Lossy and Lossless Image Compression Using Convolution Neural Network Algorithm Over Gradient Boost Algorithm with Improved Accuracy

This study focuses on enhancing the accuracy of image compression using two different techniques: the Enhanced Novel CNN and the Gradient Boost algorithm. The goal is to see which method performs better in compressing images, both with lossy and lossless compression. In the study, both algorithms were tested on image datasets with a sample size of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{1 0}$</tex> for each. The Enhanced Novel Convolutional Neural Network showed a significant improvement, achieving an accuracy rate of 94.20 %. This is notably higher than the 80.80 % accuracy rate of the Gradient Boost algorithm. To ensure the results are statistically significant, an independent sample T-test was conducted. The test showed a p-value of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{0. 0 4 4}$</tex>, indicating a meaningful difference between the two methods, with a confidence interval set at <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{9 5 \%}$</tex>. The G Power value, which measures the power of the test, was <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{9 5 \%}$</tex>, supporting the reliability of the findings.