A CAD System for Shoe Last Customization Ning Shi School of Business Sun Yat-sen University Guangzhou 510275, P.R. China
Shuping Yi College of Mechanical Engineering Chongqing University Chongqing 400030, P.R. China
Shuping Xiong* School of Mechanical Engineering Shanghai Jiao Tong University Shanghai 200240, P.R. China
[email protected]
Zuhua Jiang School of Mechanical Engineering Shanghai Jiao Tong University Shanghai 200240, P.R. China
Abstract
Therefore, there is a need of a system which can make customized shoes. The shoe last, a solid mold around which a shoe is made, is the “heart” of shoemaking since it mainly determines the shoe shape, fashion, fit and comfort qualities [1-2]. The back part of the shoe last is for fit and comfort while the toe part is mainly for fashion and style [2, 6]. Once the shoe last has been made, other shoe components (shoe upper, outsole, midsole, insole and the heel etc) can be made afterwards. Considering the great importance of the shoe last, this study focuses on the design of customized shoe last. In the traditional shoe manufacturing system, to make a pair of customized shoes is expensive, timeconsuming and complicated due to constraints imposed by manual measuring of several dimensions of a specific foot and manual manufacturing of a shoe last to fit the specific foot dimensions through a trial-anderror approach [3]. In recent years, with the rapid development of computer technology and advanced design and manufacturing technologies such as CAD/CAM, to automate the manufacturing process of customized shoe lasts becomes possible [3, 7, 8]. The objective of this paper is to propose a CAD system which can be used in shoe industries for shoe last customization.
Customized shoes are desired in shoe industries nowadays to satisfy customer needs on style, fit and comfort. This paper proposes a CAD system for designing the customized shoe last based on the customer preferred shoe style and his/her foot features. Eighteen important foot features are firstly extracted from the laser-scanned foot data. The features are then used to deform the base shoe last of the customer preferred style to the customized shoe last with better fit to the customer’s foot, while maximally maintaining the customer preferred style. Finally, the fit between the customer’s foot and the customized shoe last is evaluated through a color-coded map.
1. Introduction Previous studies have shown that ill-fitting shoes are the primary cause of various foot disorders [1-2]. In the shoemaking industry, the shoe is categorized by the length (sometimes the length and width) for the customer to select. However, to select a pair of shoes to fit a person’s feet, the fitting should include more than just foot length since different people have different foot shapes (wide vs. narrow; slim vs. fat; high-arched vs. low-arched), even though they may have the same foot length [3]. Often, a customized shoe is needed, especially for the person whose foot shape is not normal. Additionally, there is a trend among the shoe manufacturers to advance the shoe customization so that the customers’ satisfaction level and the manufacturer’s competiveness can be improved [4, 5].
2. CAD system: overall design Since the shoe style/fashion is generally the first element that attracts customers [3] and the good footshoe fit is very important for foot comfort and health [6, 9-10], the primary aim of the CAD system is to design
customized shoe lasts based on the customer preferred shoe style and his/her own foot features for good fit. 1． ．System input Customer’s foot
Preferred style, color etc
Laser scanner
3. CAD system: detailed design 2． ．CAD system
Digitized foot
Foot feature extraction
Base shoe last
Foot feature based shoe last customization
No
the base shoe last of the customer preferred style to the customized shoe last based on the extracted foot features, while maximally preserving the style of the base shoe last; (3) a color-coded map for final evaluation of the fit/match between the customer’s foot and the customized shoe last. The details of each module are described in section below.
Fit to foot？ Yes 3． ．System output
Digital customized shoe last for CNC processing and physical prototype Figure 1. Framework of CAD system for shoe last customization To achieve aforementioned aim, the CAD system begins with the customer’s selection of the preferred style (including color and shoe materials) from the huge digital database of shoes; at the same time, the customer’s two feet will be laser scanned through a YETITM foot scanner within a few seconds [4]. After two inputs from the customer (preferred style, laser scanned foot data) have been collected, the customer can leave the store and then the customized shoes will be delivered to his/her mail address within a few weeks. Based on aforementioned two inputs, the CAD system will design the customized shoe last through the procedure shown in Figure 1. Visual C++ and OpenGL have been used for the development of CAD system and it consists of three main modules (Figure 1): (1) automatic extraction of 18 important foot features from the laser-scanned foot; (2) a global grading together with the local deformation approach that can deform
3.1. Module 1: foot feature extraction Once the customer selects his/her preferred shoe style from the digital shoe database, the base shoe last (Chinese size 40) corresponding to the shoe with the preferred style has been determined since the shoe last and the shoe has a one-to-one mapping. However, the base shoe last probably does not fit to the customer’s foot in terms of both size and shape, thus, the base shoe last should be reconstructed to match the customer’s foot. In order to do this, the key features of that foot should be firstly extracted out from the laser scanned foot data, which consists of around 90,000 threedimensional points for foot size of 40 (Figure 2). In this study, 18 key foot parameters have been identified from the design requirements for a customized shoe last [7,10], they are: 5 lengths (foot length, arch length, heel to medial malleolus, heel to lateral malleolus, heel to fifth toe), 4 widths (foot width, heel width, bimalleolar width, midfoot width), 3 heights (medial malleolus height, lateral malleolus height, midfoot height), 6 girths (ball girth, instep girth, long heel girth, short heel girth, ankle girth, waist girth). The 18 foot dimensions are primarily needed for the shoe last grading. Since the calculation of lengths, widths and heights depends on the measuring-axis which can be affected by the customer’s foot orientation during scanning, an automatic alignment was first applied on the scanned foot data to adjust the foot orientation [10]. As to the girth measurements, we proposed to use a 2D convex-hull approach to take care of the non-uniform foot contours and to simulate the manual measuring procedure in shoe industry [5]. A convex hull (Hconvex) is the smallest convex set containing a set (P) of discrete points (pi) (see Figure 3), which can be mathematically represented by the following equation: k k H convex ( P) = ∑ α i pi pi ∈ P, α i ∈ R, α i ≥ 0, ∑ α k = 1 i =1 i =1 More detailed algorithms are omitted in this paper, interested readers can refer our published papers [5, 10].
Figure 2. Laser scanned foot data with a cloud of points arranged in several parallel slices with an interval of 1mm
regions on the shoe last match with the corresponding foot regions. Thus, the second step, named as the local deformation, is needed for achieving the good footshoe fit in all regions. After the proper alignment of the 3D foot and scaled shoe last, NURBS (NonUniform Rational B-Spline), one of the most popular methods for designing complex surfaces in CAD and computer graphics [11], is applied to modify local regions/surfaces of the shoe last interactively by moving the control points on NURBS surface. In the second step, the toe part of the shoe last generally is not changed to maximally keep the customer preferred style since the toe part is mainly responsible for the shoe style [2, 6]. A typical shoe last before and after reconstruction is shown in Figure 4.
Figure 3. A point set (a) and its 2D convex hull (b) for foot girth calculation
3.2. Module 2: foot-feature based shoe last customization After successfully extracting 18 foot parameters (features), the design parameters for the ideal shoe last can be established based on the relationship between the foot and shoe last according to the national standard and the industrial guidance for shoe design. The design parameters are then used to guide the reconstruction of the customized shoe last from the base shoe last. This was done through a two-step approach. First, a global grading was applied on the base shoe last to scale it in order to minimize weighted differences between the ideal design parameters (Xi foot +Ci ) and shoe last i.e., parameters after scaling (Xi shoe last), N
Min ∑W i * i =1
X i shoe last −（X i foot + Ci） X i foot
(where Xi shoe last and Xi foot are parameters of the shoe last and the customer’s foot ； Wi is the relative importance for foot-shoe fit from different shoe last design parameters ； Ci is the tolerance between the shoe last design parameter and the foot parameter). Except the change on size of the shoe last to match the foot form as close as possible, the global grading does not change the shape and the style of the shoe last, i.e., this step will exactly preserve the customer preferred shoe style. However, this step may not guarantee all the
Figure 4. Deform the base shoe last (a) into customized shoe last (b)
3.3. Module 3: foot-shoe last fit evaluation Prior to the output of the digital customized shoe last to CNC machine for physical prototype, the final fit between the scanned foot and the customized shoe last should be evaluated. In this study, a color-coded
map is adopted to directly show the mismatch between the foot and the customized shoe last (Figure 5). The mismatch is defined as the dimensional difference between the point on the shoe last and its nearest foot region along the norm of the last surface after proper alignment. The dimensional difference has the sign (+/-) as well since the tightness or looseness of the shoe last relative to the customer’s foot has quite different implications [3]. The positive (+) dimensional difference indicates where the last surface is loose for the foot, while the negative (-) dimensional difference indicates where the last surface is tight for the foot. If the dimensional differences in some shoe last regions exceed the specified tolerances, the local deformation should be further applied until all the dimensional differences are within the acceptable ranges. Once the color map shows that shoe last fits the foot well, then the digital shoe last can be outputted to the CNC machine for making the physical customized shoe last.
shoemaking industry to design a pair of fashionable and well-fitting customized shoes for the customer in a short time with reasonable cost, thus achieving customer’s high satisfaction level as well as manufacturer’s commercial success.
References [1] Clarks Ltd. Training Deparment, Manual of shoe making, Training Department Clarks, UK, 1989. [2] W.A Rossi and R Tennant, Professional shoe fitting, Pedorthic Footwear Association, New York, 2000. [3] J Leng and R Du, A CAD Approach for designing customized shoe last, Computer Aided Design and Applications, 3 (2006), 377-384. [4] Vorum Research Corporation, User manual of canfitplusTM YetiTM foot scanner, Canada, 2000. [5] J. Zhao, S. Xiong, Y. Bu and R. S Goonetilleke, Computerized girth determination for custom footwear manufacture, Computers and Industrial Engineering, 54(2008), 359-373. [6]
S. Xiong, R.S Goonetilleke, C.P Witana and E.Y.L Au, Modelling foot height and foot shape related dimensions, Ergonomics, 51(2008), 1272-1289.
[7] C Xu, Y Liu, Y Jiang and Y Pan, Design and realization of customized shoe last CAD system, Journal of Computer-Aided Design & Computer Graphics, 16 (2004),1437-1441 (In Chinese). [8] M Mochimaru, M Kouchi and M Dohi, Analysis of 3D human foot forms using the free form deformation method and its applications in grading shoe lasts, Ergonomics, 43 (2000), 1301-1313.
Figure 5. The foot-shoe last alignment (upper figure) and the mismatch (dimensional difference) color coded on the last surface (lower figure)
4. Conclusions In this paper, a CAD system for shoe last customization has been proposed. The basic approach is to deform the base shoe last with the customer preferred style into the customized shoe last that fits the scanned foot data based on the customer’s foot features, while maximally preserving the style of the base shoe last. With the continuous improvements on this CAD system, it has great potential to be applied into the
[9] S Xiong, R.S Goonetilleke, J Zhao, W Li and C.P Witana, Foot deformation under different load bearing conditions and their relationships to stature and body weight, Anthropological Science (Accepted). [10] C.P Witana, S Xiong, J Zhao and R.S Goonetilleke, Foot measurements from three-dimensional scans: A comparison and evaluation of different method, International Journal of Industrial Ergonomics, 36, (2006), 789-807. [11] S.M Hu, Y.F Li, T Ju and X Zhu, Modifying the shape of NURBS surfaces with geometric constraints, Computer-Aided Design, 33 (2001), 903-912.
