Method of using image recognition processes for improved operation of a laundry appliance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional washing machines often encounter issues with excessive suds, leading to inefficient wash cycles, increased energy and water usage, and potential skin irritation due to delayed detection of excessive detergent levels.
Innovation Solution
Integration of a camera assembly within the washing machine that uses machine learning image recognition processes to detect suds levels and adjust operating parameters, such as water addition or temperature, to rectify excessive suds conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of energy
If conventional washing machines operate without real-time suds detection, then the appliance structure remains simple and cost-effective, but excessive suds lead to delayed corrective action, suds lock mode, and increased energy and water consumption
Solution Approach 1:
The patent replaces conventional mechanical or chemical suds detection methods with an optical imaging system using a camera assembly and machine learning-based image recognition process. This substitution enables real-time suds level monitoring without significant mechanical complexity, allowing timely corrective actions that reduce energy consumption and prevent suds lock conditions
Solution Approach 2:
The patent implements a feedback control system where the camera assembly continuously monitors suds levels in the wash chamber, the controller analyzes images to determine suds thresholds, and corrective actions are automatically initiated when excessive suds are detected. This closed-loop feedback enables real-time optimization of wash cycles, reducing energy and water consumption while maintaining effective washing
2Loss of time
If conventional washing machines detect suds in a delayed manner, then the detection system remains simple, but corrective action is delayed causing suds lock mode and extended wash cycles
Solution Approach 1:
The patent performs preliminary suds detection by continuously capturing images of the wash chamber during the wash cycle. The machine learning model analyzes these images in real-time to predict when suds levels will exceed thresholds, enabling proactive corrective actions before suds lock conditions develop, thereby reducing overall wash cycle time
Solution Approach 2:
The patent replaces delayed mechanical or chemical detection methods with real-time optical image capture and machine learning analysis. This substitution enables immediate detection of suds level changes, allowing the controller to initiate corrective actions promptly and prevent extended wash cycles caused by suds lock mode
3Object-affected harmful factors
If conventional washing machines do not implement real-time suds monitoring, then the control system remains simple, but excessive suds cause skin irritation and require repeated rinse cycles
Solution Approach 1:
The patent implements automated feedback control where the camera assembly and machine learning model continuously monitor suds levels, and the controller automatically adjusts wash parameters or initiates corrective rinse cycles when excessive suds are detected. This automation ensures proper rinsing to prevent skin irritation while maintaining system simplicity through intelligent control algorithms
Solution Approach 2:
The patent enables the washing machine to self-monitor and self-correct suds level issues without user intervention. The machine learning model automatically analyzes images, determines suds thresholds, and triggers appropriate corrective actions, ensuring safe and effective washing while keeping the control system manageable through automated decision-making
Data Source
AI summary
A washing machine appliance includes a wash basket that is rotatably mounted within a wash tub and that defines a wash chamber for receiving a load of clothes. A camera assembly is mounted within the wash tub for taking images of the load of clothes during and/or after a wash or rinse cycle. A controller is operably coupled to the camera for obtaining and analyzing the images using a neural network image recognition process (such as an R-CNN process) to determine a whether the level of suds or bubbles exceeds a suds threshold. If there are too many suds in the wash tub, the controller may implement corrective action, e.g., by decreasing a water temperature, adding more water, or reducing the amount of detergent that is auto-dispensed in a subsequent wash cycle.