Oven Cloud-Based Food Recognition to Reduce Appliance Complexity

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing cooking systems face challenges in providing user-friendly and cost-effective food recognition due to the need for sophisticated computation and data storage, which is often not feasible in household appliances, resulting in unsatisfactory quality and speed of food recognition.

Innovation Solution

A cooking system that employs basic picture recognition within the oven to determine if food is present, with detailed food recognition performed externally by a cloud-based server system, utilizing adaptive self-learning systems like neural networks to achieve accurate and efficient results at lower costs.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If sophisticated food recognition is implemented in the household appliance, then food recognition quality and speed are improved, but device complexity and costs increase

Engineering Contradiction:
Improvefood recognition qualityVSAvoidcomputation and data storage requirements
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent extracts the complex food recognition computation and data storage functions from the household appliance and relocates them to an external server. The appliance only retains basic image capture and transmission capabilities, while the external server performs sophisticated food identification, parameter determination, and recipe selection. This separation resolves the contradiction by maintaining high recognition quality through external sophisticated processing while keeping the appliance itself simple and cost-effective.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent introduces an external server as an intermediary between the household appliance and the food recognition process. The server acts as a mediator that receives basic picture data from the appliance, performs sophisticated analysis using its computational resources and database, and returns cooking parameters to the appliance. This intermediary enables high-quality recognition without requiring the appliance to have complex hardware or software.

Inventive Principle:
Principle #24Intermediary (Mediator)

2Device complexity

If basic picture recognition is used in the oven, then device complexity and costs are reduced, but food recognition accuracy decreases

Engineering Contradiction:
Improvehardware and software simplicityVSAvoidfood identification accuracy
Core Design Contradiction:
Device complexityVSMeasurement precision

Solution Approach 1:

The patent segments the food recognition process into two distinct stages: basic picture recognition performed by the oven to determine food presence and trigger the process, and detailed food identification performed by the external server. This segmentation allows the oven to maintain simplicity while the server handles the computationally intensive detailed analysis, thereby achieving both low device complexity and high recognition accuracy.

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

The patent implements preliminary basic picture recognition in the oven that identifies food presence and triggers the detailed recognition process. This preliminary action by the simple oven components initiates the workflow, after which the external server performs the sophisticated detailed analysis. The preliminary basic recognition suffices to activate the system without requiring the oven to perform complex analysis itself.

Inventive Principle:
Principle #10Preliminary action

3Measurement precision

If detailed picture recognition is performed in the oven, then food parameter identification is improved, but processing time increases

Engineering Contradiction:
Improvefood parameter identificationVSAvoidrecognition processing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent extracts the time-consuming detailed picture recognition and food parameter identification processes from the oven and relocates them to the external server. The oven only performs rapid basic detection of food presence, immediately triggers the external server for detailed analysis, and receives results. This extraction eliminates the time penalty from detailed recognition while maintaining accuracy, as the server's powerful computational resources process the analysis much faster than embedded appliance processors could.

Inventive Principle:
Principle #2Taking out (Extraction)

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

This approach allows for accurate and fast food recognition with reduced hardware and software complexity, enabling more precise cooking parameter selection and user-friendly operation while maintaining a good cost-benefit ratio.

Implementation Method 1

The oven is provided with an optical sensor for obtaining picture data of the oven cavity

Methodology Applied
Scientific EffectLight reflection: Reflection

Data Source

PatentEP3608593B1Cooking system comprising an oven and an external computing means, and method of operating such system
Publication Date: 2022.03.16 ELECTROLUX APPLIANCES
  • EP3608593B1 patent drawingFigure 1
  • EP3608593B1 patent drawingFigure 2
  • EP3608593B1 patent drawingFigure 3

AI summary

A cooking system, comprises an oven (10) having an oven cavity (14) that is accessible via an oven door (16), and an external computing means (32). The oven (10) is provided with at least one optical sensor (28) for obtaining picture data of the oven cavity (14), means for detecting door closing and triggering the optical sensor to obtain picture data, processing means for performing a basic picture evaluation to determine that the oven cavity (14) no longer is empty, and communication means for providing picture data to the external computing means (32). The external computing means (32) is adapted to perform a food recognition routine to determine food parameters based on the picture data provided by the communication means, select cooking parameters based on the determined food parameters, and transmit the selected cooking parameters to the oven (10).