Thresholding Using Relaxation
Azriel Rosenfeld, Russell C. Smith · IEEE Transactions on Pattern Analysis and Machine Intelligence · 1981
If a picture contains dark objects on a light background (or vice versa), the objects can be extracted by thresholding, i.e., by classifying the pixels into ``light'' and ``dark'' classes. If the picture is noisy, so that the object and background gray level populations overlap, there will be errors in the thresholded output. A relaxation process can be used to reduce these errors; we classify the pixels probabilistically, and then adjust the probabilities for each pixel, based on its neighbors' probabilities, with light reinforcing light and dark dark. When this adjustment process is iterated, the dark probabilities become very high for pixels that belong to dark regions, and vice versa, so that thresholding becomes trivial.
Cited by 75 works. doi:10.1109/tpami.1981.4767152
Authors
References
- Digital Picture Processing
- Scene Labeling by Relaxation Operations
- Programming pattern recognition
- A survey of threshold selection techniques
- Noise Cleaning by Iterated Local Averaging
- Image segmentation by clustering
- A New Probabilistic Relaxation Scheme
- Determining Compatibility Coefficients for Curve Enhancement Relaxation Processes
- An Application of Relaxation Methods to Edge Reinforcement
- A Relaxation Method for Multispectral Pixel Classification