The Effect of Median Filtering on Edge Estimation and Detection
IEEE Transactions on Pattern Analysis and Machine Intelligence
Pattern recognition using neural networks: theory and algorithms for engineers and scientists
Pattern recognition using neural networks: theory and algorithms for engineers and scientists
Nonlinear rule-based convolution for refocusing
Real-Time Imaging
Shape Analysis and Classification: Theory and Practice
Shape Analysis and Classification: Theory and Practice
On-line evolving image classifiers and their application to surface inspection
Image and Vision Computing
| Hi-index | 0.00 |
One of the most important steps in digitalmammography is an adequate segmentation of possibleabnormalities. This obviously minimizes errors in further stagessuch as in classification. However, several factors affect theproper segmentation of mammograms. Mammograms containlow signal to noise ratio (low contrast) and a complicatedstructured background.In this article we are describing ageneric approach for detecting patterns of architecturaldistortions in mammograms that is both complete anduncommitted to any type of training. Our detection algorithmdynamically updates the pixels intensities by following theirneighboring transition zone. Such approach proved to beeffective for detecting the edges of all types of breastabnormalities including the Stellate.