You have requested a machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Neither SPIE nor the owners and publishers of the content make, and they explicitly disclaim, any express or implied representations or warranties of any kind, including, without limitation, representations and warranties as to the functionality of the translation feature or the accuracy or completeness of the translations.
Translations are not retained in our system. Your use of this feature and the translations is subject to all use restrictions contained in the Terms and Conditions of Use of the SPIE website.
30 September 1996Dynamic search of Gaussian segmentation
This paper presents a new approach, called Gaussian segmentation, to extract regions containing locally concentrated stimuli. We describe two kinds of searching methods, local search and dynamic search, to link local solutions of the estimating function in parameter space in the same and different visual fields, respectively. In the local search, we propose an iterative technique which can find a local solution from any initial state of parameters by means of searching the parametric space locally. The dynamic search is provided to move from one of the local search results as the initial state to the other local search results. Finally, the availability of the technique is shown by experiments for estimating using functional images and real images.