Back to results
US 12,602,769 B2Granted
Systems and methods for detecting vehicle defects
AssigneeFord Global Technologies, LLC
InventorsAdil Nizam Siddiqui, Mahmoud Yousef Ghannam, Jonathan Engels
Life of this patent
Estimating the term…
Abstract
A method includes obtaining an image from one or more imaging devices disposed in the interior of the vehicle, modifying the image based on a color conversion routine to generate a source image, generating a conformity value based on a comparison between the source image and a template image corresponding to the interior of the vehicle, where the conformity value is a cross-correlation value, a normalized cross-correlation value, a sum-of-squared difference value, or a combination thereof, and determining a presence of a defect based on the conformity value.
Claims14
A method for inspecting an interior of a vehicle, the method comprising:
- performing a manufacturing routine to build a vehicle, wherein the manufacturing routine comprises a plurality of steps;
- at each step of the plurality of steps, performing an inspection routine on an interior of the vehicle, wherein the inspection routine comprises:
- obtaining image data from one or more imaging devices disposed in the interior of the vehicle, wherein the image data includes one or more images corresponding to the interior of the vehicle and a vehicle identification number;
- generating, from averaging each pixel value of a first image of the image data with each pixel value of a second image of the image data, an averaged image;
- modifying the averaged image based on a color conversion routine to generate a source image;
- generating a conformity value based on a comparison between the source image and a template image corresponding to the interior of the vehicle, wherein the conformity value is a cross-correlation value, a normalized cross-correlation value, a sum-of-squared difference value, or a combination thereof;
- determining a presence of a defect based on the conformity value;
- identifying a component of the interior of the vehicle and pixel coordinates of the component based on a segmentation-based image processing routine;
- and
- determining a type of the defect as an installation defect when the pixel coordinates of the component do not correspond to expected pixel coordinates of the component;
- and
- remediating the installation defect in response to detecting the installation defect at one or more steps of the plurality of steps.
9 more elements
A system for inspecting an interior of a vehicle, the system comprising:
- a processor;
- and
- a non-transitory computer-readable medium comprising instructions that are executable by the processor, wherein the instructions comprise:
- performing a manufacturing routine to build a vehicle, wherein the manufacturing routine comprises a plurality of steps;
- at each step of the plurality of steps, performing an inspection routine on an interior of the vehicle, wherein the inspection routine comprises:
- obtaining image data from one or more imaging devices disposed in the interior of the vehicle, wherein the image data includes one or more images and a vehicle identification number;
- generating, from averaging each pixel value of a first image of the image data with each pixel value of a second image of the image data, an averaged image;
- modifying the averaged image based on a color conversion routine to generate a source image, wherein the source image is one of a binary-type image and a grayscale-type image;
- generating a conformity value based on a comparison between the source image and a template image corresponding to the interior of the vehicle, wherein the conformity value is a cross-correlation value, a normalized cross-correlation value, a sum-of-squared difference value, or a combination thereof;
- determining a presence of a defect based on the conformity value;
- identifying a component of the interior of the vehicle and pixel coordinates of the component based on a segmentation-based image processing routine;
- and
- determining a type of the defect as an installation defect when the pixel coordinates of the component do not correspond to expected pixel coordinates of the component;
- and
- remediating the installation defect in response to detecting the installation defect at one or more steps of the plurality of steps.
12 more elements
A method for inspecting an interior of a vehicle, the method comprising:
- performing a manufacturing routine to build a vehicle, wherein the manufacturing routine comprises a plurality of steps;
- at each step of the plurality of steps, performing an inspection routine on an interior of the vehicle, wherein the inspection routine comprises:
- obtaining image data from one or more imaging devices disposed in the interior of the vehicle, wherein the image data includes one or more images and a vehicle identification number;
- generating, from averaging each pixel value of a first image of the image data with each pixel value of a second image of the image data, an averaged image;
- modifying the averaged image based on a color conversion routine to generate a source image, wherein the source image is one of a binary-type image and a grayscale-type image;
- generating a conformity value based on a comparison between the source image and a template image corresponding to the interior of the vehicle, wherein the conformity value is a cross-correlation value, a normalized cross-correlation value, a sum-of-squared difference value, or a combination thereof;
- determining a presence of a defect based on the conformity value;
- identifying a component of the interior of the vehicle and pixel coordinates of the component based on a segmentation-based image processing routine;
- and
- determining a type of the defect as an installation defect when the pixel coordinates of the component do not correspond to expected pixel coordinates of the component;
- and
- remediating the installation defect in response to detecting the installation defect at one or more steps of the plurality of steps.
9 more elements
Dependencies are inferred from the claim text.