Image Analysis System for Product Identification Accuracy

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Image-based product identification techniques face challenges in accuracy due to high computational volume, leading to increased processing time and potential errors in product recognition.

Innovation Solution

An image analysis system that includes a product identification result acquisition unit and a correction necessity assessment unit, which computes a matching degree between adjacent product image regions with differing identification results and assesses the need for correction based on this matching degree, thereby improving identification accuracy while reducing overall processing volume.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If image processing is performed on the entire image to improve product identification accuracy, then identification accuracy is improved, but processing time increases and exceeds acceptable range

Engineering Contradiction:
Improveproduct identification accuracyVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent divides the entire image into multiple local regions, each containing one or more products. Instead of processing the entire image uniformly, the system performs image processing on each local region separately. This segmentation allows the system to focus computational resources on specific product regions rather than the whole image, thereby reducing overall processing time while maintaining identification accuracy for each product.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies different processing strategies to different regions based on their characteristics. For regions where products are clearly distinguishable, standard identification is applied. For regions where identification confidence is low or products are closely positioned, enhanced processing is performed. This local quality approach ensures high accuracy for difficult cases while avoiding unnecessary processing for easy cases, thus reducing overall processing time.

Inventive Principle:
Principle #3Local quality

2Measurement precision

If image processing is performed on the entire image to improve product identification accuracy, then identification accuracy is improved, but computational volume increases

Engineering Contradiction:
Improveproduct identification accuracyVSAvoidcomputational volume
Core Design Contradiction:
Measurement precisionVSQuantity of substance

Solution Approach 1:

The patent segments the image into multiple local regions and processes each region independently. This segmentation reduces the computational volume by avoiding redundant processing across the entire image. Each local region is processed with appropriate computational resources, and results are aggregated to form the final identification, thereby reducing total computational requirements while maintaining accuracy.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent applies enhanced image processing only to local regions where it is necessary (e.g., regions with low identification confidence or closely positioned products), rather than applying full processing to the entire image. This partial action approach reduces overall computational volume by performing intensive processing only where needed, while using simpler processing for other regions.

Inventive Principle:
Principle #16Partial or excessive action

Data Source

PatentUS20240412505A1Image analysis system, image analysis method, and non-transitory computer-readable medium
Publication Date: 2024.12.12 NEC CORP
  • US20240412505A1 patent drawing
  • US20240412505A1 patent drawing
  • US20240412505A1 patent drawing

AI summary

An image analysis system (1) includes a product identification result acquisition unit (110) and a correction necessity assessment unit (120). The product identification result acquisition unit (110) acquires an identification result of each of a plurality of products captured in an image. The correction necessity assessment unit (120) computes a first matching degree being a matching degree between an image region of a first product and an image region of a second product when the identification result of the first product among the plurality of products differs from the identification result of the second product adjacent to the first product. Further, the correction necessity assessment unit (120) assesses correction necessity of the identification result of the first product or the identification result of the second product, based on the first matching degree.