Image Segmentation refers to the process of partitioning a digital image into multiple segments. Image segmentation is typically used
to locate objects and boundaries (lines, curves, etc.) in images. Processing images for specific targets on a large scale has to handle
various kinds of contents with regular processing steps. To segment objects in one image, using the dual multiscale Graylevel
morphological open and close reconstructions (SEGON) algorithm. It can be used to build a Back Ground (BG) gray-level variation
mesh, which is to identify BG and object regions. SEGON roughly identifies the background and object regions in the image. To
further refine the boundaries of the objects mean-shift segmentation technique is applied on the SEGON processed image. Accuracy
of segmentation is evaluated by computing structural similarity index (SSIM), for the image segmented with and without using
SEGON. Images collected from Corel, Caltech, PASCAL, UCID and CMU-PIE databases are utilized for this work. Experimental
results showed that the proposed object segmentation method outperforms the state-of-the-art of other image segmentation techniques.
Both qualitative and quantitative comparison between different segmentation methods is performed and found SEGON to be fast and
reliable method. Content-Based Image Retrieval (CBIR) was carried out to evaluate the object segmentation capability in dealing with
large-scale database images. CBIR is a technique which uses visual contents to search image from large scale database. Relevance
feedback algorithm is used to reduce the semantic gap in CBIR inorder to make the retrieval process efficient.