Ultrasonic　tomography　(UT)　has　been　widely　applied　in　industrial　process　imaging.　Image　reconstruction　algorithm　is　one　of　the　most　popular　research　interests　in　UT　field.　In　this　paper,　a　novel　method　for　construction　of　sensitivity　matrix　was　proposed.　The　method　can　be　used　to　improve　linear　back　projection　(LBP)　algorithm　for　detection　of　the　gas　bubble　distribution　in　a　gas/liquid　two-phase　bubbly　flow.　Hybrid　reconstruction　(HR)　algorithm　was　introduced　to　improve　image　reconstruction　performance　based　on　the　presented　method.　Furthermore,　a　series　of　models　have　been　established　to　evaluate　the　performance　of　the　above　algorithms.　Three　quantitative　evaluation　criteria　were　selected　to　evaluate　the　quality　of　reconstructed　images.　The　results　indicate　that　the　fan-shaped　sensitivity　matrix　is　feasible　and　valid　for　the　LBP　algorithm　in　improving　image　quality.