Inventory Item Recognition via Vision Tunnel and Exception Pathway

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Materials handling facilities face challenges in accurately receiving shipments into inventory due to a lack of identifying information, leading to issues with recognizing and correlating items with their corresponding inventory records, resulting in incomplete or lost items.

Innovation Solution

The implementation of a vision tunnel for automated examination and an exception pathway with an image capture device and item receive application that facilitates image matching with inventory data, coupled with a weigh station and inventory control system to identify and correct issues, ensures accurate item recognition and correlation.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If automated vision tunnel and image capture devices are implemented, then item recognition accuracy is improved, but device complexity increases

Engineering Contradiction:
Improveitem recognition accuracyVSAvoidsystem complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The system captures images of items at the exception pathway before final inventory receipt, performing preliminary identification and correlation. This preliminary action allows the system to resolve identification issues early in the process flow, improving accuracy without requiring complex real-time analysis at the final receipt point.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent introduces an intermediary exception pathway with image capture devices and processing systems that act as a mediator between the main receiving flow and the inventory system. This intermediary layer handles complex identification tasks separately, allowing the main receiving system to remain relatively simple while achieving high recognition accuracy through the intermediary processing step.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Reliability

If exception pathway with image capture is added, then inventory accuracy is improved, but processing time increases

Engineering Contradiction:
Improveinventory accuracyVSAvoidprocessing time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The receiving process is segmented into multiple pathways: a primary fast track for clearly identifiable items and an exception pathway for items requiring additional verification. This segmentation allows most items to be processed quickly through the primary path while only diverting problematic items to the exception pathway for detailed image capture and analysis, thereby maintaining high overall processing speed while improving inventory accuracy.

Inventive Principle:
Principle #1Segmentation

3Measurement precision

If manual verification processes are used, then item identification accuracy is improved, but productivity decreases

Engineering Contradiction:
Improveitem identification accuracyVSAvoidreceiving throughput
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent replaces manual verification processes with automated image capture devices, computer vision systems, and electronic database matching. This substitution eliminates the need for manual inspection while maintaining high identification accuracy, thereby preserving receiving throughput. The automated systems can process multiple items simultaneously and continuously without the limitations of human operators.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS9663294B2Receiving items into inventory
Publication Date: 2017.05.30 AMAZON TECH INC
  • US9663294B2 patent drawing
  • US9663294B2 patent drawing
  • US9663294B2 patent drawing

AI summary

The present disclosure involves the receipt of inventory items in a materials handling facility. In one embodiment, a listing of inventory items associated with an inventory is maintained in a memory accessible to a computing device. An attempt is made to identify a match between a first image of a receivable item and a second image of one of the inventory items. If a match is successfully identified, an identifier associated with the one of the inventory items is rendered, where the identifier is to be associated with the receivable item.