Study of CCTV Footage Based on Lower-Limb Gait Measure for Forensic Application

Autor: Albert K. Chong, Ammar Majeed
Rok vydání: 2021
Předmět:
Zdroj: 2021 IEEE 12th Control and System Graduate Research Colloquium (ICSGRC).
Popis: Literature reveals certain lower-limb measures can be unique for every person. Hence, appropriate forensic gait measures can be developed for the identification of perpetrators from suspects of crime. CCTV multi-perspective views are readily available to provide high-resolution footage of the perpetrator of the crime for the calculation of these measures. Therefore, the reliability of these potential measures for forensic application needs to be investigated comprehensively. Twenty three male adults were recruited, and they were instructed to carry out ten gait trials to obtain the needed measures, which include the three-dimensional distance of the knee to ankle distances. Multi-perspective views of CCTV footage were processed using advanced photogrammetry techniques. Under-carpet pressure sensing plate was utilised to determine the occurrence of heel-down, mid-stance and toe-off gait phase, thus providing suitable markers for between-group comparison of the measure. To evaluate the reliability of these measures, the mean within a group and between-group significant scores (@P=0.05) were determined as (1) 0.001; and (2) 0.999 respectively. The results show that multiple CCTV footage and advanced photogrammetric methods were suitable for determining an individual’s lower limb measures during gait accurately. Also, this measure was suitable for identifying perpetrators successfully.
Databáze: OpenAIRE