Spatial calibration for accurate long-distance measurement using infrared cameras

Patent metadata

Field	Value
Publication number	US-12536701-B1
Application number	US-202418442762-A
Country	US
Kind code	B1
Filing date	Feb 15, 2024
Priority date	Feb 15, 2023
Publication date	Jan 27, 2026
Grant date	Jan 27, 2026

Abstract

A method for spatial calibration of an infrared (IR) camera allows for a large calibration area thereby providing high fidelity results for accurate long-distance measurement. Halogen lamps capable of being surveyed are spread over a relatively wide area to serve as calibration points for the infrared camera. The halogen lamps provide a precise and symmetric signature suitable for mid-wave IR cameras. An accurate model is developed for the radial and tangential distortion of the camera system. Likewise, values for the intrinsic parameters (focal lengths and optional skew coefficient) are measured to improve real-world measurements made using images recorded from a calibrated camera system.

First claim

The invention claimed is:

1 . A method for conducting spatial calibration of an infrared camera system comprising steps of:
providing an infrared camera system and a plurality of infrared source points at a desired distance from the camera system; surveying a measured location for each of the plurality of infrared source points; recording a plurality of calibration images with the infrared camera system; processing the plurality of calibration images to identify relevant information within the calibration image; determining a camera calibration parameter from the relevant information of the processed calibration images and the measured location of the infrared camera system and the measured location of each of the plurality of infrared source points; and, wherein the each of the plurality of infrared source points comprises a halogen lamp, said halogen lamp further comprising a front surface covered by a cap, said cap defining a circular central hole extending longitudinally through the cap to the front surface of the halogen lamp, wherein said central hole is covered by a survey sticker having a circular opening axially aligned with the central hole of the cap.

2 . The method of claim 1 wherein
the halogen lamp is a thirty-six millimeter diameter twelve Volt 10 Watt halogen lamp;
the cap is comprised of acrylonitrile butadiene styrene (ABS) plastic and the central hole has a diameter of six millimeters; and
the opening in the survey sticker has a diameter of three and two tenths millimeters.

3 . The method of claim 1 wherein the plurality of calibration points is at a distance from the infrared camera system such that the plurality of calibration points occupies eighty percent of a camera field of view.

4 . The method of claim 1 wherein the camera system is supported on a gimbal system at an effective focal point of the camera system and the step of surveying a location of the camera system comprises surveying locations of four surveying stickers placed equidistant from a center of the gimbal system.

5 . The method of claim 1 wherein the step of recording a plurality of calibration images by the camera system further comprises recording a plurality of static frames of the calibration points.

6 . The method of claim 1 wherein the step of recording a plurality of calibration images further comprises recording a plurality of dynamic frames while rotating a pitch and a yaw of the camera system.

7 . The method of claim 1 wherein the step of processing the plurality of the calibration images to identify relevant information within the calibration image further comprises the steps of: determining a sub-pixel location of each calibration point for each image recorded and removing any false positive results and false negative results from the determinations.

8 . The method of claim 1 wherein:
the method further comprises the step of converting the measured location for each of the plurality of calibration points from a surveyed coordinate system to an azimuth and elevation coordinate system;
the step of determining a camera calibration parameter from the relevant information of the processed calibration images and the measured location of the camera system and the measured location of each of the plurality of calibration points further comprises determining a calibration transform which corrects for a radial distortion component and a tangential distortion component of the camera system; and
wherein the step of determining a camera calibration parameter further comprises the step of employing an iterative fitting algorithm, said iterative fitting algorithm comprising an inner loop for optimizing extrinsic factors for each individual image of the plurality of calibration images and an outer loop for optimizing distortion model coefficients of the camera system and one or more intrinsic factors of the camera system.

9 . The method of claim 8 further comprising the steps of:
setting an initial guess at the one or more intrinsic factors of the camera system;
setting an initial guess at the distortion model coefficients;
providing a calculated location of each identified calibration point for each image of the plurality of calibration images;
for each individual image, optimizing the extrinsic parameters for the individual image by minimizing an error between a location of the measured calibration points and a location of the calculated calibration points below a first threshold;
calculating a summed error for the plurality of calibration images; and
optimizing the intrinsic parameters of the camera system by minimizing a reprojection error of the summed error below a second threshold.

10 . The method of claim 9 wherein the step of optimizing the optimizing the extrinsic parameters for the individual image by minimizing an error between a location of the measured calibration points and a location of the calculated calibration points below a first threshold further comprises the steps of:
setting an initial guess for the extrinsic factors for a first image as zero for each factor;
for the first image, optimizing the extrinsic parameters of the individual image by minimizing an error between a location of the measured calibration points and a location of the calculated calibration points below a first threshold; and
for each subsequent image, setting an initial guess for the extrinsic factors of the image as the optimized extrinsic factors of the preceding image and optimizing the extrinsic parameters of the individual image by minimizing an error between a location of the measured calibration points and a location of the calculated calibration points below a first threshold.

11 . The method of claim 8 wherein the extrinsic factors comprise a roll measurement, a pitch measurement, and a yaw measurement for each frame.

12 . The method of claim 8 wherein the intrinsic factors comprise a focal length and a skew coefficient.

13 . The method of claim 8 wherein the distortion model parameters comprise a principal point coordinate, one or more radial distortion coefficients and one or more tangential distortion coefficients.

Assignees

• Us Gov Sec Army
• Us Army

Inventors

• Decker Ryan
• Abruzzo Benjamin
• Hassan Ahmed
• Horvath Mihaly
• Manole Steven
• Mirza Arhum

Classifications

• G06T2207/10048

  Infrared image · CPC title

• H04N17/002 Primary

  for television cameras · CPC title

• G01C25/00

  Manufacturing, calibrating, cleaning, or repairing instruments or devices referred to in the other groups of this subclass (testing, calibrating or compensating compasses G01C17/38) · CPC title

• G01C11/02

  Picture taking arrangements specially adapted for photogrammetry or photographic surveying, e.g. controlling overlapping of pictures · CPC title

• G06T7/80 Primary

  Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration · CPC title