Dual Path Image Analysis for Autonomous Cooking Time Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing cooking devices lack intelligence and adaptability, unable to learn new recipes or adjust to real-time conditions during food preparation, and they are not networked for sharing or selling recipe files.
Innovation Solution
An autonomous cooking device that uses image processing and machine learning to compare captured images with reference images through dual path analysis, combining color histogram and one-dimensional power spectral density comparisons, to determine the remaining cooking time and adapt the cooking process accordingly, and is networked for sharing recipe files.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If basic automation is used in cooking devices, then ease of operation is improved, but adaptability and intelligence are worsened
Solution Approach 1:
The system continuously captures images of the food during cooking and compares them against reference images using image processing algorithms. This real-time feedback loop allows the cooking device to monitor cooking progress and automatically adjust cooking parameters (time, temperature, power level) to achieve the desired cooking outcome, thereby providing both automation and adaptability
Solution Approach 2:
The patent replaces traditional mechanical timers and temperature controls with an intelligent image-based control system. Instead of relying on pre-programmed fixed-time cooking cycles, the system uses computer vision to detect food state changes and dynamically controls cooking parameters, substituting mechanical automation with intelligent adaptive control
2Ease of operation
If fixed cooking programs are used, then ease of operation is improved, but manufacturing precision of cooking outcome is worsened
Solution Approach 1:
The system uses real-time image capture and comparison during cooking to provide feedback on actual cooking progress. By continuously comparing current food appearance against reference images of properly cooked food, the system can detect deviations and adjust cooking parameters dynamically, ensuring consistent cooking outcomes regardless of variations in food quantity, type, or initial state
Solution Approach 2:
The cooking control system transitions from static pre-programmed fixed-time cycles to dynamic adaptive control. The cooking time and power level are continuously adjusted based on real-time image analysis of food state changes, allowing the system to adapt to actual cooking conditions and achieve precise, consistent results for different food types and quantities
3Measurement precision
If dual path image analysis is used, then measurement precision is improved, but device complexity is worsened
Solution Approach 1:
The image analysis system is divided into two separate independent analysis paths: color histogram analysis and power spectral density analysis. Each path processes image data independently using different algorithms suited to specific features (color distribution for one path, texture/frequency patterns for the other). The results from both paths are then combined to produce the final cooking control decision, improving accuracy while keeping individual modules relatively simple
Data Source
AI summary
Image analysis for comparison utilizes a combined image distance determined between the one or more captured images and the one or more reference images from a dual path analysis. The dual path analysis can comprise a first image analysis based on histograms of image colors and a second image analysis based a one-dimensional power spectral density of a black and white instance of the one or more captured images and the one or more reference images. In one case, the one-dimensional power spectral density derived by integrating a Discrete Fourier Transform of a two-dimensional image over annular rings of substantially equal area starting from the center of the two-dimensional image.