LicenseAuthors of papers retain copyright and release the work under a Creative Commons Attribution 4.0 International License (CC-BY).

No abstract

Abstract. Oscillatory networks represent biologically inspired models that implement cognitive functions such as vision, motion and memory. Vision functions are most attractive brain ability. Despite recognition problems have been successfully solved by traditional neural networks, segmentation problems still require close attention. In this paper, we propose oscillatory network based on Kuramoto phase oscillator for image segmentation where each allocated feature is encoded by ensemble of synchronous oscillators like in biologically plausible systems. The proposed model is designed to perform color segmentation and object segmentation using synchronization phenomena. Processes of synchronization between oscillators during image processing and multi-core implementation of the network for simulation are discussed. Experimental results of segmentation by the network using formal and real images have been presented.

No abstract

No abstract