손 치수 측정을 위한 3차원 반자동 측정 방법 개발

MasterAnthropometric data is applied to designing product size and shape to develop body fitness and comfort of product. These days, anthropometric data have been obtained through 3D scanner to get human body size and shape together. To utilize the 3D measurement protocol, hand sway and hand skin deformation must be removed during 3D scanning. And the efficiency of landmarking must be evaluated. This research is performed to develop a 3D measurement protocol for hand anthropometry. At first, the spot of landmarks for the 3D scanning measurement is defined. The landmarks defined previously are not appropriate for the 3D hand measurement protocol. Referring to 8 documents, 169 measurement items for hand anthropometry are collected and 97 landmarks are defined. The landmarks are coded to be understood and used more easily. Second, a 3D semi-automatic measurement protocol (3D-SAMP) is developed to overcome the existing disadvantages of the 3D scanning measurement protocol. The 3D-SAMP is composed of 4 steps (1. Producing a hand plaster & 3D scanning, 2. Landmarking, 3. Extracting 3D coordinates, 4. Extracting hand size). Through 3D scanning a hand plaster, the hand sway and deformation could be prevented. The automatic extraction of size based on the landmarks makes the 3D-SAMP effective. Lastly, 20 persons measured 52 measurement items of 1 person’s hand 3 times respectively by 3D-SAMP & manual measurement to evaluate the 3D-SAMP. The two measurements are compared in aspect of mean difference, repeatability, time required and subjective rating. 11 items are significantly different comparing the results of two measurements. The repeatability of a 3D-SAMP (intra measurer SD = 0.70, inter measurer SD = 0.75) is higher than a manual measurement (intra measurer SD = 1.35, inter measurer SD = 1.76). There is no significant difference on the time required to measure 52 items through the manual measurement (mean = 17 min. 49 sec., SD = 4 min. 58 sec.) and the 3D-SAMP (mean = 17 min. 48 sec., SD = 4 min. 31 sec.). The subjective rating (1: very dissatisfied, 7: very satisfied) of the 3D-SAMP (mean = 4.30, SD = 0.82) is higher than the manual measurement (mean = 5.18, SD = 0.83). In summary, the spot of landmarks for a 3D hand measurement protocol are defined and a 3D-SAMP is developed by using a hand plaster and semi-automatic size extraction program. Through 3D-SAMP, 3D hand shape data can be extracted without hand sway and skin deformation