In this paper we tackle the problem of generating captions for news images. This approach leverages the vast resource of pictures available on the web. Here we have images and its associated articles. Inspired by recent work in summarization, we propose extractive caption generation model. This operates over the output of a image annotation that pre-processes the pictures and suggests keywords to describe their content. After that the article or document is pre-processed and generate keywords. Then extract the most suitable sentence from article which is more overlapped with the keywords from both image and document.Netbeans IDE 7.4 is used for developing this project
