Object Map Comparison for Discrimination Error Detection

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing systems fail to detect discrimination errors in object types due to poor learning, as multiple redundant inference units may output the same incorrect result, masking errors.

Innovation Solution

A calculation system comprising a first and second device, where each device generates an object map from image information and compares the maps to detect type and position of objects, using a majority decision to determine error correctness.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If multiple redundant inference units are used, then fault detection capability is improved, but discrimination errors due to poor learning cannot be detected

Engineering Contradiction:

Improvefault detection capabilityVSAvoiddiscrimination error detection

Core Design Contradiction:

ReliabilityVSMeasurement precision

Solution Approach 1:

The patent divides the inference system into multiple independent inference units that process the same input independently. By segmenting the inference function across multiple units and comparing their outputs, the system can detect both faults (when one unit fails) and discrimination errors (when all units agree on wrong results), resolving the contradiction between fault detection capability and discrimination error detection.

Inventive Principle:

Principle #1Segmentation

2Reliability

If multiple redundant inference units are used, then system reliability is improved, but all units may output the same incorrect result due to poor learning

Engineering Contradiction:

Improvesystem reliabilityVSAvoidinference error masking

Core Design Contradiction:

ReliabilityVSLoss of information

Solution Approach 1:

The patent implements a feedback mechanism where the output of multiple inference units is compared against each other and against reference data. This feedback loop allows the system to identify when all units produce the same incorrect result due to poor learning, preventing error masking while maintaining system reliability through redundant processing.

Inventive Principle:

Principle #23Feedback

3Measurement precision

If object maps from multiple devices are compared, then discrimination error detection is improved, but system complexity increases

Engineering Contradiction:

Improvediscrimination error detectionVSAvoidsystem complexity

Core Design Contradiction:

Measurement precisionVSDevice complexity

Solution Approach 1:

The patent merges the object map generation functions of multiple devices into a unified comparison process. By combining the outputs of multiple devices and performing centralized comparison, the system achieves improved discrimination error detection while managing complexity through integrated processing rather than separate complex systems.

Inventive Principle:

Principle #5Merging (Combining)

Data Source

PatentUS11367218B2Calculation system and calculation method

Publication Date: 2022.06.21 HITACHI LTD

AI summary

To detect a discrimination error in a type of an object. A calculation system includes a first device and a second device. The first device includes: a first object map generation unit configured to calculate, using first image information that is image information acquired by the first device, a first object map indicating a type of an object and a position of the object; and a first communication unit configured to transmit the first object map to the second device. The second device includes: a second object map generation unit configured to calculate, using second image information that is image information acquired by the second device, a second object map indicating a type of an object and a position of the object; and a comparison unit configured to compare the first object map and the second object map.