Automatic image structure analysis
Aster W. Wardhani, Rafael C. González · 2002
The rapid growth of multimedia technology has resulted in an enormous amount of data that needs to be managed and indexed efficiently. To provide effective labeling for an image indexing system requires all objects in the image to be identified. To perform better and effective object identification, the process needs to be performed automatically and without priori knowledge of the image content. We present an automatic object identification scheme, by analysing the image structural information. The image is segmented using automatic image segmentation techniques and components of objects are obtained by grouping the segments together. We present the issues and problems involved in providing such an identification scheme. Some experimental results are presented.