相关实验视频
Updated: Aug 16, 2026
09:47
Spotting Cheetahs: Identifying Individuals by Their Footprints
Published on: May 1, 2016
15.3K
使用增强的语网络方法对裸脚和穿的足迹证据进行交叉比较
Yangbo Li1, Baien Guo2, Yao Shen1,3
1School of Criminal Investigation, People's Public Security University of China, Beijing, China.
Journal of forensic sciences
|September 24, 2025
概括
这项研究引入了一种新的AI方法,用于比较赤脚和穿子的脚印,改善法医识别. 增强的语网络在挑战混合数据集方面取得了显著的准确性,有助于刑事调查.
科学领域:
- 法医科学 法医科学 法医科学
- 计算机视觉 计算机视觉
- 生物识别信息 生物识别信息
背景情况:
- 在法医分析中,比较赤脚和穿的印象带来了重大挑战.
- 犯罪分子可能会穿子来掩盖鞋子的证据,使调查复杂化.
研究的目的:
- 开发和评估一种由人工智能驱动的方法,用于对赤脚和穿的脚印进行交叉比较.
- 为了解决传统的法医鉴定技术的局限性,用于模糊的鞋类证据.
主要方法:
- 提出了一种增强的语网络,采用双分支ResNet34特征提取器.
- 实施了道级通用平均值 (GeM) 聚合策略和用硬样本挖掘进行指标学习.
- 使用了1600个足迹样本 (800名参与者,包括镜像转换) 的数据集.
主要成果:
- 在混合赤脚和穿印记的交叉比较中,达到63.4%的Top-1和90.9%的Top-10准确度.
- 具有增强 GeM 聚合功能的 ResNet34 架构优于其他替代方法.
- 在具有挑战性的检索环境中表现出卓越的性能.
结论:
- 拟议的AI方法提供了一个客观和可量化的方法来识别赤脚和穿的印记.
- 这种技术在刑事调查中具有实质性的实用价值,特别是在涉及隐蔽鞋类证据的案件中.
更多相关视频
06:25Paw-Print Analysis of Contrast-Enhanced Recordings PrAnCER: A Low-Cost, Open-Access Automated Gait Analysis System for Assessing Motor Deficits
Published on: August 12, 2019
9.0K
14:50Multifactorial Assessment of Motor Behavior in Rats after Unilateral Sciatic Nerve Crush Injury
Published on: July 31, 2021
8.0K
相关概念视频
01:35Pedigree Analysis
Overview
01:17Sign Test for Matched Pairs
The sign test for matched pairs offers a robust method for comparing two paired samples, often for the effects of an intervention in one of them. This method is very useful in situations where the underlying distribution of the data is unknown. The test compares two related samples—often pre- and post-treatment measurements on the same subjects—to determine if there are significant differences in their median values.
To conduct the sign test, we first calculate the differences in value between...
To conduct the sign test, we first calculate the differences in value between...