicle identification system. Now a days it has wide range of applications like traffic surveillance, access control etc. The images of passing vehicles are taken at surveillance system and those images will be processed. The Proposed method uses simple morphological open and close operations using different structuring elements for plate feature extraction, Labeling the connected pixels, searching the plate location based on Geometrical conditions, segmenting the number plate and character recognition with Neural Network of Multilayer Perceptron. We have proposed a new method for plate segmentation based on Labeling. This method has been tested using a database of Indian number plates and results achieved have shown the high detection rate than existing methods.

Keywords

Morphological Operations, Labeling, Plate Segmentation, Multi Layer Perceptron.

Full Text

     

Y. Ramadevi ¹, T. Sridevi ¹, B. Poornima ¹, B. Kalyani ¹

Affiliations
1 Department of CSE, Chaitanya Bharathi Institute of Technology, Gandipet, Hyderabad, IN

Abstract

Image segmentation is to partition an image into meaningful regions with respect to a particular application. Object recognition is the task of finding a given object in an image or video sequence. In this paper, interaction between image segmentation (using different edge detection methods) and object recognition are discussed. Edge detection methods such as Sobel, Prewitt, Roberts, Canny, Laplacian of Guassian(LoG) are used for segmenting the image. Expectation-Maximization (EM) algorithm, OSTU and Genetic algorithms were used to demonstrate the synergy between the segmented images and object recognition.

Keywords

EM Algorithm, OSTU, Genetic Algorithm, Image Segmentation, Object Recognition.

Full Text