研究目的
Automating the camera calibration process using coded targets to minimize localization errors and improve precision.
研究成果
The automated process using ARUCO coded targets successfully achieved sub-pixel precision in camera calibration, reducing time and errors compared to manual methods. The results demonstrated high accuracy in estimating camera parameters, with potential for full automation in future work by addressing current limitations.
研究不足
The technique is intolerant to noise or occlusions in the outer target square (corona), which can lead to failure in target identification. Future improvements could involve algorithms for handling fragmented lines and using internal points of targets.
1：Experimental Design and Method Selection:
The study involved analyzing various coded targets and selecting ARUCO for its flexibility and integration with OpenCV. A software application was developed to automatically locate and measure targets in images with sub-pixel precision.
2：Sample Selection and Data Sources:
Two calibration panels with ARUCO targets were created: Panel 1 with 24 targets and Panel 2 with 48 targets. Images were acquired using a Kodak CX 4300 camera.
3：List of Experimental Equipment and Materials:
Kodak CX 4300 camera, calibration panels with ARUCO targets, software tools including OpenCV library and custom applications in Visual Studio 2010 and Java.
4：Experimental Procedures and Operational Workflow:
Targets were generated and printed. Images were captured from different orientations. The software processed images to detect targets, extract corner coordinates, and output results for calibration.
5：Data Analysis Methods:
Calibration was performed using bundle adjustment techniques. Statistical analysis included estimating parameters and their standard deviations, comparing automated and manual measurements.
独家科研数据包，助您复现前沿成果，加速创新突破
获取完整内容
