性別以及年齡辨識可以迅速地達成人群的統計，因此與身分辨識相同是十分熱門的議題，又在不同的性別以及年齡之間會因身體的因素如身高、肢體長度以及關節活動度等不同，導致在步態表現上產生差異，且步態數據的收集具非侵入性，並可以在一定距離下測得，因此是性別以及年齡辨識經常使用的方法。
本實驗徵求健康的20歲至30歲之男性與女性各8名以及60歲至70歲之女性8名參與，並要求受測者穿戴一雙裝有F-Scan壓力墊的休閒鞋或運動鞋行走以收集步態過程中壓力中心 (center of pressure, COP) 數據，並基於COP軌跡與人體生物力學之關聯性的知識，選取欲作為支持向量機 (Support Vector Machine , SVM) 分類器輸入的特徵，對不同性別以及年輕女性與老年女性進行辨識。
本研究結果指出，擷取COP特徵並透過徑向基函數核 （Radial basis function, RBF）支持向量機 (RBF-SVM) 做性別與年齡辨識可以得到較高準確率。除此之外，藉由擷取三項關鍵特徵，在性別辨識中可以反應出男性在初次接觸階段的時間比例較長，以及在站立時期有較大的步態變異性與踝關節力矩等特性；在年齡辨識中可以反應出老年女性具較短的時間比例、較快的向前速率，以及較大的重量轉移變異性等特性。在未來的研究中將收集更多受測者的數據以提高模型辨識的穩健性，並透過融合其他類型的特徵以及比較多種分類模型以得到更高的績效。本研究之結果可應用於注重機密無法裝置攝影機之場所(如工業機密研究實驗室)，更可進一步增加監控系統之多樣性與辨識準確率。

Gender and age classifications are a popular topic of human identification and can achieve population statistics rapidly. Gait features could be used for classification because of the differences in height, limb length, range of joint motion between gender and age, which contribute to discriminate gait patterns. Furthermore, the advantages of employing gait for identification are that gait data can collect non-invasive and from a distance.
In the current study, 24 participants (8 young men, 8 young women, and 8 older women) were instructed to wear their own sneakers with F-Scan pressure insoles. Subsequently, the COP data were calculated and extracted as the input features in the SVM classifier. The feature extractions were based on the characteristics of COP trajectory and biomechanics in gait.
The RBF-SVM models of gender and age classifications showed high accuracies via extracting COP features. In addition, the result also revealed that the three key features, initial contact phase time percent, maximum medial-lateral displacement, maximum anterior-posterior displacement for gender classification and time percent, anterior-posterior speed, medial-lateral displacement standard deviation during forefoot push-off phase for age classification. In gender classification, it indicated that men spend more initial contact time percent, have greater gait variability and ankle torque during the stance phase. In age classification, it indicated that older women spend less time percent, faster anterior speed, greater weight transference variability during the forefoot push-off phase.
In future research, more subject data will be collected to enhance the robustness of classification performance. It may be possible to combine different type features and compare other classifiers to obtain better performance. The outcomes of the study indicate this approach can be applied to places where cameras cannot be installed due to confidentiality, such as industrial research and development laboratories. Furthermore, it can further enrich the diversity of the surveillance system and the accuracy of identification.

1. Ahmed, M.H., & Sabir, A.T. (2017). Human Gender Classification Based on Gait Features Using Kinect Sensor. Paper presented at the 2017 3rd Ieee International Conference on Cybernetics (Cybconf).
2. Bales, D., et al. (2016). Gender classification of walkers via underfloor accelerometer measurements. IEEE Internet of Things Journal, 3(6), 1259-1266.
3. Begg, R., & Kamruzzaman, J. (2005). A machine learning approach for automated recognition of movement patterns using basic, kinetic and kinematic gait data. Journal of biomechanics, 38(3), 401-408.
4. Bishop, C.M. (2006). Pattern recognition and machine learning: springer.
5. Bizovska, L., et al. (2014). Variability of centre of pressure movement during gait in young and middle-aged women. Gait & posture, 40(3), 399-402.
6. Blanc, Y., et al. (1999). Temporal parameters and patterns of the foot roll over during walking: normative data for healthy adults. Gait & posture, 10(2), 97-108.
7. Catalfamo, P., et al. (2008). Detection of gait events using an F-Scan in-shoe pressure measurement system. Gait & posture, 28(3), 420-426.
8. Chen, B., & Bates, B.T. (2000). Comparison of F-Scan in-sole and AMTI forceplate system in measuring vertical ground reaction force during gait. Physiotherapy Theory and Practice, 16(1), 43-53.
9. Chesnin, K.J., et al. (2000). Comparison of an in-shoe pressure measurement device to a force plate: concurrent validity of center of pressure measurements. Gait & posture, 12(2), 128-133.
10. Chiu, M.-C., et al. (2013). Gait speed and gender effects on center of pressure progression during normal walking. Gait & posture, 37(1), 43-48.
11. Chiu, M.-C., et al. (2013). Center of pressure progression characteristics under the plantar region for elderly adults. Gait & posture, 37(3), 408-412.
12. Cho, S., et al. (2004). Gender differences in three dimensional gait analysis data from 98 healthy Korean adults. Clinical biomechanics, 19(2), 145-152.
13. Choudhary, S., et al. (2017). A hybrid approach for gait based gender classification using gei and spatio temporal parameters. Paper presented at the 2017 International Conference on Advances in Computing, Communications and Informatics (ICACCI).
14. Cornwall, M.W., & McPoil, T.G. (2000). Velocity of the center of pressure during walking. Journal of the American Podiatric Medical Association, 90(7), 334-338.
15. Cortes, C., & Vapnik, V. (1995). Support-vector networks. Machine learning, 20(3), 273-297.
16. De Cock, A., et al. (2005). Temporal characteristics of foot roll-over during barefoot jogging: reference data for young adults. Gait & posture, 21(4), 432-439.
17. De Cock, A., et al. (2008). The trajectory of the centre of pressure during barefoot running as a potential measure for foot function. Gait & posture, 27(4), 669-675.
18. Debbi, E.M., et al. (2012). In-shoe center of pressure: Indirect force plate vs. direct insole measurement. The Foot, 22(4), 269-275.
19. DeVita, P., & Hortobagyi, T. (2000). Age causes a redistribution of joint torques and powers during gait. Journal of applied physiology, 88(5), 1804-1811.
20. Elble, R., et al. (1991). Stride-dependent changes in gait of older people. Journal of neurology, 238(1), 1-5.
21. Guyon, I., et al. (2002). Gene selection for cancer classification using support vector machines. Machine learning, 46(1-3), 389-422.
22. Hagedorn, T.J., et al. (2013). Factors affecting center of pressure in older adults: the Framingham Foot Study. Journal of foot and ankle research, 6(1), 1-5.
23. Haim, A., et al. (2010). The influence of sagittal center of pressure offset on gait kinematics and kinetics. Journal of biomechanics, 43(5), 969-977.
24. Han, T.R., et al. (1999). Quantification of the path of center of pressure (COP) using an F-scan in-shoe transducer. Gait & posture, 10(3), 248-254.
25. Hass, C.J., et al. (2004). The influence of Tai Chi training on the center of pressure trajectory during gait initiation in older adults. Archives of physical medicine and rehabilitation, 85(10), 1593-1598.
26. Hema, M., & Pitta, S. (2019). Human age classification based on gait parameters using a Gait Energy Image projection model. Paper presented at the 2019 3rd International Conference on Trends in Electronics and Informatics (ICOEI).
27. Hosseini, N.K., & Nordin, M.J. (2013). Human gait recognition: A silhouette based approach. Journal of Automation and Control Engineering, 1(2), 103-105.
28. Hreljac, A. (2000). Stride smoothness evaluation of runners and other athletes. Gait & posture, 11(3), 199-206.
29. Hunter, J.P., et al. (2005). Relationships between ground reaction force impulse and kinematics of sprint-running acceleration. Journal of applied biomechanics, 21(1), 31-43.
30. Jiang, S., et al. (2014). Real time gait recognition system based on Kinect skeleton feature. Paper presented at the Asian Conference on Computer Vision.
31. Jung, J.-W., et al. (2003). Dynamic-footprint based person identification using mat-type pressure sensor. Paper presented at the Proceedings of the 25th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (IEEE Cat. No. 03CH37439).
32. Kastaniotis, D., et al. (2013). Gait-based gender recognition using pose information for real time applications. Paper presented at the 2013 18th International Conference on Digital Signal Processing (DSP).
33. Katiyar, R., et al. (2013). A study on existing gait biometrics approaches and challenges. International Journal of Computer Science Issues (IJCSI), 10(1), 135.
34. Kim, S.-S., & Kim, H.-E. (2013). Gait Analysis on the Elderly Women with Foot Scan. Fashion & Textile Research Journal, 15(4), 613-619.
35. LaValle, S.M., et al. (2004). On the relationship between classical grid search and probabilistic roadmaps. The International Journal of Robotics Research, 23(7-8), 673-692.
36. Liu, Y., et al. (2014). Gait phase varies over velocities. Gait & posture, 39(2), 756-760.
37. Mason, J.E., et al. (2016). Machine Learning Techniques for Gait Biometric Recognition: Springer.
38. Menegoni, F., et al. (2009). Gender‐specific effect of obesity on balance. Obesity, 17(10), 1951-1956.
39. Michie, D., et al. (1994). Machine learning. Neural and Statistical Classification, 13(1994), 1-298.
40. Muro-De-La-Herran, A., et al. (2014). Gait analysis methods: An overview of wearable and non-wearable systems, highlighting clinical applications. Sensors, 14(2), 3362-3394.
41. Murray, M.P. (1967). Gait as a total pattern of movement: Including a bibliography on gait. American Journal of Physical Medicine & Rehabilitation, 46(1), 290-333.
42. Nichols, D.S. (1997). Balance retraining after stroke using force platform biofeedback. Physical therapy, 77(5), 553-558.
43. Nigg, B., et al. (1994). Gait characteristics as a function of age and gender. Gait & posture, 2(4), 213-220.
44. Ostrosky, K.M., et al. (1994). A comparison of gait characteristics in young and old subjects. Physical therapy, 74(7), 637-644.
45. Perry, J., et al. (2010). Gait Analysis: Normal and Pathological Function: SLACK.
46. Perry, J., & Davids, J.R. (1992). Gait analysis: normal and pathological function. Journal of Pediatric Orthopaedics, 12(6), 815.
47. Qian, G., et al. (2008). People identification using gait via floor pressure sensing and analysis. Paper presented at the European Conference on Smart Sensing and Context.
48. Røislien, J., et al. (2009). Simultaneous estimation of effects of gender, age and walking speed on kinematic gait data. Gait & posture, 30(4), 441-445.
49. Rodríguez, R.V., et al. (2007). An experimental study on the feasibility of footsteps as a biometric. Paper presented at the 2007 15th European Signal Processing Conference.
50. Rohrer, B., et al. (2002). Movement smoothness changes during stroke recovery. Journal of neuroscience, 22(18), 8297-8304.
51. Segel, J.D., & Crawford, S. (2014). Anatomy of the COP gait line and computer-aided gait analysis. Paper presented at the Pm’s Tech Forum/Orthotics & Biomechanics.
52. Song, J., et al. (1996). Foot type biomechanics. comparison of planus and rectus foot types. Journal of the American Podiatric Medical Association, 86(1), 16-23.
53. Taborri, J., et al. (2016). Gait partitioning methods: A systematic review. Sensors, 16(1), 66.
54. Tsung, B.Y.S., et al. (2004). Effectiveness of insoles on plantar pressure redistribution. Journal of rehabilitation research and development.
55. Winter, D.A. (1995). Human balance and posture control during standing and walking. Gait & posture, 3(4), 193-214.
56. Yeoh, T., et al. (2014). Genetic algorithm assisted by a svm for feature selection in gait classification. Paper presented at the 2014 International Symposium on Intelligent Signal Processing and Communication Systems (ISPACS).
57. Yu, S., et al. (2009). A study on gait-based gender classification. IEEE Transactions on image processing, 18(8), 1905-1910.
58. Yun, J. (2011). User identification using gait patterns on UbiFloorII. Sensors, 11(3), 2611-2639.
59. Zhang, D., et al. (2010). Age classification base on gait using HMM. Paper presented at the 2010 20th International Conference on Pattern Recognition.
60. Zhang, X., & Li, B. (2014). Influence of in-shoe heel lifts on plantar pressure and center of pressure in the medial–lateral direction during walking. Gait & posture, 39(4), 1012-1016.
61. Zhang, X., et al. (2013). A comparison of gait biomechanics of flip-flops, sandals, barefoot and shoes. Journal of foot and ankle research, 6(1), 1-8.