Simultaneously recovering both domain and varying density in inverse gravimetry by efficient level-set methods (Q2037218)

In this article, the authors develop new efficient algorithms for the inverse problems of gravimetry to reconstruct simultaneously both unknown domain and varying density of the anomalous measure from modulus of gravity force rather than from gravity force itself. First they establish various uniqueness theorems of the inverse gravimetry problem and design inversion algorithms to fulfill those uniqueness conditions. Further they implemented a local level-set method and a partitioned truncated singular value decomposition algorithm to improve the computational efficiency. Finally to illustrate the performance of the proposed algorithms, numerical examples are given.