A　quantitative　detection　method　for　the　buried　depth　of　internal　defects　was　proposed　for　the　detection　of　internal　defects　in　additive　manufacturing　parts.　The　wavelet　packet　decomposition　technology　was　used　to　separate　and　extract　the　ultrasonic　longitudinal　waves　in　the　signals　received　by　the　laser　ultrasonic　nondestructive　testing　technology,　which　solvesd　the　problems　that　the　coupling　of　ultrasonic　surface　wave　and　longitudinal　wave　affected　the　time-domain　feature　extraction.　According　to　the　changes　of　the　longitudinal　wave　sound　path　of　the　defects　during　the　detection　processes,　the　quantitative　detection　of　the　buried　depth　of　the　internal　defects　in　the　precision　forging　samples　was　realized,　and　the　relative　error　of　the　tests　is　1.81%.　An　abnormal　point　filtering　algorithm　was　added　on　the　basis　of　the　original　detection　method,　and　it　was　applied　to　the　detection　of　internal　defects　of　manufacturing　additive　parts　with　the　relative　error　of　detection　of　1.76%.　©　2022,　China　Mechanical　Engineering　Magazine　Office.　All　right　reserved.