Method for preparing a digital list of food ingredients and method for identifying a list of recipes corresponding to a plurality of food ingredients
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing cooking assistance apps require users to manually correct and confirm multiple times to achieve accurate ingredient recognition, leading to a time-consuming and complicated process, especially when images are suboptimal, and do not efficiently provide recipes or cooking tips.
Innovation Solution
A method using assisted interaction with a user device to recognize ingredients through video stream analysis, aided by superimposed frames and machine learning algorithms, allowing user confirmation and correction, and linking to recipes and appliance commands.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If multiple sequential recognition operations are performed to improve accuracy, then ingredient recognition precision is improved, but user time consumption and operational complexity increase significantly
Solution Approach 1:
The system performs preliminary recognition operations automatically in the background before the user needs to review results. Multiple recognition passes are executed proactively, and results are prepared in advance, so when the user interacts with the system, the heavy computational work is already completed, reducing their time commitment while maintaining high precision through multiple analysis passes.
Solution Approach 2:
The system implements a feedback mechanism where recognition results are presented to the user for confirmation or correction. User corrections are fed back into the recognition system to refine future predictions. This allows the system to learn from user interactions and improve accuracy over time without requiring users to manually perform multiple recognition operations each time.
2Measurement precision
If multiple sequential recognition operations are performed to improve accuracy, then ingredient recognition precision is improved, but operational complexity increases
Solution Approach 1:
The system performs self-service by automatically executing multiple recognition operations without requiring user intervention at each step. The complex multi-stage recognition process is handled autonomously by the system, while the user only needs to provide initial input and final confirmation, dramatically simplifying the operational complexity despite using sophisticated multi-step recognition algorithms.
Solution Approach 2:
The system introduces an intermediary processing layer that manages the complexity of multiple recognition operations. This intermediary layer handles the coordination between different recognition passes, consolidates results, and presents simplified information to the user. The intermediary absorbs the operational complexity, shielding the user from the intricate multi-step process while maintaining high precision through comprehensive analysis.
3Ease of operation
If image recognition is performed on suboptimal images to maintain system usability, then ease of operation is improved, but recognition accuracy deteriorates
Solution Approach 1:
The system performs preliminary enhancement operations on uploaded images before recognition. Suboptimal images undergo automatic preprocessing steps such as brightness adjustment, contrast enhancement, and noise reduction. This preliminary action improves image quality without requiring users to manually optimize their photos, maintaining ease of operation while improving the foundation for accurate recognition.
Solution Approach 2:
The system implements feedback loops where recognition results are analyzed to identify potential errors caused by suboptimal image quality. When uncertainty is detected, the system can request additional images or provide feedback to users about specific image quality issues. This feedback mechanism allows the system to compensate for suboptimal inputs while maintaining usability, as users don't need to understand technical image quality parameters.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A method for preparing a digital list of food ingredients based on an assisted interaction with a user is described. The method comprises the steps of acquiring at least one digital video stream of one or more respective food ingredients by means of a user device, and interacting with the aforesaid at least one digital video stream, by the user by means of the user device, to select in the digital video stream at least one digital image of each of said one or more food ingredients which are desired to be included in the digital list of ingredients. The method then includes recognizing, through electronic processing means, a plurality of ingredients based on the at least one respective digital image selected by the user in the digital video stream; and preparing, through electronic processing means, a suggested digital list of ingredients, comprising the list of one or more recognized ingredients. The method then comprises the steps of providing the aforesaid suggested digital list of ingredients to the user, through a digital interface of the user device; and confirming and/or correcting, by the user, through said digital interface of the user device, the aforesaid suggested digital list of ingredients, to prepare a definitive digital list of ingredients. A method for identifying a list of recipes corresponding to a plurality of food ingredients, a method for controlling a cooking apparatus for cooking food, and a method for using/controlling a refrigeration apparatus are further described, which are based on the results of the aforesaid method for preparing a digital list of food ingredients.