# US 12,118,740 B2 · Method and device for detecting propellers with high-precision

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 12,118,740 B2
- Document ID: US12118740B2
- Status: Granted
- Kind: B2
- Application number: 18/105,364
- Filed: 2023-02-03
- Published: 2024-06-20
- Granted: 2024-10-15
- Earliest claimed: 2022-12-19
- Assignees: Qingdao Polaris Intelligent Manufacturing Co., Ltd.
- Applicants: Qingdao Polaris Intelligent Manufacturing Co., Ltd.
- Inventors: Yongshun Wang; Zhidong Fu; Gongtao Xu
- Web page: https://patentgrep.com/patent/US12118740B2

## CPC

- G06T 7/0004: {Industrial image inspection}
- G06T 7/60: Analysis of geometric attributes
- G06T 3/60: Rotation of whole images or parts thereof
- G06T 7/10: Segmentation; Edge detection (motion-based segmentation G06T7/215)
- G06T 2207/10028: Range image; Depth image; 3D point clouds
- G06T 2207/20021: Dividing image into blocks, subimages or windows
- G06T 2207/30164: Workpiece; Machine component
- Y02T 90/00: Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- Y02T: CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION

## Abstract

A method and device for detecting propellers with high-precision are provided, the method includes obtaining a measurement point cloud of the propeller, filtering the measurement point cloud to obtain a target point cloud; processing the target point cloud to obtain a first measurement plane, and finding a center point of a pressure cone point cloud; obtaining a measuring axis according to the first measurement plane and the center point; segmenting the target point cloud to obtain a blade point cloud of each blade, and analyzing the blade point cloud and the measuring axis to obtain the parameters of the propeller, so as to determine whether the propeller is qualified. A high-precision structured light 3D camera is used, simple and high-precision detection solution for propeller products is developed based on the open-source 3D algorithm framework PCL. The device can improve the detection efficiency and the feasibility of promotion.

## Claims

### Claim 1

A method for detecting propellers with high-precision, comprising:
obtain a measurement point cloud of the propeller, and filter the measurement point cloud to obtain a target point cloud;
process the target point cloud to obtain a first measurement plane, and find a center point of a pressure cone point cloud;
obtain a measuring axis according to the first measurement plane and the center point;
segment the target point cloud to obtain the blade point cloud of each blade, and analyze the blade point cloud and measuring axis to obtain parameters of the propeller, so as to determine whether the propeller is qualified.

### Claim 2 (depends on 1)

The method according to claim 1, wherein obtain the measurement point cloud of the propeller, and filter the measurement point cloud to obtain the target point cloud comprises:
detect the propeller by means of a structured light 3D camera and obtain the measurement point cloud of the propeller;
filter the measurement point cloud to remove stray point cloud in order to obtain the target point cloud.

### Claim 3 (depends on 1)

The method according to claim 1, wherein process the target point cloud to obtain the first measurement plane comprises:
Step A: randomly select three non-collinear points from the target point cloud, and determine a plane through the three points;
Step B: find inliers from the target point cloud, wherein take the points whose distance to the plane is less than a preset distance as inliers;
Step C: update the three points selected in step A, re-determine a plane, and perform step B to obtain inliers of the updated plane;
iteratively process the step C, after a preset number of iterations, select the plane with the most inliers as the first measurement plane.

### Claim 4 (depends on 1)

The method according to claim 1, wherein obtain the measuring axis according to the first measurement plane and the central point comprises:
obtain a linear equation of the measuring axis according to the center point of the pressure cone point cloud and a normal vector of the first measurement plane;
transform a symmetric linear equation of the measuring axis into an intersection linear equation of the measuring axis to obtain the measuring axis.

### Claim 5 (depends on 1)

The method according to claim 1, wherein segment the target point cloud to obtain the blade point cloud of each blade, and analyze the blade point cloud and measuring axis to obtain parameters of the propeller, so as to determine whether the propeller is qualified, comprises:
segment the target point cloud to obtain the blade point cloud of each blade;
calculate the distance value of each blade point cloud from the measuring axis, and add the blade point cloud whose distance value satisfies a specific distance to a point set p;
calculate point P_max farthest from the first measurement plane in the point set P;
take the distance from the point P_max to the first measurement plane as the blade height, and calculate the full pitch of the propeller according to the full pitch measurement formula to determine whether the blade height and full pitch meet the requirements.

### Claim 6 (depends on 5)

The method according to claim 5, wherein segment the target point cloud to obtain the blade point cloud of each blade, and analyze the blade point cloud and measuring axis to obtain parameters of the propeller, so as to determine whether the propeller is qualified, further comprises:
calculate point P_min closest to the first measurement plane in the point set P;
calculate a second measurement plane formed by the point P_max and the measuring axis, and calculate a third measurement plane formed by the point P_min and the measuring axis;
calculate a target angle between the second measurement plane and the third measurement plane;
segment the second measurement plane according to a local equal fraction n and the target angle to obtain n equal division planes;
calculate a local pitch according to the point set P and each equal division plane.

### Claim 7 (depends on 6)

The method according to claim 6, wherein segment the second measurement plane according to a local equal fraction n and the target angle to obtain n equal division planes, comprises:
rotate the second measurement plane around the measuring axis by 1/n times, 2/n times . . . (n−1)/n times of the target angle according to the local equal fraction n to obtain n equal division planes.

### Claim 8 (depends on 6)

The method according to claim 6, wherein calculate the local pitch according to the point set P and each equal division plane, comprises:
calculate the intersection of the point set P and each equal division plane, and use the intersection as the input of the local pitch, and calculate the local pitch according to the pitch calculation method and the intersection to determine whether the local pitch meets the requirements.

### Claim 9 (depends on 6)

The method according to claim 6, wherein segment the target point cloud to obtain the blade point cloud of each blade, and analyze the blade point cloud and measuring axis to obtain parameters of the propeller, so as to determine whether the propeller is qualified, further comprises:
find the flanging point P_invert, obtain a vertical height difference between the flanging point P_invert and the point P_max from the first measurement plane, and use the height difference as the flanging height to determine whether the flanging height meets the requirements.

### Claim 10 (depends on 1)

A device for detection with high-precision, comprising:
one or more processors;
memory; and
one or more application programs, wherein the one or more application programs are stored in the memory and configured to be executed by the processor to implement the method according to claim 1.