Human body type category prediction method for automobile seat and target seat posture acquisition method

• Main Authors: GU LEI, GUO WEILONG, LIU FEI, LI LIANCHAO, CHANG LIN, FU BAO'AN
• Format: Patent
• Language: Chinese; English
• Published: 19.03.2024
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; MEASURING; PERFORMING OPERATIONS; PHYSICS; SEATS SPECIALLY ADAPTED FOR VEHICLES; TESTING; TRANSPORTING; VEHICLE PASSENGER ACCOMMODATION NOT OTHERWISE PROVIDEDFOR; VEHICLES IN GENERAL; VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISEPROVIDED FOR; WEIGHING

The invention discloses a human body type category prediction method for an automobile seat and a target seat posture acquisition method, and relates to the field of automobile seat adjusting.The method comprises the steps that original pressure point cloud data on the seat are acquired on the basis of pressure detection points arranged on the seat when multiple users in the same human body type are located at preset seat positions; preprocessing the pressure point cloud feature data to obtain pressure point cloud feature data, namely sample data; binding a corresponding body type category label on the sample data to obtain a training sample; training an SVM model through the training sample to obtain an SVM target model; the body type category corresponding to the current user is predicted through the SVM target model, so that the height range and the weight range of the user are obtained, namely, the height and the weight of the user are predicted based on the pressure sensor, and the problem that the heigh