Enhancing the quality of compressed images using rounding intensity followed by novel dividing technique

The most extensive data storage size is significantly increasing, and storing such data, including images to make them obtainable over the network, has become a significant problem. Thus, efficient compression techniques are necessary because the original images need much disk space. Generally, the primary purpose of image compression is to decrease the amount of data required for representing digital images. Image compression techniques can be categorized into lossless and lossy. Rounding Intensity followed by the Dividing Technique (FIRD) is a known lossy technique utilized in the proposed method that aims to reduce the range of the intensities and increase redundancy, achieving better compression performance. The main idea of this paper is to separate the (8*8 block of mage) into two parts: the matrix of hundreds and the single matrix. The elements in the hundreds matrix will be divided into 10, and the resulted matrix will be compressed using the Huffman technique. The elements in the single matrix will be rounded as follow: 0,1,2,3, and 4 to 2, and the elements 5, 6, 7, 8, and 9 are rounded to 7. Then, the Huffman technique will be applied to the resulting matrix. The previously generated values are combined, and the initial results indicate a reduction in distortion with better compression performance. This paper deals with whether the proposed technique enhancing the quality of Compressed Images using Rounding Intensity Followed by the Dividing Technique (E-RFID) can be used as a distinctive option in the distortion reduction process image a high compression ratio. In addition, image compression technology and quality enhancement with high compression explain each step's work, which will be presented in detail. The research factors have been positively impacted by improving the results and comparing them through MSE, PSNR, and MAE. The E-RIFD technique experiments showed an improved rate in MSE 66.82% and PSNR 18.31%, while the MAE improvement rate was 21.09% for grayscale images. However, when testing color images, the results show that the rate of improvement in MSE was 66.59% and PSNR 15.28, while the improvement rate in the MAE was 18.64%.