Analysis of Cross-Media Web Information Fusion for Text and Image Association- A Survey Paper
The image comprises of the text- and content-based features. Images can be represented using both text-and content-based features. Web information fusion can be defined as the problem of collating and tracking information related to specific topics on the World Wide Web. But the main concern is image and text association, a cornerstone of crossmedia web information fusion. Two methods have been described .The first method based on vague transformation measures the information similarity between the visual features and the textual features through a set of predefined domainspecific information categories. Another method uses a neural network to learn direct mapping between the visual and textual features by automatically and incrementally summarizing the associated features into a set of information templates. Despite their distinct approaches, our experimental results on a terrorist domain document set show that both methods are capable of learning associations between images and texts from a small training data set. Another method en- compasses a variety of techniques relating to document summarization and text- and content-based image retrieval. The text-based approaches described uti- lize the Unified Medical Language System (UMLS) synonymy to identify concepts in information requests and image-related text in order to re-trieve semantically relevant images. Our image content-based approaches utilize similarity metrics based on computed visual concepts” and low- level image features to identify visually similar images. The edge model for image and text association detects sharp changes in image brightness and captures important events and changes in properties of the world. This paper discuss about the analysis of the previous work done so that we can suggest an enhanced fusion method for text and Image association.
