|
| 透镜成像模型约束的大视场图像测量相机成像参数标定方法研究* |
| Research on Calibration Method for Camera Imaging Parameters in Large Field-of-View Image Measurement Constrained by Lens Imaging Model |
| 投稿时间：2024-12-24 修订日期：2025-02-26 |
| DOI： |
| 中文关键词: 图像测量 相机标定 大视场 透镜成像模型 |
| 英文关键词:Image measurement Camera calibration Large field-of-view Lens imaging model. |
| 基金项目:国家自然科学基金项目（面上项目，重点项目，重大项目）；上海交通大学深蓝计划 |
|
| 摘要点击次数: 237 |
| 全文下载次数: 0 |
| 中文摘要: |
| 图像测量技术因其全场、非接触的特点而被广泛应用于航空、航天、航海和轨道交通等领域。图像测量过程中需要通过相机标定来确定二维图像与三维场景之间的转换关系,可以说,标定精度直接决定了图像测量精度。为了保证标定精度,标定中通常要求标定物视场占比不低于1/2。然而,在一些大视场测量场景中,标定物尺寸受限而无法满足视场占比要求,导致相机标定精度不足。为此,需要研究大视场相机成像参数标定方法,以保证图像测量技术在大视场测量场景中的有效应用。本文实验研究了不同物距对成像参数标定结果的影响,并通过透镜成像模型分析了影响的机制,据此提出了一种透镜成像模型约束的大视场相机成像参数标定方法。实验结果和分析表明：物距变化对等效焦距标定结果存在显著影响；透镜成像模型约束下,本文所提标定方法在近距离小视场下获得成像参数初值后,在远距离大视场下仅需拍摄单幅标定物图像即可获得高精度的成像参数标定结果,且标定物视场占比要求可降至1/8。 |
| 英文摘要: |
| Image measurement technology is widely used in fields such as aviation, aerospace, navigation, and rail transportation due to its full-field and non-contact characteristics. The camera calibration is necessary in the image measurement process to determine the transformation relationship between two-dimensional images and three-dimensional scenes. It can be said that the calibration accuracy directly determines the precision of image measurement. To ensure calibration accuracy, the calibration object typically needs to occupy no less than 1/2 of the field of view. However, in some large field-of-view measurement scenarios, the size of the calibration object is limited, failing to meet the field of view occupancy requirement, leading to insufficient camera calibration accuracy. Therefore, it is necessary to study the calibration method for large field-of-view camera imaging parameters to ensure the effective application of image measurement technology in large field-of-view measurement scenarios. This paper experimentally investigates the influence of different object distances on the imaging parameter calibration results and analyzes the mechanism of influence through the lens imaging model. Based on this, a lens imaging model-constrained large field-of-view camera imaging parameter calibration method is proposed. Experimental results and analysis show that the object distance variation has a significant impact on the equivalent focal length calibration results; under the constraint of the lens imaging model, the calibration method proposed in this paper can obtain high-precision imaging parameter calibration results with only a single calibration object image at a long distance and large field of view after obtaining the initial imaging parameters at a short distance and small field of view, and the requirement for the calibration object’s field of view occupancy can be reduced to 1/8. |
| 下载PDF阅读器 |
| 关闭 |
|
|
|
|
|
