Method and system for sensor maintenance

Patent metadata

Field	Value
Publication number	US-11680712-B2
Application number	US-202117201953-A
Country	US
Kind code	B2
Filing date	Mar 15, 2021
Priority date	Mar 13, 2020
Publication date	Jun 20, 2023
Grant date	Jun 20, 2023

Abstract

The method for dirty camera detection including: detecting a first predetermined state change event; sampling a set of cavity measurements; optionally determining a set of features of the set of cavity measurements; determining a class label based on the cavity measurements; optionally verifying the classification; and facilitating use of the appliance based on the classification.

First claim

We claim:

1. A system for dirty camera detection, comprising:
an appliance, comprising a camera and a cook cavity; and
a processor, configured to:
detect a first state change event at the appliance;
after detection of the first state change event, sample a set of cavity measurements;
determine a class label for the cook cavity using the set of cavity measurements, wherein the class label represents multi-component state combinations for the cook cavity and the camera; and
facilitate use of the appliance based on the class label.

2. The system of claim 1 , wherein the appliance is an oven.

3. The system of claim 1 , wherein the processor is further configured to:
store preceding class labels;
retrieve the preceding class labels in response to the class label indicative of a dirty camera state; and
verify that the camera is dirty when at least a predetermined number of dirty camera state classifications appear within the preceding class labels.

4. The system of claim 3 , wherein the preceding class labels that are retrieved comprise class labels from a predetermined time period prior to a current time.

5. The system of claim 3 , wherein the predetermined number of dirty camera state classifications are for images that depict empty cook cavities.

6. The system of claim 1 , wherein the class label is determined using a neural network, wherein the neural network is trained using a plurality of images that depict an empty cook cavity.

7. The system of claim 6 , wherein the neural network is trained using images from a plurality of different appliances.

Assignees

• June Life Inc

Inventors

• Bhogal Nikhil
• Kumar Nishit
• Cottrell Robert
• Paruchuri Jithendra
• Perry Ryan
• Van Horn Matthew
• Clark Christopher Russell
• Kuo Eugenia

Classifications

• G06V10/82

  using neural networks · CPC title

• G06V10/764

  using classification, e.g. of video objects · CPC title

• F24C7/085 Primary

  on baking ovens · CPC title

• G06V20/68

  Food, e.g. fruit or vegetables · CPC title

• G06V10/993

  Evaluation of the quality of the acquired pattern · CPC title