Cooking phase identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Consumers face challenges in determining the doneness of food items during cooking due to variations in cooking processes and environments, with existing methods being cumbersome or inaccurate, especially for cooking methods like air frying and water bathing.
Innovation Solution
A cooking method that utilizes image processing to identify the phase of a cooking process by comparing the color of a food item at different times, using a region of interest to derive pixel intensity data and apply functions to determine the phase, which can be implemented by a cooking apparatus with a camera and controller.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If computer vision technology is used for food quality detection in commercial settings, then food quality consistency can be maintained, but the technology cannot be applied during the cooking process itself due to complexity and resource requirements
Solution Approach 1:
The patent segments the cooking process into distinct phases (e.g., raw, cooking, cooked, overcooked) and identifies them separately through color analysis. The food item is divided into a region of interest that is monitored specifically, rather than analyzing the entire image, which reduces computational complexity while maintaining identification accuracy.
Solution Approach 2:
The patent extracts only the essential feature (color) from the food item images for phase identification, rather than performing comprehensive computer vision analysis. By taking out just the color information from the region of interest and comparing it across time points, the system achieves accurate phase detection with minimal computational resources.
2Measurement precision
If comprehensive computer vision analysis is performed to identify cooking phase, then accurate phase identification can be achieved, but significant computational resources and cloud-based systems are required
Solution Approach 1:
The patent extracts only the essential feature (color) from the food item images for phase identification, rather than performing comprehensive computer vision analysis. By taking out just the color information from the region of interest and comparing it across time points, the system achieves accurate phase detection with minimal computational resources.
Solution Approach 2:
The patent uses simple, lightweight image processing techniques that can be executed locally on the cooking device rather than relying on expensive cloud-based computer vision systems. The approach uses basic color comparison algorithms that consume minimal energy and can run on embedded processors in consumer appliances.
3Measurement precision
If weight change detection is used to determine doneness, then cooking phase can be identified for some food types, but the method is incompatible with water bathing and air frying processes
Solution Approach 1:
The patent implements a universal color-based detection method that works across multiple cooking methods including air frying, water bathing, and conventional heating. By monitoring color changes in the food item, the system adapts to different cooking processes without requiring method-specific calibration, making the doneness detection versatile and broadly applicable.
Solution Approach 2:
The patent directly utilizes color changes in the food item as the primary indicator of cooking phase. Different food types and cooking methods produce characteristic color transformations (e.g., browning, fading, darkening) that can be detected and interpreted to determine doneness, providing a universal approach that works regardless of the specific cooking technique used.
4Ease of operation
If consumers manually check food by inserting a probe 10 minutes before recipe time, then doneness can be assessed, but the process is cumbersome and may result in undercooking or overcooking by more than 10 minutes
Solution Approach 1:
The patent implements continuous monitoring of the food item throughout the entire cooking process by capturing images at multiple time points and continuously analyzing color changes. This eliminates the need for discrete manual checking at fixed intervals, providing uninterrupted observation that accurately tracks the cooking progression and determines the precise moment when doneness is achieved.
Solution Approach 2:
The patent provides real-time feedback to the consumer about the cooking phase by analyzing color changes and communicating the current state (e.g., raw, cooking, cooked, overcooked). This continuous feedback loop allows consumers to know exactly when to stop cooking without manual probing, improving both convenience and timing accuracy.
Data Source
AI summary
A cooking method includes receiving first image data and second image data corresponding to a view of a food item at a first time and at a second time of a cooking process, respectively. A perimeter of the food item visible in the view encloses a first area of the food item. A region of interest in the view that maps to a second area of the food item is selected, which is less than the first area. A phase of the cooking process is identified based on a comparison of a first color of the food item at the first time with a second color of the food item at the second time. The first and second colors are derived from a part of the first image data and a part of the second image data, respectively, that correspond to the region of interest.