In this paper, the detection methods and evaluation systems for material surface cleanliness after laser cleaning are systematically reviewed. These encompass various online detection techniques such as laser-induced breakdown spectroscopy (LIBS), fluorescence spectroscopy, reflectance spectroscopy, acoustic signal analysis, and image recognition, as well as offline methods like wettability testing, surface morphology observation, and elemental analysis. The research finds that these diverse methods can effectively identify the characteristic stages of insufficient cleaning, complete cleaning, and excessive cleaning during the cleaning process, with distinct signal features reflecting the evolution of the cleaning state. The collaborative analysis of multi-source detection information aids in enhancing the accuracy of state recognition and process monitoring. At present, there are differences among these methods in terms of response mechanisms, evaluation dimensions, and application scopes. Techniques such as LIBS, spectroscopy, and acoustic signal analysis are more suited for laboratory conditions, whereas image acquisition and analysis, as well as roughness and wettability assessments, are more practical for on-site industrial monitoring. Each detection method yields distinct results with its own characteristics. Future research can further explore areas such as multi-modal integration, cross-scale modeling, and the establishment of standardized systems, thereby advancing laser cleaning quality detection towards higher precision, intelligence, and engineering applicability. This progression will help meet the technical demands for surface cleanliness control in aerospace manufacturing and other advanced industrial fields.