A Computational Approach to Edge Detection
IEEE Transactions on Pattern Analysis and Machine Intelligence
Speckle Noise MAP Filtering Based on Local Adaptive Neighborhood Statistics
SIBGRAPI '99 Proceedings of the XII Brazilian Symposium on Computer Graphics and Image Processing
Hi-index | 0.00 |
Two individual Neuro-Fuzzy systems are used to enhance sonographical images. This system reduces noise and detects edge in sonographical images sequentially. Noise reduction and edge detection are two different Genetic-Neuro-Fuzzy systems. Furthermore the noise reduction and edge detection are opposite algorithms, so these operations can't be in one system. Although noise reduction is an edge preserving and edge detection is noise free algorithm. Noise reduction is a low pass filter and edge detection is a high pass filter, So this system is composite of two independent serial systems. Noise reduction filter is 4 layers ANFF network and edge detection system is competitive network. Results of this system are compared with combination of weighted adaptive median filter and Canny edge detector. Results show that the performance of this system is the better than median-Canny system.