对数空间中基于点云配准的颜色校正算法研究-通信与信息系统专业论文.docx

摘 摘 要 I 万方数据 万方数据 万方数据 万方数据 摘 要 在计算机视觉中，常见的一个问题就是对于同一个成像物体来说，在不同的 拍摄条件下，得到的图像颜色是不同的。产生这个问题的原因主要有两个：一是 在成像过程中，数码相机的光学系统，光电传感器器件和相机的参数设置会对图 像颜色造成影响；二是不同的光源照射也会使物体表面显示的颜色有一定的偏差。 这一问题降低了计算机视觉中很多算法的性能，例如物体跟踪、图像拼接、3D 重 建、特殊图像处理等等。为了克服这一问题，对颜色校正技术的研究具有十分重 要的意义。 为了消除不想要的颜色差异，本文提出了一种新颖的基于点云配准的颜色校 正算法。该算法基于朗伯体反射模型。由于一幅图像的颜色用对数表示会有很多 优点，RGB 颜色空间首先被投影到对数颜色空间，然后在对数颜色空间中，本文 证明了同一物体或者相似物体在不同图像中的点云可以用一个三维平移向量来表 示；其次，在对数颜色空间中，使用这个三维平移向量来配准对应点云；最后， 被这个三维平移向量变换过的源图像跟参考图像显示为相似的颜色外貌。各种实 验结果表明了本文所提算法的有效性。 关键词：颜色校正 朗伯体反射模型 颜色空间 点云 Abstract III 万方数据 万方数据 万方数据 万方数据 Abstract A common problem in computer vision is that a digital camera may acquire different color responses of the same scene under different imaging environments responses to an imaged object. There are two main reasons for this problem. On the one hand, the physical factors such as the optical system, photoelectric sensor device and different parameters of the camera introduce a variation that differs for each sensor. On the other hand, even slight changes in light can introduce a substantially different ‘feel’ for shots taken at different times under natural light. This problem degrades the performance of colour computer vision processes such as object tracking, image mosaicing, 3D reconstruction, special effects processing and so on. In these cases, it becomes utterly important to make corrections for this variation. Therefore, research on the color correction technique has a very important significance. Caused by different illuminations and different camera settings in imaging process, color distortion incurs many problems. To overcome these problems, color correction is required. In this paper, a novel method for point cloud alignment based color correction is proposed to eliminate unwanted color variation between images of the same or similar scenes automatically. The method is based on the Lambertian reflection model. For the advantages inherent in the logarithmic representation of an image, we first project the