18072659. IMAGE PROCESSING TECHNIQUES TO VALIDATE SENSOR CLEANING simplified abstract (GM Cruise Holdings LLC)

IMAGE PROCESSING TECHNIQUES TO VALIDATE SENSOR CLEANING

Organization Name

GM Cruise Holdings LLC

Inventor(s)

Kilian Leport of San Francisco CA (US)

Jace Artichoker of San Francisco CA (US)

Wesley Newhouse of San Francisco CA (US)

Nicholas Decker of Sausalito CA (US)

Chin-Kai Chang of Sunnyvale CA (US)

IMAGE PROCESSING TECHNIQUES TO VALIDATE SENSOR CLEANING - A simplified explanation of the abstract

This abstract first appeared for US patent application 18072659 titled 'IMAGE PROCESSING TECHNIQUES TO VALIDATE SENSOR CLEANING

Simplified Explanation

The disclosed technology involves a method for confirming a sensor cleaning event of a sensor mounted on an autonomous vehicle by analyzing image frames from an optical sensor.

• Receiving a first set of image frames from the optical sensor
• Segmenting each image frame into interest regions
• Calculating image statistics for each interest region
• Creating an image mask based on the image statistics
• Receiving a second set of image frames from the optical sensor
• Applying the image mask to the second set of image frames
• Calculating image statistics for each frame in the second set
• Determining the status of a sensor cleaning event based on the image statistics

1. 1. 1. Potential Applications

- Autonomous vehicles - Sensor technology - Image processing

1. 1. 1. Problems Solved

- Ensuring sensors on autonomous vehicles are clean and functioning properly - Automating the process of confirming sensor cleaning events

1. 1. 1. Benefits

- Improved safety of autonomous vehicles - Reduced maintenance costs - Enhanced reliability of sensor data

1. 1. 1. Potential Commercial Applications
         1. Optimizing Sensor Maintenance in Autonomous Vehicles

1. 1. 1. Possible Prior Art

There may be prior art related to image processing techniques for sensor maintenance in various industries, but specific prior art related to confirming sensor cleaning events on autonomous vehicles may be limited.

1. 1. 1. 1. Unanswered Questions
         2. How does this technology compare to manual sensor cleaning verification methods?

This article does not provide a direct comparison to manual sensor cleaning verification methods. It would be interesting to know the efficiency and accuracy of this technology compared to traditional methods.

1. 1. 1. 1. What are the potential limitations of this technology in different environmental conditions?

The article does not address the potential limitations of this technology in various environmental conditions. It would be important to understand how factors such as weather, lighting, and terrain could impact the effectiveness of this sensor cleaning event confirmation method.

Original Abstract Submitted

The disclosed technology provides solutions for confirming a sensor cleaning event of a sensor mounted on an autonomous vehicle. A method of the disclosed technology can include steps for receiving a first plurality of image frames from an optical sensor; segmenting each of the first plurality of image frames into one or more interest regions; computing image statistics for each of the one or more interest regions; generating an image mask based on the image statistics for each of the one or more interest regions; receiving a second plurality of image frames from the optical sensor; applying the image mask to the second plurality of image frames; computing image statistics for each of the second plurality of image frames; and determining a status of a sensor cleaning event based on the image statistics for each of the second plurality of image frames. Systems and machine-readable media are also provided.