Demo of my Melody and Drawing recognition program using MultiLevelPopulationCoding

Spatiotemporal-Sample Transform(SST) is a transform for any signals from Spatiotemporal domain to Sample domain. The computational model of this transform is based on multi-level population coding network, which is a hypothesis of how human brain interpert the external information. With certain co-effitient and connection, multi-level population coding(MLPC) of Discrete SST from audio signal to frequent domain became to Fast discrete Fourier Transform. We use a nature method to build MLPC network, with any input signal, MLPC network of SST can give the most exact estimate of signal domain distribution. SST with other input, such as tone, paint-ing, video and passages, can successfully transform these signal to their Sample domain and finally recognize and understand them.

Category:

Science & Technology

Tags:

• melody
• recognition
• drawing
• Spatiotemporal-Sample
• Transform
• Digtial
• Signal
• Processing
• Pattern
• Recognition
• Neuro
• Network
• On
• Intelligence

[suxpert]

Up...