论文部分内容阅读
从人类手印中提取信息确定性别以推断史前手印岩画中男女的性别角色。人们曾以手工的方式测算手指长度的比例和其他物理特征来判断性别角色。绝大多数传统的研究方法是基于手工测量长度,因此常常受制于出版的图片中信息量的缺失。我们已经探索了一种通过测量获取手部图像的信息并基于现代机器学习确定性别角色。这是目前已知的用自动化程序代替耗时的手工测量以此来确定史前岩画作者性别身份的方法。我们的研究为两性异形和旧石器时代晚期社会的劳动分工提供了定量的实证。此外,除了分析历史中的手印图像,该研究方法也具有应用于犯罪取证和人机交互方面的潜在可能。
Extracting information from human fingerprints Determining gender to infer male and female gender roles in prehistoric handprint rock paintings. People have manually measured the proportion of finger length and other physical characteristics to determine the gender role. The vast majority of traditional research methods are based on manual measurement of length and are therefore often subject to the lack of information in the published pictures. We have explored a way to get the hand imagery by measuring the information and determine the gender role based on modern machine learning. This is a known method of determining the gender identity of a prehistoric rock artist by replacing the time-consuming manual measurement with an automated procedure. Our research provides a quantitative evidence of the gender division and the division of labor in the late Paleolithic society. In addition, in addition to analyzing handprint images in history, this research method also has the potential to be used in criminal forensics and human-computer interaction.