Gender recognition from torso features using elliptic fourier descriptors

摘要: Gender recognition has important applications in apparel design, social security, and human-computer interaction systems. In this paper, we investigate gender-recognition technologies using 3-D human body shape. The front side silhouettes from 459 female subjects 107 male were extracted then modeled normalized Elliptic Fourier descriptors. Principal Component Analysis (PCA) was conducted to summarize the information contained by EF coefficients. A back propagation (BP) neural network with 33 inputs, 2 outputs 10 hidden layers adopted gender recognition. research demonstrates that torso features achieved a considerably high rate. Moreover, combination of PCA BP have provided effective ways for overcome some limitations other technologies.