Skip to Main Content
Skip Nav Destination
ASME Press Select Proceedings
International Conference on Advanced Computer Theory and Engineering, 5th (ICACTE 2012)
Xie Yi
Xie Yi
Search for other works by this author on:
No. of Pages:
ASME Press
Publication date:

This paper presents an object-centred image representation based on spatial pyramid for the work of image classification. Unlike other representation, we only consider the salient part of an image, ignoring the unimportant part, as the salient part contributes most of the image meaning. So we first detect the closed contour of the salient object in the image. And next we find a minimum bounding rectangle that bounds the salient object’s contour in order to keep rotation invariant. In the final step, we build the image representation based on the spatial pyramid inside the rectangle. The first advantage of the approach is that it offers a better representation for images with a salient object against a wide background. And the second is that our representation is rotation and translation invariant for the object contained. Experimental results show that our approach is valid.

1. Introduction
2. Related Work
3. Object-Centred Image Representation
4. Experimental Results
5. Conclusion
This content is only available via PDF.
You do not currently have access to this chapter.
Close Modal

or Create an Account

Close Modal
Close Modal