US-12618184-B2 - Systems and methods using image recognition processes for improved operation of a laundry appliance

US12618184B2US 12618184 B2US12618184 B2US 12618184B2US-12618184-B2

Abstract

A method of operating a washing machine appliance includes obtaining an image of the wash basket and a load of articles therein from a camera assembly of a remote user interface device. The method also includes analyzing the obtained image using a machine learning image recognition process. Analyzing the obtained image may include determining a ratio of a diameter of an area occupied by the load of articles to a diameter of the wash basket or determining a ratio of a maximum height of an area occupied by the load of articles to a major axis of the wash basket. The method further includes estimating a load size of the load of articles based on the analysis and directing a wash cycle within the washing machine appliance based on the estimated load size.

Inventors

Myunggeon Chung
Hyeonsoo Moon
Je Kwon Yoon

Assignees

HAIER US APPLIANCE SOLUTIONS, INC.

Dates

Publication Date: 20260505
Application Date: 20230829

Claims (19)

1 . A method of determining a load size for a washing machine appliance, the washing machine appliance defining a vertical direction, a lateral direction, and a transverse direction, the vertical direction, the lateral direction, and the transverse direction being mutually perpendicular, the washing machine appliance comprising a cabinet, a wash tub mounted within the cabinet, and a wash basket, the wash basket defining a wash chamber, the wash basket rotatably mounted within the wash tub whereby the wash basket is rotatable about the vertical direction, the method comprising: obtaining one or more images of the wash basket and a load of articles therein from a camera assembly of a remote user interface device; analyzing at least one obtained image using a machine learning image recognition process, wherein analyzing the at least one obtained image comprises determining a ratio of a diameter of an area occupied by the load of articles to a diameter of the wash basket; estimating the load size of the load of articles based on the analysis; and directing a wash cycle within the washing machine appliance based on the estimated load size.
2 . The method of claim 1 , further comprising estimating an additional load attribute based on the analysis, wherein the additional load attribute comprises at least one of a fabric type or a load color.
3 . The method of claim 1 , further comprising displaying, on a display of the remote user interface device, a reticle configured to align with the wash basket.
4 . The method of claim 1 , further comprising receiving a plurality of angle readings from the remote user interface device, determining a position of the remote user interface device relative to the washing machine appliance based on the plurality of angle readings, and determining a set camera angle for a camera assembly of the remote user interface device is met based on the determined position of the remote user interface device.
5 . The method of claim 4 , wherein the plurality of angle readings are detected by an accelerometer of the remote user interface device.
6 . The method of claim 1 , wherein directing the wash cycle within the washing machine appliance based on the estimated load size comprises setting a dispense dosage for the wash cycle based on the estimated load size.
7 . The method of claim 1 , wherein directing the wash cycle within the washing machine appliance based on the estimated load size comprises setting a fill volume for the wash cycle based on the estimated load size.
8 . The method of claim 1 , wherein obtaining one or more images comprises receiving a video signal from the camera assembly, and wherein the method further comprises: presenting a real-time feed of the camera assembly at the remote user interface device according to the received video signal; and overlaying a two-dimensional reference shape over the real-time feed.
9 . The method of claim 1 , wherein the diameter of the area occupied by the load of articles is determined based on an image segmentation map.
10 . A method of determining a load size for a washing machine appliance, the washing machine appliance defining a vertical direction, a lateral direction, and a transverse direction, the vertical direction, the lateral direction, and the transverse direction being mutually perpendicular, the washing machine appliance comprising a cabinet, a wash tub mounted within the cabinet, and a wash basket, the wash basket defining a wash chamber, the wash basket rotatably mounted within the wash tub whereby the wash basket is rotatable about the vertical direction, the method comprising: obtaining one or more images of the wash basket and a load of articles therein from a camera assembly of a remote user interface device; analyzing at least one obtained image using a machine learning image recognition process, wherein analyzing the at least one obtained image comprises determining a ratio of a maximum height of an area occupied by the load of articles to a minor axis of the wash basket; estimating the load size of the load of articles based on the analysis; and directing a wash cycle within the washing machine appliance based on the estimated load size.
11 . The method of claim 10 , further comprising estimating an additional load attribute based on the analysis, wherein the additional load attribute comprises at least one of a fabric type or a load color.
12 . The method of claim 10 , further comprising displaying, on a display of the remote user interface device, a reticle configured to align with the wash basket.
13 . The method of claim 10 , further comprising determining a position of the remote user interface device relative to the washing machine appliance based on a plurality of angle readings, and determining a set camera angle for a camera assembly of the remote user interface device is met based on the determined position of the remote user interface device.
14 . The method of claim 13 , wherein the position of the remote user interface device is determined by analyzing a geometry of the wash basket in the at least one obtained image.
15 . The method of claim 14 , wherein the wash basket appears as an ellipse in the at least one obtained image, and wherein geometry of the wash basket is an eccentricity of the ellipse.
16 . The method of claim 10 , wherein directing the wash cycle within the washing machine appliance based on the estimated load size comprises setting a dispense dosage for the wash cycle based on the estimated load size.
17 . The method of claim 10 , wherein directing the wash cycle within the washing machine appliance based on the estimated load size comprises setting a fill volume for the wash cycle based on the estimated load size.
18 . The method of claim 10 , wherein obtaining one or more images comprises receiving a video signal from the camera assembly, and wherein the method further comprises: presenting a real-time feed of the camera assembly at the remote user interface device according to the received video signal; and overlaying a two-dimensional reference shape over the real-time feed.
19 . The method of claim 10 , wherein the maximum height of the area occupied by the load of articles is determined based on an image segmentation map.

Description

FIELD OF THE INVENTION The present subject matter relates generally to washing machine appliances, or more specifically, to systems and methods for using image recognition processes to improve or optimize operation of washing machine appliances. BACKGROUND OF THE INVENTION Washing machine appliances generally include a tub for containing water or wash fluid, e.g., water and detergent, bleach, or other wash additives. A basket is rotatably mounted within the tub and defines a wash chamber for receipt of articles for washing. During normal operation of such washing machine appliances, the wash fluid is directed into the tub and onto articles within the wash chamber of the basket. The basket or an agitation element can rotate at various speeds to agitate articles within the wash chamber, to wring wash fluid from articles within the wash chamber, etc. During a spin or drain cycle, a drain pump assembly may operate to discharge water from within sump. Notably, it is frequently desirable to understand characteristics of a load of articles, e.g., clothes, within the washing machine appliance, e.g., in order to optimize water usage, agitation time, agitation profile selection, and other wash parameters. For example, certain loads (e.g., towels or linens) may require more water and detergent, increased water temperature, and stronger agitation cycles. By contrast, other loads (e.g., such mixed color loads or delicates) may require cooler water and a gentler agitation profile. However, conventional washing machine appliances require a user to select operating cycles or specify the type of load added to the wash chamber, often resulting in inaccurate inputs or sub-optimal cycle settings. Attempts have been made to automatically (e.g., without direct user input or estimations) detect certain attributes of a load using sensors or detection assemblies within the washing machine appliance. Unfortunately, though, such systems may increase the expense and complexity of an appliance. Moreover, it can be difficult for a user to know if or when any detected attributes have been detected accurately or correctly. Accordingly, a washing machine appliance with features for improved wash performance would be desirable. More specifically, a system and method for automatically detecting characteristics of the load of articles and determining preferred operating parameters would be particularly beneficial, especially if such systems or methods could be achieved without requiring additional or dedicated sensing assemblies to be installed on the washing machine appliance. Additionally or alternatively, it may be beneficial to provide a system or method wherein a user could be confident that characteristics were detected in the correct manner (e.g., to ensure accuracy of such detections). BRIEF DESCRIPTION OF THE INVENTION Aspects and advantages of the invention will be set forth in part in the following description, or may be obvious from the description, or may be learned through practice of the invention. In one exemplary aspect of the present disclosure, a method of operating a washing machine appliance is provided. The washing machine appliance defines a vertical direction, a lateral direction, and a transverse direction. The vertical direction, the lateral direction, and the transverse direction are mutually perpendicular. The washing machine appliance includes a cabinet, a wash tub mounted within the cabinet, and a wash basket. The wash basket defines a wash chamber. The wash basket is rotatably mounted within the wash tub whereby the wash basket is rotatable about the vertical direction. The method includes obtaining one or more images of the wash basket and a load of articles therein from a camera assembly of a remote user interface device. The method also includes analyzing at least one obtained image using a machine learning image recognition process. Analyzing the at least one obtained image includes determining a ratio of a diameter of an area occupied by the load of articles to a diameter of the wash basket. The method further includes estimating a load size of the load of articles based on the analysis and directing a wash cycle within the washing machine appliance based on the estimated load size. In another exemplary aspect of the present disclosure, a method of operating a washing machine appliance is provided. The washing machine appliance defines a vertical direction, a lateral direction, and a transverse direction. The vertical direction, the lateral direction, and the transverse direction are mutually perpendicular. The washing machine appliance includes a cabinet, a wash tub mounted within the cabinet, and a wash basket. The wash basket defines a wash chamber. The wash basket is rotatably mounted within the wash tub whereby the wash basket is rotatable about the vertical direction. The method includes obtaining one or more images of the wash basket and a load of articles therein from a camera assembly of a remote u