In this article, we present the design and validation of a non-contact scanning system for the development of a three-dimensional (3D) model of moist biological samples. Due to the irregular shapes and low stiffness of soft tissue samples, the use of a non-contact, reliable geometry scanning system with good accuracy and repeatability is required. We propose a reliable 3D scanning system consisting of a blue light profile sensor, stationary and rotating frames with stepper motors, gears and a five-phase stepping motor unit, single-axis robot, control system, and replaceable sample grips, which once mounted onto the sample, are used for both scanning and mechanical tests. The proposed system was validated by comparison of the cross-sectional areas calculated based on 3D models, digital caliper, and vision-based methods. Validation was done on regularly-shaped samples, a wooden twig, as well as tendon fascicle bundles. The 3D profiles were used for the development of the 3D computational model of the sample, including surface concavities. Our system allowed for 3D model development of samples with a relative error of less than 1.2% and high repeatability in approximately three minutes. This was crucial for the extraction of the mechanical properties and subsequent inverse analysis, enabling the calibration of complex material models.