To obtain a seamless remote sensing image with wide field-of-view and high resolution, this study proposes a remote sensing image mosaic algorithm based on distribution measure and saliency information. Outlier removal, optimal seamline detection, and smooth transition during image mosaic are the main areas of improvement. To improve image alignment, the optimal inliers are first chosen based on the distribution quality of inliers in the overlapping area. Secondly, the saliency information of images is determined by line segments and guided filter to avoid the seamline crossing obvious features. Lastly, a two-scale image fusion is performed using a guided filter, and spatial consistency is used to achieve a smooth transition between the images on both sides of the seamline. The simulation results demonstrate that, when compared with the RANSAC algorithm, the proposed algorithm enhances the mutual information by 1.93% and the stability by 46.55% in outlier removal. In comparison to the QESE algorithm, the proposed algorithm improves structural similarity (SSIM) by 3.21% and peak signal-to-noise ratio (PSNR) by 2.55% in optimal seamline detection and smooth transition. A high-quality, wide field-of-view, and high-resolution remote sensing image is produced with uniform brightness and no ghosting.