The Internet image retrieval is an interesting task that needs efforts from image processing and relationship structure analysis. In this paper, has been proposed compressed method when you need to send more than a photo via the internet based on image retrieval. First, face detection is implemented based on local binary patterns. The background is notice based on matching global self-similarities and compared it with the rest of the image backgrounds. The propose algorithm are link the gap between the present image indexing technology, developed in the pixel domain, and the fact that an increasing number of images stored on the computer are previously compressed by JPEG at the source. The similar images are found and send a few images instead of huge images with the same parameters. This provides users in social media with efficient tools to search their images and establish a database or a collection of special images to their interests. Experiments using face detection and background notice based indexing techniques support the idea that the proposed algorithm achieves significantly better results in terms of computing cost. This technology will support control the explosion of media-rich content by offering users an efficient automated image indexing and retrieval tool for compressed images on the Internet
In order to serve the field of knowledge in the field of design and analysis of experiments by adopting formulas and methods that are easy and simple in arithmetic work, especially in experiments that require many calculations for squares, including tests of type