To overcome the limitations of traditional measurement techniques and achieve quantitativemeasurement of the compressibility of liquids, this sludy combines ulrasonic characterization technology andschlieren imaging technology to design and implement a high-precision, non-contact liquid compressibilitymeasurement system. This system utilizes advanced computer image processing technology to signiicantlyenhance the accuracy of experimental data and effectively eliminate external interference during themeasurement proeess.Further, this sludy introduces machine learing algorithms to deeply analyzes the intrinsicrelationship between the compressibility of sodium chloride solutions and their density and ultrasonicwavelength, and constructs corresponding, predictive models.These research results provide new experimentalmethods for in-depth studies of liquid physical properties, and at the same time offer solid data support andheoretical foundations for scientifie research and practical applications in related fields.