Method for calibration of camera and lidar, and computer program recorded on recording medium for executing method therefor
Proposed is a calibration method for a camera and a lidar using a calibration board having a specific pattern. The method may include extracting one frame for an image captured by a camera and point cloud data acquired by a lidar, by a data generator, identifying a feature point for a calibration board included in one extracted frame for the image and the lidar, by the data generator, and performing calibration for the camera and the lidar based on the identified feature point, by the data generator. The present method is a technology developed with support from the Ministry of Trade, Industry and Energy/Korea Planning and Evaluation Institute of Industrial Technology (Project No. 20022003/Project name-Automotive industry technology development project/Project name-Development of industrial autonomous driving and safety assurance technology based on longitudinal and lateral interconnection).
1. A calibration method comprising:
extracting one frame for each of an image captured by a camera and point cloud data acquired by a lidar;
identifying a feature point for a calibration board included in the one frame for the each of the image and the point cloud data; and
performing calibration for the camera and the lidar based on the feature point for the calibration board,
wherein the calibration board is a rectangular checker board, the calibration board comprises a border identification region that is formed at a border thereof by attaching along outer edges of the calibration board with a high-brightness tape formed of a material with an intensity higher than a preset value, and the calibration board comprises a top point identification region formed by removing a portion of the high-brightness tape from a top vertex of the board identification region to have an intensity of the top point identification region lower than the preset value,
wherein the identifying the feature point identifies a corner of the calibration board in the one frame,
wherein the identifying the feature point identifies a point corresponding to the corner using a preset pattern based on the feature point located on a top portion or a bottom portion of the calibration board and then assigns an index to the point corresponding to the corner,
wherein the identifying the feature point assigns the index in such a way that a number for the index increases down and right from the feature point located on the top portion of the calibration board, or assigns the index in such a way that the number for the index increases up and left from the feature point located on the bottom portion of the calibration board,
wherein the identifying the feature point further comprises:
generating a virtual plane based on the point cloud included in the border identification region;
connecting outermost points for each channel in the virtual plane to generate a plurality of straight lines;
detecting vertices of the calibration board based on intersections of the plurality of straight lines; and
identifying the feature point of the calibration board based on the vertices,
wherein the performing the calibration calculates external parameters including a rotation value and a translation value based on the feature point of the calibration board included in the one frame for the each of the image and the point cloud data, and performs calibration based on the external parameters, and
wherein the performing the calibration individually calculates the rotation value and the translation value.
2. The calibration method of claim 1 , wherein the identifying the feature point matches the index of the feature point identified from the image with the feature point identified from the lidar, based on the top point identification region of the calibration board.
3. The calibration method of claim 1 , wherein the identifying the feature point detects the calibration board from the point cloud data based on the border identification region.
4. The calibration method of claim 1 , wherein the performing the calibration calculates the translation value through a loss function based on Re-Projection Error (RPE).
5. A computer program recorded on a recording medium,
wherein the computer program is coupled to a computing device comprising:
a memory;
a transceiver; and
a processor processing an instruction loaded in the memory,
whereby the computer program, when executed by the processor, performs a method comprising:
extracting a calibration board from an image captured by a camera and point cloud data acquired by a lidar, by the processor;
identifying a feature point of the calibration board included in the image and the point cloud data, by the processor; and
performing calibration of the camera and the lidar based on the feature point, by the processor,
wherein the calibration board is a rectangular checker board, the calibration board comprises a border identification region that is formed at a border thereof by attaching along outer edges of the calibration board with a high-brightness tape formed of a material with an intensity higher than a preset value, and the calibration board comprises a top point identification region formed by removing a portion of the high-brightness tape from a top vertex of the board identification region to have an intensity of the top point identification region lower than the preset value,
wherein the identifying the feature point identifies a corner of the calibration board in the image,
wherein the identifying the feature point identifies a point corresponding to the corner using a preset pattern based on the feature point located on a top portion or a bottom portion of the calibration board and then assigns an index to the point corresponding to the corner,
wherein the identifying the feature point assigns the index in such a way that a number for the index increases down and right from the feature point located on the top portion of the calibration board, or assigns the index in such a way that the number for the index increases up and left from the feature point located on the bottom portion of the calibration board,
wherein the identifying the feature point further comprises:
generating a virtual plane based on the point cloud included in the border identification region;
connecting outermost points for each channel in the virtual plane to generate a plurality of straight lines;
detecting vertices of the calibration board based on intersections of the plurality of straight lines; and
identifying the feature point of the calibration board based on the vertices,
wherein the performing the calibration calculates external parameters including a rotation value and a translation value based on the feature point of the calibration board included in the image and the point cloud data, and performs calibration based on the external parameters, and
wherein the performing the calibration individually calculates the rotation value and the translation value.
6. The computer program of claim 5 , wherein the identifying the feature point matches the index of the feature point identified from the image with the feature point identified from the lidar, based on the top point identification region of the calibration board.
7. The computer program of claim 5 , wherein the identifying the feature point detects the calibration board from the point cloud data based on the border identification region.
8. The computer program of claim 5 , wherein the performing the calibration calculates the translation value through a loss function based on Re-Projection Error (RPE).