Extraction of Blood Droplet Flight Trajectories from Videos for Forensic Analysis. L.A. Zarrabeitia, D.A. Aruliah, F.Z. Qureshi Faculty of Science, University of Ontario Institute of Technology, Oshawa, ON, Canada {Luis.Zarrabeitia, Dhavide.Aruliah, Faisal.Qureshi}@uoit.ca Keywords: Stereo Reconstruction : Tracking : Droplets : Blood flight : Multi-target Tracking Abstract: We present a method for extracting three-dimensional flight trajectories of liquid droplets from video data. A high-speed stereo camera pair records videos of experimental reconstructions of projectile impacts and ensuing droplet scattering. After background removal and segmentation of individual droplets in each video frame, we introduce a model-based matching technique to accumulate image paths for individual droplets. Our motion detection algorithm is designed to deal gracefully with the lack of feature points, with the similarity of droplets in shape, size, and color, and with incomplete droplet paths due to noise, occlusions, etc. The final reconstruction algorithm pairs two-dimensional paths accumulated from each of the two cameras’ videos to reconstruct trajectories in three dimensions. The reconstructed droplet trajectories constitute a starting point for a physically accurate model of blood droplet flight for forensic bloodstain pattern analysis. (a) (b) (c) Figure 1: (a) BB pellet impacting ballistic gel containing transfer blood. (b) Tracking individual blood droplets in high-speed video (1300 frames per second). (c) Reconstructed blood droplet trajectories. Notice the effects of gravity and viscous drag forces even for short trajectories. 1 Introduction Bloodstain pattern analysis (BPA) comprises tech- niques for inferring spatial locations of bloodletting events from bloodstains found at crime scenes (Bevel and Gardner, 2008; Buck et al., 2011). At present, BPA, to a large extent, is a qualitative sub-discipline of forensic science. Our present goal is to im- prove computational models for bloodletting events and bloodstain pattern formation. These models, we believe, will be of immense value to forensic inves- tigators and BPA specialists for reasoning accurately from images of bloodstain patterns at crime scenes. Furthermore, such models are required to develop the next generation of BPA software for inference and as- sessing uncertainties in BPA. Stringing (Buck et al., 2011) is a common method for locating the bloodletting event responsible for a particular bloodstain pattern. This method relies on the assumption that blood droplets move in straight lines, ignoring the effects of gravity and aerodynamic drag. Stringing can provide reasonable approximate locations projected onto a horizontal plane. At short distances, stringing may also provide estimates for the height of the bloodletting event. More accurate bal- listic models that incorporate viscous drag forces and gravity, are only used when the stringing method pro- duces unreasonable locations or speeds (Buck et al.,