Display Rack Image Processing for Accurate State Evaluation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing technologies fail to accurately evaluate the state of display racks in retail environments, particularly in identifying changes due to customer interactions or inventory replenishment, which can lead to sales opportunities being missed.

Innovation Solution

An image processing device that detects changes in a display rack, classifies these changes, and evaluates the display state of goods based on captured images, using a combination of detection, classification, and evaluation units to identify types of changes such as goods being taken or replenished.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Difficulty of detecting and measuring

If background subtraction method is used to detect change areas, then detection capability is improved, but classification accuracy of change types deteriorates

Engineering Contradiction:
Improvedetection capabilityVSAvoidclassification accuracy
Core Design Contradiction:
Difficulty of detecting and measuringVSMeasurement precision

Solution Approach 1:

The patent segments the change detection process into two independent stages: first detecting change areas using background subtraction, then classifying change types using image recognition. This segmentation allows each stage to specialize in its function, with detection handling the 'what changed' and classification handling the 'how it changed', thereby resolving the contradiction between detection capability and classification accuracy.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent introduces an intermediary classification unit that receives change area information from the detection unit and provides classified change type information to the evaluation unit. This intermediary layer bridges the gap between simple detection and accurate classification, enabling the system to first identify where changes occurred and then determine the nature of those changes with high accuracy.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Device complexity

If evaluation is based on disturbed display indicators only, then evaluation simplicity is maintained, but evaluation accuracy deteriorates

Engineering Contradiction:
Improveevaluation simplicityVSAvoidevaluation accuracy
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent merges multiple evaluation approaches into a unified system: it combines traditional disturbed display indicators with change type classification results. The evaluation unit integrates information from both sources, using classification results to enhance the interpretation of disturbed display indicators, thereby improving evaluation accuracy without significantly increasing overall system complexity.

Inventive Principle:
Principle #5Merging (Combining)

Solution Approach 2:

The patent implements feedback by using classification results to refine the evaluation of disturbed display indicators. The system continuously refines its evaluation by feeding back the classified change type information to adjust the interpretation of display disturbances, creating an iterative evaluation process that improves accuracy over time.

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS11250570B2Display rack image processing device, image processing method, and recording medium
Publication Date: 2022.02.15 NEC CORP
  • US11250570B2 patent drawing
  • US11250570B2 patent drawing
  • US11250570B2 patent drawing

AI summary

A state of a display rack is evaluated accurately. An image processing device includes a detection unit configured to detect a change area related to a display rack from a captured image in which an image of the display rack is captured, a classification unit configured to classify a change related to the display rack in the change area, and an evaluation unit configured to evaluate a display state of goods, based on a classification result.