Watersheds in Digital Spaces: An Efficient Algorithm Based on Immersion Simulations
IEEE Transactions on Pattern Analysis and Machine Intelligence
Two-dimensional entropic segmentation
Non-Linear Analysis
Digital Image Processing
A New Approach for Calculating Implications of Fuzzy Rules
ICAIS '02 Proceedings of the 2002 IEEE International Conference on Artificial Intelligence Systems (ICAIS'02)
Image Analysis and Mathematical Morphology
Image Analysis and Mathematical Morphology
Translation, scaling and rotation invariant spot matching using delaunay triangulation
ACS'08 Proceedings of the 8th conference on Applied computer scince
| Hi-index | 0.00 |
The proteomic analysis is a set of process involving modern analytical techniques: -Realization of 2D Electrophoresis-Gels (2DE-G). - Processing of 2DE-G image and location of new or modified proteins. - Peptidic identification by mass spectrometry. - Location in data bases of proteins whose peptidic card correlates with respect to that found. The 2DE-G image processing is a crucial stage because it allows to localize pertinently the protein spots candidate for a peptidic identification by mass spectrometry. One of the major axes of this stage turns around the protein-spots detection. The reliability of this stage depends among others on the quality of the detection tools. In this paper we present a segmentation technique combining both the Top-Hat transformation and the watershed. The segmentation process is an iterative process leading to a simple and effective detection of protein spots. Before this segmentation process, image must be preprocessed in order to improve the image quality in term of signal to noise ratio. Image preprocessing algorithm uses the sub-band technique combined with Retinex Method. The detection algorithm operates by finding the threshold gray level that minimizes the entropy of the fuzziness measure.