System and method for sensing underwater point cloud
AssigneeMetal Industries Research & Development Centre
InventorsWei-Te Chen, Shang-Ching Chuang, Chen-Shun Chen, Pin-Xian Lin, Chen-Han Kuo, Yun-Hui He, Yao-Ming Wang
Life of this patent
Estimating the term…
Abstract
The present invention proposes a method for sensing underwater point cloud. The method includes: acquiring point cloud information through an underwater sonar array, where the point cloud information contains multiple points; rearranging these points into a matrix which includes the aforementioned points and multiple missing points; and for each missing point, interpolating its depth value based on the depth values of several neighboring points.
Claims13
A system for sensing underwater point cloud, comprising:
- an underwater sonar array configured to acquire point cloud information comprising a plurality of points;
- an image sensor configured to capture a digital image comprising a plurality of pixels;
- and
- a computing module communicatively connected to the underwater sonar array, wherein the computing module is configured to rearrange the points into a matrix comprising the points and a plurality of missing points;
- wherein the computing module performs an interpolation calculation for each of the missing points based on depth values of neighboring points among the points to obtain a depth value of each of the missing points;
- wherein each of the pixels comprises a red gray value, a green gray value and a blue gray value, and the computing module is configured to calibrate a depth value of one of the missing points based on the following equation 2:
- P
- C
- =
- P
- ″
- +
- [
- (
- (
- R
- ″
- -
- R
- a
-
- v
-
- g
- )
- ×
- W
- R
-
- C
- )
- +
- (
- (
- G
- ″
- -
- G
- a
-
- v
-
- g
- )
- ×
- W
- G
-
- C
- )
- +
- (
- (
- B
- ″
- -
- B
- a
-
- v
-
- g
- )
- ×
- W
- B
-
- C
- )
- ]
- W
- R
-
- C
- =
- R
- a
-
- v
-
- g
- 1
- P
- ″
- W
- G
-
- C
- =
- G
- a
-
- v
-
- g
- 1
- P
- ″
- W
- B
-
- C
- =
- B
- a
-
- v
-
- g
- 1
- P
- ″
- wherein PC is a calibrated depth value, P″ is the depth value of the one of the missing points, R″ is the red gray value of the one of the missing points, G″ is the green gray value of the one of the missing points, B″ is the blue gray value of the one of the missing points, Ravg is an average red gray value of a plurality of similar pixels among the pixels, Gavg is an average green gray value of the similar pixels among the pixels, and Bavg is an average blue gray value of the similar pixels among the pixels.
A method for sensing underwater point cloud, executed by a computer system, wherein the method for sensing underwater point cloud comprises:
- acquiring point cloud information through an underwater sonar array, wherein the point cloud information comprises a plurality of points;
- capturing a digital image through an image sensor, wherein the digital image comprises a plurality of pixels, each of the pixels comprises a red gray value, a green gray value and a blue gray value;
- rearranging the points into a matrix comprising the points and a plurality of missing points;
- performing an interpolation calculation for each of the missing points based on depth values of neighboring points among the points to obtain a depth value of each of the missing points;
- and
- calibrating a depth value of one of the missing points based on the following equation 2:
- P
- C
- =
- P
- ″
- +
- [
- (
- (
- R
- ″
- -
- R
- a
-
- v
-
- g
- )
- ×
- W
- R
-
- C
- )
- +
- (
- (
- G
- ″
- -
- G
- a
-
- v
-
- g
- )
- ×
- W
- G
-
- C
- )
- +
- (
- (
- B
- ″
- -
- B
- a
-
- v
-
- g
- )
- ×
- W
- B
-
- C
- )
- ]
-
- W
- R
-
- C
- =
- R
- a
-
- v
-
- g
- 1
- P
- ″
-
- W
- G
-
- C
- =
- G
- a
-
- v
-
- g
- 1
- P
- ″
-
- W
- B
-
- C
- =
- B
- a
-
- v
-
- g
- 1
- P
- ″
- [
- Equation
-
- 2
- ]
- wherein PC is a calibrated depth value, P″ is the depth value of the one of the missing points, R′ is the red gray value of the one of the missing points, G″ is the green gray value of the one of the missing points, B″ is the blue gray value of the one of the missing points, Ravg is an average red gray value of a plurality of similar pixels among the pixels, Gavg is an average green gray value of the similar pixels among the pixels, and Bavg is an average blue gray value of the similar pixels among the pixels.
Dependencies are inferred from the claim text.