7 6 5 4 3 2 1 0 19 9 19 5 9 19 7 9 20 9 0 20 1 0 20 3 0 20 5 0 20 7 0 20 9 1 20 1 1 20 3 1 20 5 17 The Total Number of Publications LS PU methods can be generally divided into two groups: regular subinterferogram-based and irregular subinterferogram-based methods. Figure 16 summarizes the processing flowcharts of these two groups of methods. It is FIGURE 15. Statistics for LS PU publications from 1995 to 2017. apparent that these groups are both based on the divideand-conquer framework. Comparing small- and mediumscale PU methods, both need extra computation to guarantee consistency between local and global PU solutions. To be specific, the methods in the first group use a simple tiling strategy to regularly divide the input interferogram into fixed and rectangular subinterferograms (the size of each subinterferogram is chosen by algorithm users) and unwrap each subinterferogram individually. Then, because the PU result of each subinterferogram is not usually consistent with that of the whole interferogram, a complicated splicing strategy that recognizes and corrects areas exhibiting inconsistency between local and global PU solutions is used to combine the PU results of all of the subinterferograms. The tiling step of the methods in the first group consumes only a little computation, but the computational burden of the splicing step can be very high. As examples of representative PU methods in the first group, [105] used a statistics-based strategy to recognize the areas with inconsistency between local and global LS Input Interferogram Regular Subinterferogram-Based Methods Irregular Subinterferogram-Based Methods Directly Divide LS Input Interferogram Into Fixed and Rectangular Subinterferograms Divide LS Input Interferogram Into Unfixed and Irrectangular Subinterferograms for Keeping Consistency Between Local and Global PU Solutions Unwrap Each Subinterferogram Individually Unwrap Each Subinterferogram Individually Recognize and Correct Inconsistency Between Local and Global PU Solutions, Then Combine All the Local PU Results Directly Combine All the Local PU Results LS PU Result FIGURE 16. The processing flowcharts of the LS PU methods. 52 ieee Geoscience and remote sensing magazine march 2019