{"title":"EZW Coding System with Artificial Neural Networks","authors":"Saudagar Abdul Khader Jilani, Syed Abdul Sattar","country":null,"institution":"","volume":39,"journal":"International Journal of Computer and Information Engineering","pagesStart":418,"pagesEnd":424,"ISSN":"1307-6892","URL":"https:\/\/publications.waset.org\/pdf\/14955","abstract":"Image compression plays a vital role in today-s\r\ncommunication. The limitation in allocated bandwidth leads to\r\nslower communication. To exchange the rate of transmission in the\r\nlimited bandwidth the Image data must be compressed before\r\ntransmission. Basically there are two types of compressions, 1)\r\nLOSSY compression and 2) LOSSLESS compression. Lossy\r\ncompression though gives more compression compared to lossless\r\ncompression; the accuracy in retrievation is less in case of lossy\r\ncompression as compared to lossless compression. JPEG, JPEG2000\r\nimage compression system follows huffman coding for image\r\ncompression. JPEG 2000 coding system use wavelet transform,\r\nwhich decompose the image into different levels, where the\r\ncoefficient in each sub band are uncorrelated from coefficient of\r\nother sub bands. Embedded Zero tree wavelet (EZW) coding exploits\r\nthe multi-resolution properties of the wavelet transform to give a\r\ncomputationally simple algorithm with better performance compared\r\nto existing wavelet transforms. For further improvement of\r\ncompression applications other coding methods were recently been\r\nsuggested. An ANN base approach is one such method. Artificial\r\nNeural Network has been applied to many problems in image\r\nprocessing and has demonstrated their superiority over classical\r\nmethods when dealing with noisy or incomplete data for image\r\ncompression applications. The performance analysis of different\r\nimages is proposed with an analysis of EZW coding system with\r\nError Backpropagation algorithm. The implementation and analysis\r\nshows approximately 30% more accuracy in retrieved image\r\ncompare to the existing EZW coding system.","references":null,"publisher":"World Academy of Science, Engineering and Technology","index":"Open Science Index 39, 2010"}