Smart Tote Camera System for Automated Item Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current materials handling systems require manual scanning of items at checkout, which is inefficient and can be avoided by using a tote equipped with cameras and classifiers to automatically identify items and generate a list for checkout.
Innovation Solution
A tote with integrated cameras and classifiers that analyze video data to identify item identifiers, such as barcodes, and generate a virtual item listing, allowing users to skip traditional checkout processes by automatically updating the list based on item placement and removal.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If manual scanning at checkout is used, then item identification is performed, but checkout efficiency is reduced and time is lost
Solution Approach 1:
The system performs item identification in advance during the shopping process by capturing images with the tote camera and processing them through classifiers to generate a preliminary item list. This preliminary action eliminates the need for manual scanning at checkout, as the identification work is completed beforehand while the customer is still shopping.
Solution Approach 2:
The tote system provides self-service item identification by automatically capturing images, processing them through classification algorithms, and maintaining an updated item list without requiring customer action. The system serves itself by continuously monitoring the tote contents and autonomously generating the checkout list.
2Ease of operation
If automated image recognition is implemented, then manual scanning is eliminated, but system complexity increases
Solution Approach 1:
The tote system serves multiple functions: it acts as a shopping container, an automated camera system, an image processing unit, and a checkout management device. By combining these functions into a single universal system, the patent reduces the need for separate manual operations while managing complexity through integration rather than adding separate components.
Solution Approach 2:
The system replaces manual mechanical scanning operations with automated optical image capture and digital classification. The physical act of manually presenting items to scanners is substituted by automated camera-based image recognition and computational classification, eliminating the need for mechanical interaction at checkout.
3Measurement precision
If multiple images are captured and processed, then identification accuracy is improved, but processing time increases
Solution Approach 1:
The system captures and processes multiple images during the shopping period rather than at checkout. By performing this processing in advance, the system accumulates identification data over time, improving accuracy through multiple samples while distributing the processing time across the shopping duration rather than concentrating it at checkout.
Solution Approach 2:
The image capture and processing operates continuously throughout the shopping experience rather than as a discrete batch operation. The camera continuously monitors the tote, and images are processed as they are captured, maintaining continuous useful action that improves identification accuracy through ongoing data collection without creating time delays at checkout.
Data Source
AI summary
This disclosure describes, in part, techniques for collecting image data representing item identifiers, such as barcodes. For instance, system(s) may receive image data representing images, where the images depict at least a portion of an identifier located on an item. The system(s) may then identify a first portion of the image data representing an image that that is associated a low confidence level. Next, the system(s) may identify a second portion of the image data representing additional images that are associated high confidence levels. Using results for this the second portion of the image data, the system(s) may determine a ground truth result for the first portion of the image data. The system(s) may then store, in one or more databases, data representing the ground truth result in association with the first portion of the image data.