Smart Cart Image Highlighting for Accurate Item Localization
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Traditional systems for object localization in environments rely on isolated device data, limiting their ability to recognize and situate objects without network connectivity, leading to incomplete assessments and misinterpretations due to limited image data capture.
Innovation Solution
A system that utilizes smart shopping carts equipped with cameras and sensors to capture and process images, applies machine-learning models to identify items, and communicates with a remote system to modify images to highlight target items on display screens as users approach their locations.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional systems use isolated device data for object localization, then device complexity is reduced and operation is simpler, but measurement precision and reliability of object location deteriorate due to limited image data capture
Solution Approach 1:
The patent merges data from multiple smart shopping carts and devices to achieve accurate object localization. By aggregating sensor data, image data, and location information from multiple sources, the system overcomes the limitations of single-device analysis and improves measurement precision without requiring each individual device to be overly complex
Solution Approach 2:
The patent introduces a centralized system as an intermediary that collects and processes data from multiple smart shopping carts. This intermediary aggregates sensor data, image data, and location information to determine object locations accurately, while the individual carts remain relatively simple devices that only need to capture and transmit data
2Loss of information
If devices capture limited image data based on their movement path, then device complexity and energy consumption are reduced, but loss of information occurs regarding objects not visually captured
Solution Approach 1:
The patent combines image data from multiple devices to achieve complete object recognition. Even though individual devices only capture data in their movement path, the aggregation of data from multiple devices provides comprehensive coverage of the environment, eliminating information loss about objects not captured by any single device
Solution Approach 2:
The system performs preliminary actions by capturing image data from multiple devices as they move through the environment, building up a comprehensive database before object localization is needed. This preliminary data collection ensures that when object search is required, complete information is already available from the aggregated data
3Measurement precision
If systems process sensor data device-by-device, then processing resources are reduced and operation is simpler, but measurement precision of environmental understanding deteriorates due to lack of cross-device context
Solution Approach 1:
The patent merges sensor data from multiple devices to achieve accurate environmental understanding. By aggregating and cross-referencing data from different smart shopping carts, the system gains comprehensive contextual information about the environment that cannot be obtained from single-device analysis, improving measurement precision while distributing the processing workload
Data Source
AI summary
A system may store a plurality of images depicting items within an environment, where each image was captured by a camera coupled to a shopping cart. The system identifies a target item associated with a user device that is located within the environment. The system identifies a location of the target item within the environment based on item data associated with the target item and environment map data describing the environment. The system selects, from the plurality of images, an image depicting the item at the location within the environment based on the environment map data and the location data associated with each of the plurality of images. The system identifies a portion of the identified image that depicts the item by applying a machine-learning model to the identified image. The system modifies the identified portion of the identified image to highlight the target item.