Multi-Image Food Sorting for Duplicate Defect Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing high-speed, mass-flow food sorting devices often make inaccurate sorting decisions due to multiple sensors identifying the same defect in an object, leading to incorrect ejections and inaccurate statistics, which can result in financial losses.
Innovation Solution
The method involves generating multiple images with different perspectives of objects, classifying features as parent or child, fusing complementary regions, and forming aggregated region representations to enhance sorting accuracy by synchronizing detection device operations in real-time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple sensors are used to detect defects in objects, then detection coverage is improved, but misidentification and erroneous quantification of defects occur
Solution Approach 1:
The patent segments the detection process by classifying features as either 'parent features' (primary defects) or 'child features' (secondary characteristics). This segmentation allows the system to distinguish between actual defects and secondary effects, preventing misidentification when multiple sensors detect the same object from different perspectives.
Solution Approach 2:
The patent merges data from multiple detection devices by fusing complementary regions and creating aggregated region representations. This combining process integrates information from multiple sensors while eliminating duplicate defect identifications, resolving the contradiction between improved detection coverage and reduced misidentification.
2Productivity
If multiple detection devices operate simultaneously, then real-time detection capability is improved, but interference between detection devices occurs
Solution Approach 1:
The patent implements periodic action by selectively actuating detection devices in a predetermined sequence rather than simultaneously. Each detection device operates at specific time intervals, allowing the system to maintain real-time detection capability while preventing interference between devices. The controller coordinates the periodic operation of multiple sensors to capture object features without mutual interference.
3Measurement precision
If comprehensive object features are collected, then sorting accuracy is improved, but processing time increases
Solution Approach 1:
The patent applies preliminary action by pre-classifying detected features as parent or child features during the detection phase. This preliminary classification organizes the data structure in advance, allowing the controller to quickly process and evaluate only the most relevant parent features when making sorting decisions, thereby reducing processing time while maintaining comprehensive feature collection.
Solution Approach 2:
The patent extracts and focuses on parent features as the primary basis for sorting decisions, separating them from child features that provide supplementary information. By extracting only the essential parent features for final decision-making, the system maintains high sorting accuracy while significantly reducing the processing time required to evaluate all collected object features.
Data Source
AI summary
A method and apparatus for sorting is described, and which includes providing a product stream formed of individual objects of interest having feature aspects which can be detected; generating multiple images of each of the respective objects of interest; classifying the feature aspects of the objects of interest; identifying complementary images by analyzing some of the multiplicity of images; fusing the complementary images to form an aggregated region representation of the complementary images; and sorting the respective objects of interest based at least in part upon the aggregated region representation which is formed.