Details

Autor(en) / Beteiligte

• Gu, Xiaodong
• Zhang, Liming
• Yi, Zhang
• Liao, Xiao-Feng
• Wang, Jun

Titel

Global Icons and Local Icons of Images Based Unit-Linking PCNN and Their Application to Robot Navigation

Ist Teil von

• Advances in Neural Networks – ISNN 2005, 2005, p.836-841

Ort / Verlag

Berlin, Heidelberg: Springer Berlin Heidelberg

Erscheinungsjahr

2005

Quelle

Alma/SFX Local Collection

Beschreibungen/Notizen

• PCNN-Pulse Coupled Neural Network is a new artificial neural network supported by biological experimental results. Unit-linking PCNN we have developed from PCNN can be efficiently used for image processing. Using Unit-linking PCNN can produce icons of images, namely time signatures of images. This paper describes how to use Unit-linking PCNN to produce global icons, and local icons of images. A global icon of an image based Unit-linking PCNN has the translation, rotation invariance. In some situations, such as object detection, this invariance is an advantage. However, in other situations, such as navigation, Unit-linking PCNN icons should reflect the local changes of scenes and here the invariance of the global icon is not the advantage. Therefore we also introduce the local icons of images based Unit-linking PCNN, which can reflect the local changes of images and have no translation, rotation invariance that global icons have. In the meantime, the robot navigation based global and local icons is shown as an example.