Image processing apparatus, image processing method and recording medium
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing image processing techniques for recognizing commodities in store displays are influenced by environmental factors like lighting and camera angle, leading to higher failure rates in recognition due to inappropriate threshold settings, resulting in either false negatives or false positives.
Innovation Solution
An image processing apparatus that includes recognition and detection units, where the detection unit identifies regions in a captured image that are likely to have failed recognition based on store fixture information, allowing for higher precision in detecting unrecognized products and reducing false recognition.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a higher recognition threshold is set to prevent false recognition, then false recognition rate is reduced, but recognition failure rate increases
Solution Approach 1:
The patent divides the recognition evaluation into two separate dimensions: recognition threshold (for preventing false recognition) and recognition failure detection (for identifying missed products). By segmenting the problem into false recognition prevention and recognition failure detection, the system can independently optimize each aspect without trade-offs.
Solution Approach 2:
The patent introduces store fixture information as an intermediary element that mediates between the recognition result and the final evaluation. This intermediary provides contextual information about expected product placements, enabling the system to detect recognition failures without lowering the recognition threshold.
2Measurement precision
If a lower recognition threshold is set to prevent recognition failure, then recognition failure rate is reduced, but false recognition rate increases
Solution Approach 1:
The patent separates the functions of recognition threshold setting and failure detection into independent processes. The recognition threshold can be set high for reliability, while recognition failures are detected through comparison with store fixture information, eliminating the need to lower the threshold.
Solution Approach 2:
The system uses store fixture information as feedback to evaluate whether products are properly recognized in their expected locations. This feedback mechanism allows the system to identify recognition failures without affecting the recognition threshold setting.
3Device complexity
If environmental factors like lighting and camera angle are not considered, then system complexity is reduced, but recognition accuracy deteriorates
Solution Approach 1:
The patent introduces store fixture information as an intermediary that compensates for environmental variations without requiring complex environmental sensing or processing. This intermediary provides a reference framework that remains stable despite changes in lighting, camera angle, or other environmental factors.
Solution Approach 2:
The system changes the parameter being used for evaluation from direct image analysis to comparison with store fixture information. This parameter change makes the recognition system more robust to environmental variations while maintaining simplicity.
Data Source
AI summary
An image processing apparatus includes: a recognition unit configured to recognize products from a captured image obtained by capturing an image of displayed products; and a detection unit configured to detect, based on store fixture information related to a store fixture in which the products are displayed, a region of a product that is included in the captured image but is not recognized by the recognition unit.