Seam line automatic generation method based on Dijkstra algorithm with weight

Abstract

The invention provides a seam line automatic generation method based on a Dijkstra algorithm with weight. The seam line automatic generation method comprises the following steps that S1 an overlapped area between two adjacent images is calculated according to a digital ortho-image through ortho-rectification; S2 a Dijkstra algorithm reference point is calculated according to the overlapped area, and a digraph is generated; S3 weight parameters are set, the digraph shortest path is calculated through the Dijkstra algorithm based on the weight parameters, and a seam line is generated and displayed in real time; S4 the seam line is optimized, and a mosaic surface is generated and correlated with the images; and S5 the images correlated with the mosaic surface are outputted to the designated position according to control information. The seam line is extracted through the Dijkstra algorithm with the weight so that manmade objects can be effectively avoided by extraction of the seam line, color difference of water areas can be eliminated, large amount of manual editing work can be saved, the degree of automation of extraction of the seam line can be further enhanced and the extracted seam line is ensured to have great accuracy.

Description

technical field [0001] The invention relates to the technical field of image data processing, in particular to a mosaic line automatic generation method based on weighted Dijkstra algorithm. Background technique [0002] At present, the relevant algorithm for automatically generating mosaic lines for orthorectified images is very mature, which greatly improves the efficiency of image mosaic compared with the traditional method of manually drawing mosaic lines. [0003] For the field of image mosaicking, the research focus is mainly on how to further optimize the mosaic line algorithm. For example, the new generation of grid-based digital photogrammetry systems such as DPGRIDMapping mainly use the mosaic line automatic extraction algorithm based on the ant colony algorithm To extract the mosaic line, through this algorithm, the point whose gray value is greater than the threshold on the difference image of the overlapping area is regarded as the possible obstacle area. Optim...

AI Technical Summary

Problems solved by technology

[0004] However, compared with the traditional method of manually drawing mosaic lines, the current method of automatically generating mosaic lines saves the workload of manual editing to a large extent and makes the extraction of mosaic lines faster. , so the effect of the extracted mosaic line is not good, and a certain amount of manual editing is still required to optimize the mosaic line extraction; at the same time, the automatic mosaic line extraction method based on the ant colony algorithm still has thresholds and search times in the process of its implementation. and algorithm convergence is difficult to determine and other issues

[0005] Moreover, due to the complex environment in the image, that is, the image contains both water body data, surface data and artificial feature data, when the mosaic line is extracted, the mosaic line cannot effectively avoid the artificial features, and when the mosaic When the line crosses images in different seasons, there will be color differences in the water body, so that the extraction of the mosaic line still requires manual editing work

Images