An Analysis of the Traffic Loads in the Servers Using Thermal Images Utilizing Empirical Wavelet Transform with Dyadic Wavelet Transform

This paper mainly focuses on analyzing the traffic loads in the servers using thermal images for green computing based on different load conditions. For training and investigation, an achievement motivation data database with 25,000 objects and five characteristics is used. A classification innovative model was used to analyze the traffic loads of the server using thermal images by incorporating empirical wavelet transform and dyadic wavelet transform algorithms. The classification accuracy of the empirical wavelet transform was 92.09%, and that of dyadic wavelet transform is 88.07%. With a significance score of P = 0.001 ( P < 0.05), there exists a significant statistical distinction between the research institutions. The experimental results provide evidence that the traffic load analysis in the thermal servers provide significant results using thermal images in respect between both precision and exactness.