Description

Here we compress the document image by the use of LZMA algorithm. It is stands for Lempel-Ziv-Markov chain algorithm. It is used for lossless compression. In pre-processing we remove the noise from the image. Then we apply the wavelet transform to the image. It will decompose the image. Here we quantize the image by the use of wavelet. It is more robust under transmission. Then we apply the LZMA algorithm. The total image pixels are considered as a stream bits. Then the bits are separated as a number of packets. Then the packets bits are encoded. For encoding process it will use the adaptive binary range coder. Finally we get the encoded value from the algorithm. After compression we apply the Decompression algorithm. It is the reverse process of compression. The output of decompression will give to the Inverse wavelet transform. Finally it will produce the reconstructed image. we proposed the LZMA algorithm as a best performing one. Here we achieved compression ratios were improved with predictive modeling in place – typically we observed CR improvement between 5% to 15% in comparing with the plain best performing lossless compression algorithm (LZMA). Significance of this extra saving is crucial in the emerging data intensive smart grid technology where huge amount of data will be hosted in enterprise data warehouses but only relatively small portion of that data will be actively analyzed.