| Sonstiges: |
- Nachgewiesen in: USPTO Patent Applications
- Sprachen: English
- Document Number: 20200077383
- Publication Date: March 5, 2020
- Appl. No: 16/670446
- Application Filed: October 31, 2019
- Claim: 1. A three-dimensional (3D) depth imaging system for assessing an orientation with respect to a container, the 3D depth imaging system comprising: a 3D-depth camera configured to capture 3D image data, the 3D-depth camera oriented in a first direction to capture the 3D image data of a shipping container located in a space, the shipping container having a shipping container type; and a container feature assessment application (app) configured to execute on one or more processors and to receive the 3D image data, the container feature assessment app configured to determine, based on the 3D image data, a container point cloud representative of the shipping container, wherein the container feature assessment app is further configured to execute on the one or more processors to: (a) convert the 3D image data into 2D depth image data; (b) compare the 2D depth image data to one or more template image data, each of the one or more template image data corresponding to a respective shipping container type; (c) in response to a match between the 2D depth image data and the one or more template image data, perform segmentation to extract 3D point cloud features; (d) from the 3D point cloud features, determine exterior features of the shipping container and assess the exterior features using an exterior features metric; (e) from the 3D point cloud features, determine interior features of the shipping container and assess the interior features using an interior features metric; and (f) in response to assessing the exterior features and assessing the interior features, generate an orientation adjustment instruction for indicating to an operator to orient the 3D-depth camera in a second direction for use during a shipping container loading session, wherein the second direction is different than the first direction.
- Claim: 2. The 3D depth imaging system of claim 1, wherein the container feature assessment app is further configured to convert the 3D image data into 2D depth image data by generating the 2D depth image data as grayscale image data, and wherein the container feature assessment app is further configured to compare the 2D depth image data to the one or more template image data in grayscale to identify a template matching portion.
- Claim: 3. The 3D depth imaging system of claim 2, wherein the container feature assessment app is further configured to execute on the one or more processors to: receive 3D image point cloud data for a template shipping container, convert the 3D image point cloud data to a depth image data, apply a penalized least squares process to introduce missing data into the depth image data and generate a template depth image, and normalize the template depth image to a grayscale to generate the template image data.
- Claim: 4. The 3D depth imaging system of claim 2, wherein the container feature assessment app is further configured to execute on the one or more processors to: segment the 3D image data into a shipping container ground plane and a shipping container back wall based on the template matching portion; and feed the shipping container ground plane and the shipping container back wall to a plane regression pipeline and determine and segment a shipping container front plane, wherein the 3D point cloud features comprise the shipping container ground plane, the shipping container back wall, and the shipping container front plane.
- Claim: 5. The 3D depth imaging system of claim 4, wherein the container feature assessment app is further configured to assess the exterior features using the exterior features metric by: dividing the shipping container front plane into a left side portion, a right side portion, and a top edge portion, determining completeness ratios for each of the left side portion, the right side portion, and the top edge portion, and determining, from the completeness ratios, a completeness of the shipping container front plane as the exterior features metric.
- Claim: 6. The 3D depth imaging system of claim 5, wherein the container feature assessment app is further configured to assess the interior features using the interior features metric by: determining a height of the shipping container front plane, determining a height of the shipping container back wall, and determining an occlusion ratio as the interior features metric.
- Claim: 7. The 3D depth imaging system of claim 6, wherein the container feature assessment app is further configured to execute on the one or more processors to: determine six planes of the shipping container; determine a bounding box of the shipping container based on the six planes; and generate the orientation adjustment instruction in response to determining the bounding box, the orientation adjustment instruction comprising at least one of an orientation to right instruction, an orientation to left instruction, a lower/raise instruction, and a tilt up/down instruction.
- Claim: 8. The 3D depth imaging system of claim 1, wherein the 3D-depth camera and the one or more processors are housed in a load monitoring unit (LMU).
- Claim: 9. The 3D depth imaging system of claim 1, wherein the one or more template image data corresponds to one or more different universal loading device (ULD) types.
- Claim: 10. The 3D depth imaging system of claim 1, further comprising an installation visualization app, the installation visualization app executing on a client device implementing a graphical user interface (GUI), the GUI graphically indicating the orientation adjustment instruction on a digital display of the client device.
- Claim: 11. The 3D depth imaging system of claim 10, wherein the installation visualization app is configured to display the orientation adjustment instruction on a heads up display (HUD) communicatively coupled to the 3D-depth camera via a digital network.
- Claim: 12. The 3D depth imaging system of claim 1, further comprising an installation voice instruction app, the installation voice instruction app executing on a client device implementing a speaker for audibly communicating the orientation adjustment instruction to the operator of the client device.
- Claim: 13. A method for assessing an orientation with respect to a container of a 3D depth imaging system in an active installation mode, the method comprising: capturing, by a 3D-depth camera, 3D image data, the 3D-depth camera oriented in a first direction to capture the 3D image data of a shipping container located in a space during the active installation mode, the shipping container having a shipping container type; determining, based on the 3D image data, a container point cloud representative of the shipping container; converting the 3D image data into 2D depth image data; comparing the 2D depth image data to one or more template image data, each of the one or more template image data corresponding to a respective shipping container type; in response to a match between the 2D depth image data and the one or more template image data, performing segmentation to extract 3D point cloud features; determining, from the 3D point cloud features, exterior features of the shipping container and assessing the exterior features using an exterior features metric; determining, from the 3D point cloud features, interior features of the shipping container and assessing the interior features using an interior features metric; and in response to assessing the exterior features and assessing the interior features, generating an orientation adjustment instruction for indicating to an operator to orient the 3D-depth camera in a second direction for use during a shipping container loading session, wherein the second direction is different than the first direction.
- Claim: 14. The method of claim 13, further comprising: converting the 3D image data into 2D depth image data by generating the 2D depth image data as grayscale image data; and comparing the 2D depth image data to the one or more template image data in grayscale to identify a template matching portion.
- Claim: 15. The method of claim 14, further comprising: receiving 3D image point cloud data for a template shipping container; converting the 3D image point cloud data to a depth image data; applying a penalized least squares process to introduce missing data into the depth image data and generate a template depth image; and normalizing the template depth image to a grayscale to generate the template image data.
- Claim: 16. The method of claim 14, further comprising: segmenting the 3D image data into a shipping container ground plane and a shipping container back wall based on the template matching portion; and feeding the shipping container ground plane and the shipping container back wall to a plane regression pipeline and determine and segment a shipping container front plane, wherein the 3D point cloud features comprise the shipping container ground plane, the shipping container back wall, and the shipping container front plane.
- Claim: 17. The method of claim 16, wherein assessing the exterior features using the exterior features metric comprises: dividing the shipping container front plane into a left side portion, a right side portion, and a top edge portion, determining completeness ratios for each of the left side portion, the right side portion, and the top edge portion, and determining, from the completeness ratios, a completeness of the shipping container front plane as the exterior features metric.
- Claim: 18. The method of claim 17, wherein assessing the interior features using the interior features metric comprises: determining a height of the shipping container front plane, determining a height of the shipping container back wall, and determining an occlusion ratio as the interior features metric.
- Claim: 19. The method of claim 18, further comprising: determining six planes of the shipping container; determining a bounding box of the shipping container based on the six planes; and generating the orientation adjustment instruction in response to determining the bounding box, the orientation adjustment instruction comprising at least one of an orientation to right instruction, an orientation to left instruction, a lower/raise instruction, and a tilt up/down instruction.
- Claim: 20. The method of claim 13, wherein the 3D-depth camera is housed in a load monitoring unit (LMU).
- Claim: 21. The method of claim 13, wherein the one or more template image data corresponds to one or more different universal loading device (ULD) types.
- Claim: 22. The method of claim 13, further comprising graphically indicating, by an installation visualization application (app) executing on a client device implementing a graphical user interface (GUI), the orientation adjustment instruction on a digital display of the client device.
- Claim: 23. The method of claim 22, wherein graphically indicating the orientation adjustment instruction includes displaying the orientation adjustment instruction on a heads up display (HUD) communicatively coupled to the 3D-depth camera via a digital network.
- Claim: 24. The method of claim 13, further comprising audibly communicating, by an installation voice instruction app executing on a client device implementing a speaker, the orientation adjustment instruction to the operator of the client device.
- Current International Class: 04; 04; 04; 04; 04
|
