Model based foot shape classification using 2D foot outlines

doi:10.1016/j.cad.2011.01.005

Computer-Aided Design

Volume 44, Issue 1, January 2012, Pages 48-55

https://doi.org/10.1016/j.cad.2011.01.005 Get rights and content

Abstract

This study introduces a novel technique to identify foot outline characteristics and to classify feet into groups using turning functions and clustering techniques so that shape can complement anthropometry in producing good fitting shoes. The digital 3D foot scans, obtained from 50 Hong Kong Chinese subjects (25 males and 25 females) were processed to generate the foot outlines at heights of 2 mm and 40 mm. The outlines were represented as turning functions and the similarity among shapes was determined using average linkage clustering. The results show that there are two distinct shape groups for the 40 mm foot outlines on both medial and lateral sides of the foot. The presence (46%) or absence (54%) of a medial bulge characterizes the medial side, while the two shape groups on the lateral side are mainly due to the lateral concavity in the mid-foot region. The group with a lateral concavity consists of more females (68%) and thus lateral side of foot outline appears to be gender related. Furthermore, the medial and lateral side clusters are not related to each other. The medial side shape from the 2 mm foot outline is a good indicator of fallen arches. Based on the analyses, four types of feet were identified: feet with (1) lateral concavity and a medial bulge, (2) a medial bulge and no lateral concavity, (3) lateral concavity and no medial bulge and (4) lateral concavity and a medial bulge. These shape differences can be useful in the design of shoe lasts and in the manufacture of compatible footwear so that trial and error fitting can be minimized.

Research highlights

► Turning function and clustering can differentiate foot shape characteristics. ► Medial and lateral side outlines differentiate distinct groups. ► Medial clusters are independent of the lateral side clusters. ► Medial side outline characterizes the arch. ► Females exhibit concavity on the lateral side outline of the rearfoot region.

Introduction

A trend towards mass customized footwear for military personnel, health care workers and even for high-performance sportsmen such as skiers is common today. Such customization requires that good fitting lasts be designed and manufactured to acceptable manufacturing standards. However, the process of generating a customized last has been through trial and error with a series of fitting trials to achieve the required fit. This is primarily because, in most instances, an existing last is transformed to be similar to the person’s foot dimensions and the various intricacies of a foot are neglected. Custom-made footwear requires the dimensions as well as the characteristic features of a foot to be identified.

Characteristics such as arch angle, arch height, rearfoot alignment, presence or absence of a talonavicular bulge and navicular drift, and their similarities and differences among people are well-known [1] even though not all are used in the design of footwear or shoe lasts. The variations in the arch are used to design straight or flared shoes; rearfoot alignments are accommodated with hard or soft or post-type of cushioning and so on. To account for the differences in ball of foot length on the lateral side of foot, Wunderlich and Cavanagh [2] suggested that the fifth metatarsal head expansion in a woman’s last should be at a more proximal location than in a man’s last. Thus, identifying the similarity characteristics and classifying them into groups can be useful in the design of and mass customization of footwear for improved comfort [3], [4], [5], [6], [7], [8], [9]. Unfortunately, most of the identified differences among feet have been primarily on the forefoot. Some researchers [10] have claimed that differences in the rearfoot are minimal and suggested that the rear-part of foot lasts can be the same for any given size. This study investigates shape differences among feet and between genders so that shape can supplement anthropometry in the design of lasts so that footwear can be designed and fitted right the first time.

Identifying and classifying foot shapes can have a significant impact on last design [11], [12], [13] and the treatment or prevention of foot related disorders [14]. Techniques that can be used to classify shapes include turning functions [15], Fourier descriptors [16], autoregressive coefficient [17], stochastic labeling [18], convolution [19], curve bending function and variations [20], etc. Turning functions are an efficient method for “free” curve matching especially for complex and open shapes [21] that are convex as well as concave [15], [22], [23].

Thus the objective of this study was to identify the significant characteristics on the lateral and medial sides of the foot.

Section snippets

Participants

Fifty students (25 males and 25 females) from the Hong Kong University of Science and Technology were recruited and consented to participate in this study, which was approved by the institutional research ethics committee. The descriptive statistics of the participants are given in the Table 1. None of them had any visible foot abnormalities or foot illnesses.

Foot shape determination

The YETI $^{TM} 1$ foot scanner was used to obtain the 3D point cloud of each participant’s foot surface. The scan sections were set to be 1 mm

Analysis and results

A cluster analysis was performed on the similarity indices for the medial and lateral side of 2 and 40 mm foot outlines separately. The number of combinations for the distance comparisons were $^{50} C_{2} = 1225$ , as there were fifty subjects and the dissimilarity calculations were between any two subjects. The descriptive statistics of the distance functions among subjects are in Table 2 and it was found that the distance functions follow a normal distribution (Fig. 6). The maximum distance is highest

Discussion

Feet can be classified in many different ways, but most methods require extensive data gathering and analysis [37]. The pattern recognition method adopted here is easy to use, especially, with the availability of many types of scanners and imaging methods these days.

The distance functions among subjects showed a normal distribution for each type of outline. The medial side of the 2 mm outline showed higher distances in the distance function matrix (mean = 0.2299, SD = 0.0759) than other

Acknowledgement

The authors would like to thank the Research Grants Council of Hong Kong for funding this study under grant HKUST 613008.

References (38)

M. Razeghi et al.
Foot type classification: a critical review of current methods
Gait and Posture
(2002)
T.J. Hwang et al.
Derivation of template shoe-lasts for efficient fabrication of custom-ordered shoe-lasts
Computer-Aided Design
(2005)
M.S. Rathleff et al.
Perspectives for clinical measures of dynamic foot function—reference data and methodological considerations
Gait & Posture
(2010)
P.W. Huang et al.
Planar shape recognition by directional flow-change method
Pattern Recognition Letters
(1999)
A.A. Eftekharian et al.
Distance functions and skeletal representations of rigid and non-rigid planar shapes
Computer-Aided Design
(2009)
S. Hou et al.
Structure-oriented contour representation and matching for engineering shapes
Computer-Aided Design
(2008)
S.R. Urry et al.
The accuracy of footprint contact area measurements: relevance to the design and performance of pressure platforms
The Foot
(2001)
D. McFadden et al.
Relative lengths of fingers and toes in human males and females
Hormones and Behavior
(2002)
J. Fu et al.
Shape differentiation of freeform surfaces using a similarity measure based on an integral of Gaussian curvature
Computer-Aided Design
(2008)
S.E.M. Salhieh
A methodology to redesign heterogeneous product portfolios as homogeneous product families
Computer-Aided Design
(2007)

P.R. Cavanagh et al.

The arch index: a useful measure from footprints

Journal of Biomechanics

(1987)

Wunderlich RE, Cavanagh PR. External foot shape differences between males and females and among races. Abstract book,...

M.R. Hawes et al.

Ethnic differences in forefoot shape and the determination of shoe comfort

Ergonomics

(1994)

B.M. Nigg et al.

Shoe inserts and orthotics for sport and physical activities

Medicine & Science in Sports & Exercise

(1999)

Goonetilleke RS, Luximon A, Tsui KL. The quality of footwear fit: what we know, don’t know and should know. In:...

M. Dohi et al.

Foot shape and shoe fitting comfort for elderly Japanese women

Japanese Journal of Ergonomics

(2001)

Goonetilleke RS, Luximon A. Designing for comfort: a footwear application. In: Proceedings of computer-aided ergonomics...

C.P. Witana et al.

Dimensional differences for evaluating the quality of footwear fit

Ergonomics

(2004)

F.T. Cheng et al.

A systematic approach for developing a foot size information system for shoe last design

International Journal of Industrial Ergonomics

(1999)

Cited by (20)

Morphological analysis of pottery based on the turning function using photogrammetry
2023, Journal of Archaeological Science: Reports
Geometric morphometry is a method used to analyse the morphological characteristics of research objects and has many applications in biological analysis. With the development of 3D digital technology, geometric morphometry is widely used on 3D models. The pottery samples were obtained from the Ganguya site (1850 and 1500 BCE) in Gansu Province. Three-dimensional models of the pottery were obtained by photogrammetry, and the outline curves of the pottery were extracted. The outline curves were converted into angles by a turning function, and four characteristic indices were extracted according to the features of the angle function curve. Two principal components were extracted, and the samples were divided into three categories by the K-means clustering algorithm. The characteristics and differences of the three categories were analysed. The turning function provides a new method for generating an outline in geometric morphometry and a new perspective for the study of pottery morphology.
Subject-specific identification of three dimensional foot shape deviations using statistical shape analysis
2020, Expert Systems with Applications
Citation Excerpt :
For some people, this pain is linked to foot deformities, with common conditions including hallux valgus (Garrow et al., 2001; Nix, Vicenzino, Collins, & Smith, 2012), collapsed foot arches (Xiong, Goonetilleke, Witana, Weerasinghe, & Au, 2010; Young, Niedfeldt, Morris, & Eerkes, 2005), and club feet (Agarwal & Rastogi, 2018; Ganesan, Luximon, Al-Jumaily, Balasankar, & Naik, 2017). For others, foot pain has been associated with improperly fitting footwear (de Castro, Rebelatto, & Aurichio, 2010; Dobson, Riddiford-Harland, Bell, & Steele, 2018), indicating that a more precise characterization of foot shape would be valuable in footwear fitting and design (Deselnicu, Vasilescu, Mihai, Purcarea, & Militaru, 2016; Rodrigo, Goonetilleke, & Witana, 2012; Sarghie, Mihai, & Herghiligiu, 2016; Wunderlich & Cavanagh, 1999). Despite a clear link between foot shape and foot pain, one study has reported that more than half of its participants who experienced debilitating foot pain did not seek professional help (Garrow et al., 2004).
The high prevalence of foot pain, and its relation to foot shape, indicates the need for an expert system to identify foot shape abnormalities. Yet, to date, no such expert system exists that examines the full 3D foot shape and produces an intuitive explanation of why a foot is abnormal. In this work, we present the first such expert system that satisfies those goals. The system is based on the concept of model-based outlier detection: a statistical shape model (SSM) is generated from 186 3D optical foot scans of healthy feet. This model acts as a knowledge base which is then parameterized by one’s demographic characteristics (e.g., age, weight, height, shoe size) through a multivariate regression. This regression introduces model flexibility as it allows the model to be fine tuned to a specific individual. This fine tuned model is then used as a baseline to which the individual’s 3D foot scan can be compared using standard statistical tests (e.g. t-tests). These statistical tests are performed at each vertex along the foot surface to identify the degree and location of shape outliers. Our expert system was validated on foot scans from patients with hallux valgus and abnormal foot arches. As expected, our results varied per patient, confirming that feet with the same clinical classification still show high shape variability. Additionally, the foot shape abnormalities identified by our system not only agreed with the expected location and severity of the tested foot deformities, but our analysis of the full 3D foot shape was able to completely characterize the extent of those abnormalities for the first time. These results show that the combination of statistical shape modelling, multivariate regression, and statistical testing is powerful enough to perform outlier detection for 3D foot shapes. The resulting insights provided by this system could prove useful in both shoe design and clinical diagnosis.
Study on the foot shape characteristics of the elderly in China
2017, Foot
Citation Excerpt :
Asanka et al. measured the outline of the foot, and divided the foot into four types. They processed data through the methods of clustering and rotation analysis, and supplemented the morphological tissue characteristics of the foot for the design of footwear and shoe lasts [21]. Golightly et al. studied ethnic-related differences in the incidence of foot diseases, flatfoot deformity, and high arch foot, and obtained prevalent foot disease types among the subjects over the age of 50 [6].
With aging, the feet of the elderly above 60 years old in China present degenerative changes, deformities, and diseases, which significantly affect their daily activities.
The authors aimed to study the morphological characteristics of the feet and identify the foot type according to size (length and width) and defect characteristics of elderly feet in China.
A convenient sample of 1000 subjects above 60 years old was recruited mainly in the regions of Shanghai, Shaanxi, Henan, Hebei, and Sichuan in China. Foot images were collected, and 800 (male 398, female 402) valid questionnaires were recovered. A total of 800 elderly subjects as the test group were invited to measure their foot sizes by means of a Footprint Collector (Tong Yuan Tang Health Management Limited, Qingdao in Shandong province). The foot type of the elderly was compared with that of the general adult Chinese population as the control group using the t-test for independent samples.
Hallux valgus (46.9%) and flat foot (50.0%) were the most common foot shape deformities. The most frequent foot diseases were foot scaling (91.2%) and calluses (96.3%). The medial width of the first metatarsal-toe joint of the elderly was significantly higher (elderly female, 44.95 ± 4.86 mm; elderly male, 48.55 ± 4.94 mm) than that of the general adult population (adult female, 40.18 ± 3.43 mm; adult male, 43.22 ± 3.20 mm) (p < 0.01).
The foot length of the elderly was not significantly different from that of the general adult Chinese population. The width of the first metatarsal-toe joint in the forefoot of the elderly was significantly higher than that of the general adult Chinese population, which was consistent with the result that a high proportion of elderly subjects presented hallux valgus.
Classification of the Elderly Foot Types Based on Plantar Footprints
2017, Procedia Engineering
The article presents the methodology and the results of the studies conducted in order to develop a rational classification technique of the elderly foot typologies, based on parameters derived from the plantar footprint. The plantar footprints were taken from 67 women, aged between 52-84 years old with RSscan pressure plate and the associated system, Footscan 7 Gait, 2nd Generation, 0,5 Gait Scientific System. To classify the foot typology, Chippaux-Simark Index and Hallux-Valgus Angle have been used. According to the obtained results, the subjects were divided into five categories, subjects with Normal Foot, High Arched Foot, Flat Foot, Hallux-Valgus Foot and Hallux Varus Foot. The authors have developed a method of determining the middle area of a plantar footprint for a more accurate measurement, being able to register it in the same way, for all subjects. Using statistical analysis, differences between age groups have been found, demonstrating the necessity of modelling and designing customized shoe lasts, footwear and prophylactic components according to the age of the subject. A good fit of the shoe on the elderly foot involves modifying the shoe last, but this can only be achieved under certain standards. Having biomechanical data of the foot and a clear classification of it, the customization of the product is simplified, reducing the number of tests and increasing the footwear technological process efficiency.
A framework for geometry acquisition, 3-D printing, simulation, and measurement of head-related transfer functions with a focus on hearing-assistive devices
2016, CAD Computer Aided Design
Citation Excerpt :
This model was then used to synthesize plausible body shapes as input to product design. In another example, foot shape was investigated in several studies driven by the large footwear industry and recently, 50 surface scanned feet were used as input to a statistical analysis of shape with the goal to produce optimized shoe lasts [27]. Similarly, in the design of HADs, the position of the microphones is essential, not only for the design of the casings, but also for capturing spatial acoustic cues.
Individual head-related transfer functions (HRTFs) are essential in applications like fitting hearing-assistive devices (HADs) for providing accurate sound localization performance. Individual HRTFs are usually obtained through intricate acoustic measurements. This paper investigates the use of a three-dimensional (3D) head model for acquisition of individual HRTFs. Two aspects were investigated; whether a 3D-printed model can replace measurements on a human listener and whether numerical simulations can replace acoustic measurements. For this purpose, HRTFs were acoustically measured for four human listeners and for a 3D printed head model of one of these listeners. Further, HRTFs were simulated by applying the finite element method to the 3D head model. The monaural spectral features and spectral distortions were very similar between re-measurements and between human and printed measurements, however larger deviations were observed between measurement and simulation. The binaural cues were in agreement among all HRTFs of the same listener, indicating that the 3D model is able to provide localization cues potentially accessible to HAD users. Hence, the pipeline of geometry acquisition, printing, and acoustic measurements or simulations, seems to be a promising step forward towards in-silico design of HADs.
Foot volume estimation formula in healthy adults
2014, International Journal of Industrial Ergonomics
Citation Excerpt :
They reported this method with R2 equal to 0.81 with some measurement issues. With the advent of the 3-D scanning technology, 3-D foot form could now be measured with great accuracy, and had been used widely in many foot researches (Witana et al., 2006; Hu et al., 2007; Xiong et al., 2008; Luo et al., 2009; Park et al., 2009; Xiong et al., 2009; Yu and Tu, 2009; Han et al., 2010; Telfer and Woodburn, 2010; Xiong et al., 2010; Yu et al., 2010; Ma et al., 2011; Lee et al., 2012; Rodrigo et al., 2012; Sims et al., 2012; Telfer et al., 2012). Nevertheless, the procedure is still time consuming, and sometimes cumbersome for patients.
Foot volume (FV) is an important measurement in clinics or industrial applications. Direct measurement and anthropometric formula were methods usually used to evaluate FV. However, direct measurement is usually cumbersome for patients. Three anthropometric formulas were proposed to estimate FV, including wedge method, prism method, and figure-of-eight linear regression model, but none of them could estimate FV with good accuracy and ease of use. The purpose of this study is to derive a formula to estimate FV easily and accurately. 3-D foot forms of the right feet of 122 males and 96 females were scanned with good accuracy. FV and nine 1-D foot measurements were extracted from these 3-D foot scans. From these foot scans, a new FV estimation formula based on foot length, ball girth, and instep girth was derived with R² as 0.973. It is of simple and geometrically meaningful form, and is easier to be applied in clinics or industrial applications with accuracy.
Foot Volume (FV) is an important measurement in clinics or industrial applications. This study had provided a simple FV estimation formula with R² as 0.973 by three 1-D foot measurements.

View all citing articles on Scopus

View full text

Model based foot shape classification using 2D foot outlines

Abstract

Research highlights

Introduction

Section snippets

Participants

Foot shape determination

Analysis and results

Discussion

Acknowledgement

Gait and Posture

Computer-Aided Design

Gait & Posture

Pattern Recognition Letters

Computer-Aided Design

Computer-Aided Design

The Foot

Hormones and Behavior

Computer-Aided Design

Computer-Aided Design

Journal of Biomechanics

Ethnic differences in forefoot shape and the determination of shoe comfort

Ergonomics

Shoe inserts and orthotics for sport and physical activities

Medicine & Science in Sports & Exercise

Foot shape and shoe fitting comfort for elderly Japanese women

Japanese Journal of Ergonomics

Dimensional differences for evaluating the quality of footwear fit

Ergonomics

A systematic approach for developing a foot size information system for shoe last design

International Journal of Industrial Ergonomics