Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

< Back to Article

Fig 1.

Framework diagram of the proposed method.

More »

Fig 1 Expand

Fig 2.

Illustration of intensity differences in overlapping regions.

More »

Fig 2 Expand

Fig 3.

Superpixel merging process based on the RAG.

(a) Adjacency relationships among superpixels. (b) Initial region adjacency graph. (c) Region adjacency graph after merging.

More »

Fig 3 Expand

Fig 4.

Voronoi diagram construction process.

(a) Seed points. (b) Delaunay triangulation. (c) Voronoi diagram construction. (d) Voronoi diagram.

More »

Fig 4 Expand

Fig 5.

AVDO Illustration.

More »

Fig 5 Expand

Fig 6.

Illustration of Valid Regions.

(a) Orthorectified SAR image. (b) Contour lines of valid regions.

More »

Fig 6 Expand

Fig 7.

Illustration of Overlapping Area Extraction.

(a) Searching Process. (b) Overlapping Area.

More »

Fig 7 Expand

Fig 8.

Skeleton Lines of Overlapping Regions.

More »

Fig 8 Expand

Fig 9.

Initial seamline network among multiple SAR images.

More »

Fig 9 Expand

Fig 10.

Optimization of Network Nodes.

(a) Intersection points of skeleton lines in overlapping regions. (b) Optimized network nodes.

More »

Fig 10 Expand

Fig 11.

Segmentation results of the PB-SLIC algorithm.

(a) Segmentation result of image A. (b) Segmentation result of image B. (c) Overlay of the two segmentation results.

More »

Fig 11 Expand

Fig 12.

Segmentation results of the proposed algorithm.

(a) Segmentation result of image A. (b) Segmentation result of image B.

More »

Fig 12 Expand

Fig 13.

Results of Superpixel Merging and Candidate Seamline Extraction.

(a) Before merging. (b) After merging. (c) Candidate seamlines.

More »

Fig 13 Expand

Fig 14.

Extraction of the Optimal Seamline.

(a) Left image, with the blue area representing the overlapping region. (b) Right image, with the blue area representing the overlapping region. (c) Cost matrix. (d) Candidate seamline. (e) Refined result: green curves indicate candidate seamlines, black dots represent network nodes, and red curves indicate the optimal seamline. (f) Illustration of the optimal seamline overlaid on the images to be mosaicked: red curves show the optimal seamline, and the blue area represents the overlapping region.

More »

Fig 14 Expand

Fig 15.

Data Coverage.

More »

Fig 15 Expand

Table 1.

Data Information.

More »

Table 1 Expand

Fig 16.

Overall Results.

(a) Initial seamline network. (b) Result of the method proposed by Pan et al. (c) Result of the method proposed by Wang et al. (d) Result of the proposed method in this study.

More »

Fig 16 Expand

Fig 17.

Local Area Enlargements.

(a1–h1) Enlarged local details of the initial seamline network, corresponding to the red boxes in Fig 16(a). (a2–h2) Enlarged local details of the results from Pan et al.’s method, corresponding to the red boxes in Fig 16(b). (a3–h3) Enlarged local details of the results from Wang et al.’s method, corresponding to the red boxes in Fig 16(c). (a4–h4) Enlarged local details of the results from the proposed method in this study, corresponding to the red boxes in Fig 16(d).

More »

Fig 17 Expand

Table 2.

Quantitative Metric Results.

More »

Table 2 Expand