Spiking Neuron Network for Visual Data Encoding

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing artificial retinal systems fail to effectively encode visual data into spike output using spiking neuron networks, particularly in reproducing spatial characteristics and achieving optimal configuration for different applications with efficient data compression.

Innovation Solution

A computerized spiking neuron signal processing apparatus is developed, comprising processors that execute modules to encode signals into spike outputs with distinct response durations, allowing for luminance and chromaticity encoding, and reconfiguring processors to adapt to different input filters and mappings for optimal visual data processing.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If artificial retinal systems use conventional encoding methods, then implementation is simpler, but encoding efficiency and spatial characteristic reproduction are insufficient

Engineering Contradiction:

Improveencoding efficiencyVSAvoidsystem complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent implements dynamic spiking neuron networks where neurons adapt their response durations based on input signal characteristics. The system transitions from static conventional encoding to dynamic spike-based encoding, allowing neurons to adjust their temporal response patterns to optimize encoding efficiency for different visual stimuli while maintaining biological plausibility

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system changes the fundamental encoding parameter from continuous analog signals to discrete spike trains with variable response durations. By adjusting the response duration parameter of spiking neurons to match natural retinal ganglion cell characteristics, the system achieves more efficient encoding of visual information while reproducing spatial characteristics more accurately

Inventive Principle:

Principle #35Parameter changes

2Loss of substance

If spiking neuron networks are implemented with distinct response durations, then data compression is improved, but computational complexity increases

Engineering Contradiction:

Improvedata compressionVSAvoidcomputational complexity

Core Design Contradiction:

Loss of substanceVSDevice complexity

Solution Approach 1:

The patent employs periodic spiking patterns where neurons fire action potentials at specific intervals rather than continuous signaling. This periodic action converts continuous visual information into discrete temporal codes, achieving data compression by encoding information in spike timing and frequency patterns while reducing the overall data transmission burden

Inventive Principle:

Principle #19Periodic action

Solution Approach 2:

The system replaces conventional mechanical or electronic signal processing with biologically-inspired spiking neuron computation. By substituting traditional computational approaches with spike-based temporal coding, the system achieves more efficient data representation and compression while mimicking natural neural processing mechanisms

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

3Adaptability or versatility

If processors are reconfigured for different input filters and mappings, then adaptability to various applications is improved, but device complexity increases

Engineering Contradiction:

Improveapplication adaptabilityVSAvoidprocessor configuration complexity

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent implements universal spiking neuron processors that can be configured to perform multiple encoding functions. The same processor architecture can be adapted to different input filters (e.g., luminance, chromaticity) and mapping schemes by reconfiguring connection weights and neuron parameters, enabling a single system to serve multiple visual processing applications without requiring separate dedicated hardware for each function

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS9311594B1Spiking neuron network apparatus and methods for encoding of sensory data

Publication Date: 2016.04.12 BRAIN CORP

AI summary

Sensory encoder may be implemented. Visual encoder apparatus may comprise spiking neuron network configured to receive photodetector input. Excitability of neurons may be adjusted and output spike may be generated based on the input. When neurons generate spiking response, spiking threshold may be dynamically adapted to produce desired output rate. The encoder may dynamically adapt its input range to match statistics of the input and to produce output spikes at an appropriate rate and/or latency. Adaptive input range adjustment and/or spiking threshold adjustment collaborate to enable recognition of features in sensory input of varying dynamic range.