In this paper, we present a neural approach for unsupervised natural color-texture image segmentation, which is based on both Kohonen maps and mathematical morphology, using a combination of the texture and the image color information of the image, namely, the fractal features based on fractal dimension are selected to present the information texture, and the color features presented in RGB col...

The major drawback of watershed transformation is over-segmentation. It also has a significant advantage: very good edge extraction. Thresholding methods usually utilize only global information such as an image histogram; however, they have the ability to group pixels into clusters by their value. The method presented in this paper combines the advantages of watershed segmentation and multileve...

A watershed-based segmentation approach that uses iterative bilateral smoothing with an adapting photometric similarity parameter for pre-filtering is adopted for the segmentation of rock scenes. But the resultant segmented images also contain non-rock watershed regions which are not desired for measuring rock sizes. A proximity-based classifier is applied for the removal of the latter using fe...

This paper presents the statistical analysis for the quality check of membrane filter, used for the bacteria filtration, using the image processing techniques. The quality of membrane filter is determined by the uniformity factor of the polymer beads under test. We compare the performance of well known watershed method and active contour method used for the segmentation process. The simulation ...

In this paper, we propose a video object segmentation algorithm based on spatio-temporal information and marked-watershed for extracting the moving objects from a video sequence. The algorithm begins with difference image between two adjacent frames and, using the Canny operator on the difference image, determines the initial edge mask for the object in motion. Morphological techniques are appl...

A new watershed-based technique is proposed for the segmentation of multiresolution remote-sensing images. These images are composed by a high-resolution panchromatic band and a low-resolution multispectral set. To achieve a segmentation with the high resolution of the panchromatic image and the high accuracy granted by the spectral information, the two components are processed jointly, using b...

This paper addresses the problem of motion segmentation using a multilayer representation. At first, the coarse optic flow is estimated using the robust Simultaneous-Over-Relaxation (SOR) algorithm, meanwhile the intensity segmentation is performed by a watershed algorithm. The image is divided into nonoverlapping rectangular tiled regions, within those subregions affine motion models are fitte...

Character segmentation is the major step of document image analysis and optical character recognition (OCR). The character segmentation is necessary to detect all the character regions in the image document. The proposed method preprocesses the image document with edge detection techniques to enhance the character edges. Further, the watershed algorithm is implemented to identify the regions of...

In this paper, a new segmentation technique for multivalued images is elaborated. The technique accesses multiscale edge information of a multivalued image by a concept, called multiscale fundamental form. At different scales, an edge map of the multivalued images is obtained, at which a watershed-based algorithm is applied. The multiscale behavior of the obtained watershed regions is employed ...

The watershed transform is the method of choice for image segmentation in the field of mathematical morphology. We present a critical review of several definitions of the watershed transform and the associated sequential algorithms, and discuss various issues which often cause confusion in the literature. The need to distinguish between definition, algorithm specification and algorithm implemen...