Yao, Chang; Chen, Houjin; Li, Jupeng Analysis of dynamic behaviors of an improved pulse coupled neural network in image processing. (Chinese. English summary) Zbl 1174.68766 Acta Autom. Sin. 34, No. 10, 1291-1297 (2008). Summary: In order to further exploit the application depth of Pulse Coupled Neural Network (PCNN) in image processing, combined with network parameters, dynamic behaviors of an improved PCNN are analyzed theoretically and deduced by formula in detail. The influence of parameters’ change on dynamic behaviors is proved, and the rule of parameters determination is proposed under the condition that each neuron only fires once. Finally, the experiments implemented on the practical image segmentation indicate that the relationship between dynamic behaviors and parameters determination is accurate and the algorithm is effective. MSC: 68U10 Computing methodologies for image processing 94A08 Image processing (compression, reconstruction, etc.) in information and communication theory 92B20 Neural networks for/in biological studies, artificial life and related topics 68T05 Learning and adaptive systems in artificial intelligence Keywords:pulse coupled neural network; dynamic behaviors; image processing; neuron PDFBibTeX XMLCite \textit{C. Yao} et al., Acta Autom. Sin. 34, No. 10, 1291--1297 (2008; Zbl 1174.68766) Full Text: DOI