Study of a imaging indexing technique in JPEG Compressed domain

In our computers all stored images are in JPEG compressed format even when we download an image from the internet that is also in JPEG compressed format, so it is very essential that we should have content based image indexing its retrieval conducted directly in the compressed domain. In this paper we used a partial decoding algorithm for all the JPEG compressed images to index the images directly in the JPEG compressed domain. We also compare the performance of the approaches in DCT domain and the original images in the pixel domain. This technology will prove preciously in those applications where fast image key generation is required. Image and audio techniques are very important in the multimedia applications. In this paper, we comprise an analytical review of the compressed domain indexing techniques, in which we used transform domain techniques such as Fourier transform, karhunen-loeve transform, Cosine transform, subbands and spatial domain techniques, which are using vector quantization and fractrals. So after comparing other research papers we come on the conclusion that when we have to compress the original image then we should convert the image by using the 8X8 pixels of image blocks and after that convert into DCT form and so on. So after doing research on the same concept we can divide image pixels blocks into 4X4X4 blocks of pixels. So by doing the same we can compress the original image by using the steps further