The paper presents a method for globally optimal stitching of tiled 3D microscopic image acquisitions, which is crucial for high-resolution imaging of large biological specimens. The authors, Stephan Preibisch, Stephan Saalfeld, and Pavel Tomancak, developed a method based on the Fourier Shift Theorem to compute all possible translations between pairs of 3D images, ensuring the best overlap in terms of cross-correlation. This approach avoids the propagation of errors from consecutive registration steps and compensates for brightness differences between tiles using a smooth, non-linear intensity transition. The method is fast, works on both 2D and 3D images, and does not require prior knowledge about the tile configuration for small image sets. The implementation is available as an ImageJ plugin as part of the Fiji project. The results demonstrate that the method reconstructs images without stitching artifacts and can handle mosaics with unknown tile configurations. The program is fully multi-threaded and applicable to various types of microscopy image data, making it a valuable tool for analyzing large biological specimens.The paper presents a method for globally optimal stitching of tiled 3D microscopic image acquisitions, which is crucial for high-resolution imaging of large biological specimens. The authors, Stephan Preibisch, Stephan Saalfeld, and Pavel Tomancak, developed a method based on the Fourier Shift Theorem to compute all possible translations between pairs of 3D images, ensuring the best overlap in terms of cross-correlation. This approach avoids the propagation of errors from consecutive registration steps and compensates for brightness differences between tiles using a smooth, non-linear intensity transition. The method is fast, works on both 2D and 3D images, and does not require prior knowledge about the tile configuration for small image sets. The implementation is available as an ImageJ plugin as part of the Fiji project. The results demonstrate that the method reconstructs images without stitching artifacts and can handle mosaics with unknown tile configurations. The program is fully multi-threaded and applicable to various types of microscopy image data, making it a valuable tool for analyzing large biological specimens.