T. Karthikeyan ¹, P. Manikandaprabhu ¹

Affiliations
1 PG and Research Department of Computer Science, PSG College of Arts and Science, Coimbatore - 641014, Tamil Nadu, IN

Abstract

In this work, an urban area land cover is proposed to classify the large resolution image. It aims to extract the features like texture, shape, size and spectral information in the feature extraction process. Embedded Zero tree Wavelet transform is a lossy image compression algorithm. Most of the coefficients at low bit rates bent through a sub band transform will be zero, or very close to zero. These features data are used for the classification process. Here, we used various classification algorithms namely, Radial Basis Function, SMO, Multilayer Perceptron and Random Forest are implemented. The classification accuracy constantly depends on the efficiency of the extracted features and classification algorithms. The result of the proposed classification algorithms are merged with EZW. Experimental results illustrate that the better accuracy performance is obtained by the Multilayer Perceptron algorithm than other classification algorithms.

Keywords

Artificial Neural Network, Embedded Zero Tree Wavlet, Feature Extraction, Image Classification, Multilayer Perceptron, Radial Basis Function

Full Text

   

T. Karthikeyan ¹, C. Thirumoorthi ²

Affiliations
1 Department of Computer Science, PSG College of Arts and Science, Coimbatore, IN
2 Department of Computer Science, Hindustan College of Arts and Science, Coimbatore, IN

Abstract

Objectives: This study focuses on Image compression and compares different methods. Methods/Statistical Analysis: In this work we simulated four image compression methods. The first method is focused on Karhunen-Loève Transforms (KLT), second method is focused on Walsh-Hadamard Transforms (WHT), third method based on FFT and fourth one is proposed sFFT. Findings: The experimental outcomes are compared with the different quality of parameters applying on numerous lung cancers CT scan images. The Proposed SFFT method algorithm was given better results like Peak Signal to Noise Ratio (PSNR), Structural Content (SC), Mean Square Error (MSE) and Compression Ratio (CR) are compare to other Transform methods. Application /Improvement: The Proposed SFFT technique gives improved result compared with other methods in all evaluation measures.

Keywords

CR, FFT, Image Compression, KLT, Lung Cancer CT Images, MSE, PSNR, Proposed sFFT, SC, WHT.

Full Text

   

M. Dev Anand ¹, G. Glan Devadhas ², N. Prabhu ³, T. Karthikeyan ⁴

Affiliations
1 Department of Mechanical Engineering, Noorul Islam Centre for Higher Education, Kumaracoil - 629 180, Thuckalay, Kanyakumari District, Tamil Nadu, IN
2 Department of Electronics and Instrumentation Engineering, Noorul Islam University, Kumaracoil - 629 180, Thuckalay, Kanyakumari District, Tamil Nadu, IN
3 Department of Mechanical Engineering, Kottayam Institute of Technology and Science (KITS), Chengalam East, Pallickathodu, Kottayam - 686585, Kerala, IN
4 Department of Mechanical Engineering, Saveetha Nagar, Thandalam, Chennai - 602105, Tamil Nadu, IN

Abstract

A ceramic monolith heat exchanger has been learnt for finding out heat transfer performance and effectiveness on computing numerically and ξ-NTU method. In entire domain computation numerically has been performed along with fluid region in rectangular ducts of exhaust gas side, ceramic core and rectangular duct fluid region in air side with the air exhaust in direction of cross flow. Additionally, the heat exchanger has been examined for estimating the functionality via ξ-NTU technique that is conventional along numerous Nusselt number links for rectangular duct flow for the literature. Based on the research, it has been performed on the ceramic heat exchangers and on the ceramic materials and the demand in utilizing the ceramic materials in heat exchangers. Then the recuperator is modeled by using GAMBIT and it is analyzed using FLUENT. The effectiveness and the heat transfer rate are also calculated. Then those outcomes have been assessed along the experimental data. By comparison of both functionality by computing numerically and the ξ-NTU technique, the efficiency by ξ-TU technique has been identified to be nearest to product by the numerical computation among the associative of 2.15% when Stephan's Nusselt number association has been adapted to the ξ-NTU technique within numerous connections. The total heat transfer and effectiveness by ξ-NTU method relative errors utilizing five Nusselt number correlations from literature have been lesser than 14.5% comparative to numerical computation. Associated to Nusselt number correlations, the entire heat transfer utilizing ξ-NTU technique with Stephan's correlation is highly nearest to numerical computation. For that reason, the exit temperature by ξ-NTU method with Stephan's correlation simulates within 1.2% of the relative error for exhaust exist temperature and 0.45% for the air exit temperature assessed against the numerical computation. Overall heat transfer coefficient's relative errors by ξ-NTU technique utilizing five Nusselt number correlations for the literature have been more than 17.5% to that on computing numerically.

Keywords

Ceramic Recuperators, Cross Flow, Effectiveness, Heat Transfer, Pressure Drop

Full Text