Layered Network Calculation Scheme Selection for Recognition Devices

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Recognition devices, such as mobile terminals and cameras, face increased processing time and resource requirements due to enhanced image sensors and communication data capacity, necessitating a calculation scheme decision system that optimizes calculation costs for network structures and weight data processing.

Innovation Solution

A calculation scheme decision system that includes a pre-calculation unit, cost acquisition unit, and decision unit to select optimal calculation schemes for each layer of a network structure based on predetermined data, considering calculation costs, and performs calculations using these selected schemes to optimize processing efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If image sensor performance is enhanced and communication data capacity is increased, then recognition accuracy and data processing capability are improved, but processing time and resource requirements increase

Engineering Contradiction:
Improverecognition accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent segments the network structure into multiple layers and applies different calculation schemes to different layers. This segmentation allows the system to optimize processing time for each layer independently, reducing overall processing time while maintaining recognition accuracy.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent dynamically selects calculation schemes based on the specific layer and data characteristics. By making the calculation scheme selection dynamic rather than static, the system can adapt to different processing requirements at different layers, reducing processing time without sacrificing accuracy.

Inventive Principle:
Principle #15Dynamics

2Measurement precision

If image sensor performance is enhanced and communication data capacity is increased, then recognition accuracy and data processing capability are improved, but resource requirements increase

Engineering Contradiction:
Improverecognition accuracyVSAvoidresource requirements
Core Design Contradiction:
Measurement precisionVSUse of energy by moving object

Solution Approach 1:

The patent divides the network into layers and assigns different calculation schemes to different layers based on their specific requirements. This segmentation enables resource optimization by applying computationally intensive schemes only where necessary, reducing overall resource consumption while maintaining recognition accuracy.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies different calculation schemes to different layers based on their local characteristics and requirements. This local quality approach ensures that resources are allocated efficiently to each layer, using more resources only where needed for accuracy while conserving resources in layers where simpler schemes suffice.

Inventive Principle:
Principle #3Local quality

3Productivity

If multiple calculation schemes are prepared and evaluated for each layer, then optimal calculation scheme selection is achieved, but system complexity increases

Engineering Contradiction:
Improveprocessing efficiencyVSAvoidsystem complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The patent performs pre-calculation to evaluate multiple calculation schemes before actual processing. This preliminary action identifies the optimal scheme for each layer in advance, allowing the system to achieve high processing efficiency during runtime without the complexity of dynamic decision-making, as the optimal paths are predetermined.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3614312B1System, method and processing program for determining a calculation technique
Publication Date: 2024.09.18 MORPHO INC
  • EP3614312B1 patent drawingFigure 1
  • EP3614312B1 patent drawingFigure 2
  • EP3614312B1 patent drawingFigure 3

AI summary

A calculation scheme decision system includes a pre-calculation unit performing, in an execution environment in which calculation is performed, calculation for each of respective layers of the network structure using at least one of calculation schemes prepared in advance for the respective layers, a cost acquisition unit acquiring a calculation cost of at least one calculation scheme for each layer based on a result of the calculation by the pre-calculation unit, a decision unit selecting one calculation scheme for each layer based on the calculation cost from among at least one of the calculation schemes prepared in advance for the respective layers to associate the layer with the selected one calculation scheme, and a calculation unit performing the calculation for each of the respective layers of the network structure on input data in the execution environment using the calculation scheme associated with each layer.