Study of 3D Shape Reconstruction from Medical and Industrial Endoscope Images

在本研究中，我们建立了一种基于医学和工业内窥镜图像的三维形状重建方法，并构建了实验系统来验证其性能。主要工作是接下来两种新方法的开发：1。因式分解方法在内窥镜图像序列三维形状重建中的新应用。该方法利用大量的点和图像，鲁棒地应用了易于理解的矩阵计算，实现了高精度的形状重建。我们承认，这种方法易于理解，易于实现，并提供足够的精度的情况下，近似的光学系统保持良好。然而，该方法的后半部分过程，称为归一化，并不像前半部分使用奇异值分解那样容易理解。然后，在本研究中，我们重建了详细的理论基础，并提高了实际应用的可行性。在点光源位于投影中心的情况下，从阴影图像重建形状。这种情况很接近内窥镜的成像系统。在这种情况下，图像的灰度不仅取决于物体表面的梯度，还取决于光源到表面上每个点的距离。为了解决这一难题，我们引入了曲面等距轮廓的演化方程。通过求解这个方程来传播这个轮廓，我们可以重建一个形状。对人体胃内壁的真实医学内窥镜图像的实验结果表明了该方法的可行性，为人体内脏肿瘤的形态学分析提供了一种有前景的技术。

In this research, we established a methodology for 3-dimensional shape reconstruction from Medical and Industrial Endoscope Images, and constructed experimental system to verify its performances.The main works were development of next two new method :1.A new application of the factorization method for 3-D shape reconstruction from endoscope image sequences.This method achieves high accuracy of shape reconstruction by using a large number of points and images, and robustly applying a well-understood matrix computations. We admitted that this method was easy to understand, easy to implement, and providing enough accuracy for the case where the approximation on the optical system holds well.However, the latter half process of the method, named normalization, was not so wellunderstandable as the use of singular value decomposition in its first half. Then, in this study, we reconstructed the detail theoretical basis and inproved the feasibility of actual use for medical applications2.Reconstructing a shape from it's shaded image in the case where a point light source is at the projection center. This condition well approximates the imaging system of an endoscope. In this case, the image gray level depends on not only the gradient of the object surface but also the distance from the light source to each point on the surface. To deal with this difficulty, we introduce the evolution equation for equal-range contours on the surface. Propagating this contour by solving the equation, we can reconstruct a shape. Experimental results for real medical endoscope images of a human stomach inner wall show feasibility of this method, and present a promising technique for morphological analysis of tumors on human inner organs.