Close-Range Photogrammetry
Written by Lee F. DeChant   

Miss the photos and figures?
View, read, share, save, and print this article
as it appeared in the print edition now, online!

POLICE INVESTIGATORS and forensic technicians often face challenges in obtaining accurate measurements when they are documenting crime scenes. Often the scene is in cramped quarters, making diagramming difficult. In many cases, it is unfeasible or nearly impossible to map a crime scene with an electronic total station due to unstable footing, line-of-sight issues, or limited space in the area requiring the instrument’s setup. Baseline tape measurements are often prone to error for 3D data, and in many situations the same can be said for producing reliable 2D data.

In this article, recent advancements in close-range photogrammetry (CRP) are discussed. The advancements include the use of color-coded targets for automatic camera orientation, and automatic measurement of passive-style and signalized synthetic targets. These advancements were developed for forensic 3D mapping—but many other fields and applications benefit from the sub-millimeter accuracy that is achieved using standard consumer-grade digital cameras and specialized photogrammetric software.

In addition to the 3D image-based measurement approach, the photogrammetric method discussed here was supplemented with a laser range-finder to accomplish 1D measurement inside a residential home. The results of the combined measurement methods were then completed through computer-aided diagramming (CAD) animation, producing a 3D “virtual tour” of the home’s interior.

Introduction

The goal of crime scene measurement is to reconstruct a real-world incident into a scaled diagram, represented as a 2D plan-view or a 3D model that can be used for visualization and animation. The initial step is to accurately characterize the dimensions of the incident scene. This is true whether the measurements are used to assist in the determination of a suspect’s location, bullet trajectory, blood spatter, location of weapons or drugs, or a myriad of other criminal factors.

The level of detail in the model varies depending on the kind of mapping data used. For example, in the case of a crime scene that involves bullet trajectory, the bullet path may travel through two or more adjacent rooms. In this scenario, an accurate 3D model is beneficial in order to document the precise bullet flight path through the rooms. The ability to use non-contact measurement methods is highly attractive when recording a combination of bullet trajectory, blood spatter, blood-pool areas, and other biohazard conditions. At other crime scenes, a simple documentation of evidence using more traditional measurement approaches may meet the needs of the investigator.

When diagramming rooms in a residential home, apartment building, or other interior crime scene, each environment can present its own unique set of diagramming challenges. No single “perfect measuring tool” is available that meets all requirements for crime scene recording. The key is to build an affordable measurement tool kit that provides the crime scene investigator a set of effective tools at the scene—ensuring accuracy that ultimately produces an irrefutable record of the scene through permanent data archival.

Measurement Objective

In the example detailed in this article, a simulated crime scene was measured in a four-bedroom home. The home consisted of an upper level and a lower-level daylight basement. The goal of the diagramming task was to use the automation of coded and signalized photogrammetric targets for the home’s irregular floor plan. The “irregular” floor plan layout is described as the most difficult portion of the floor to accurately measure two-dimensionally.

The bulk of the upper-level floor plan was recorded using CRP, while the width and length of the rooms were measured by clicking a few vector distances per room using a laser range-finder. Photogrammetry targets were used to model the hallways, foyer, kitchen, and all door entrances of the home’s upper-level rooms. This allowed accurate measurement of the rooms, and captured data that would be used later in a CAD program.

In close-range photogrammetry, the level of measurement detail is only limited by the number of images acquired on site. For example, if certain rooms are not considered immediately critical to the investigation, one does not need to spend time collecting precise measurement points at the scene. Instead, these areas can be photographed in a matter of minutes—
and then mapped later using photogrammetry methods. And if the investigator requires high-level details, 2D or 3D models can be rendered photogrammetrically using synthetic targets or natural-feature point marking taken from the photographs captured at the scene.

A Few Components of a 3D-Mapping Tool Kit

To measure a good portion of the home’s floor plan and evidence in this example, a number of tools were utilized:

  • Digital SLR camera
  • Close-range photogrammetry software system
  • Specialized targets for marking (or “modeling”) features
  • Laser range-finder
  • Computer-aided diagramming (CAD) program

Triplet Offset Targets

Some features in a scene—such as the right-angled corner of a countertop or the curved edge of a kitchen sink—are considered “untargetable”. To facilitate the automatic 3D measurement of untargetable features, photogrammetry specialists DeChant Consulting Services and Photometrix developed specialized markers called Triplet Offset Targets (or simply “Triplets”). The targets are made from red retroreflective material and are production die cut to provide high 3D measurement accuracy when they are combined with the iWitness Pro close-range photogrammetry system. The tip of the Triplet designates the point of interest.

Photogrammetry at Work

In order to compute a point of interest of the Triplet Offset Target, the camera positions are automatically solved in iWitnessPRO by measuring “coded targets” that are spatially separated in the image’s field of view. In order to accurately compute the Triplet Offset Target point, at least four camera positions should be used to image the coded targets and Triplets from perspective-imaging angles. The left-hand side of Figure 1 is an expanded view of two Triplet targets and six coded targets. The white arrows denote the Triplets in Figure 1.

Due to the overall size of the project, two photogrammetric networks (Network 1 and Network 2) were “linked” by conducting a least-squares coordinate transformation from six shared passive-style targets. These six coordinate-transformation linking points are illustrated on the right half of Figure 1 (“3D view”) as yellow circles. The coordinate system is automatically defined by giving common names to the six points in the two networks (e.g. “CPn”). In this case, the six point names were designated CP1 through CP6.

Approximately 80 coded targets were randomly placed on the hardwood flooring of Network 1, shown on the left half of Figure 1. Network 2 used approximately 60 of the codes from Network 1. It took one person less than an hour to image the entire photogrammetric Network 1 and 2. The processing time to produce the photogrammetric 3D points—accurately measured to a fraction of a millimeter using the iWitnessPRO software and a laptop computer—was about six minutes.

Computation of the positions of the Triplet Offset Targets is accomplished automatically after the project has been unit-scaled. The scale distances are illustrated as two precision scale bars, shown as bold red lines in the right-hand image of Figure 1 (titled “3D View”). The Triplets were also line connected in Figure 1 (“3D View”) prior to producing a DXF export. The photogrammetry software combines all networks into one file for import into any CAD program that reads the DXF file format.

Computer-Aided Diagramming

Depending on litigation requirements, a comprehensive 3D animation or a 2D plan-view drawing usually determines how much work is required in a CAD program using the 3D photo-grammetry and 1D laser measurements. Figure 2 is a plan view of the upper level of the home, displayed in the Crime Zone CAD software.

To position the two levels of the home, the lower staircase was accurately measured with the digital-camera images and photogrammetry, providing a natural “linking” of the upper- to lower-floor levels. The remainder of the downstairs rooms and hallway were then measured with the laser range-finder.

Figures 3 and 4 are oblique views of the measured results of the photogrammetry and laser measurements in CAD.

Finally, a “walkthrough movie” of the measured results was created in the Crime Zone software. You can watch the walkthrough here.

About the Author

This e-mail address is being protected from spam bots, you need JavaScript enabled to view it is the Principal of DeChant Consulting Services—DCS Inc. of Bellevue, Washington. DeChant is the co-developer and provider of the iWitness close-range photogrammetry software system, offering photogrammetric consulting, sales of iWitness, iWitnessPRO, metrically calibrated digital cameras, and a wide range of close-range photogrammetry products specifically designed for accident reconstruction and forensic image-based measurement.

Return to the December 2011 Featured Products & Services Main Page

 
< Prev   Next >






Forensic Podiatry (Part Two of Two)

THE DISCIPLINE of forensic podiatry—or, in other words, the examination of pedal evidence—has progressed significantly over the past ten years. It is no longer a question of “What can you do with a footprint?” but rather, “Who can we use to evaluate the footprint?” Cases involving pedal evidence, especially bloody footprints and issues of determining shoe sizing or fit issues compared to questioned footwear, have become more common over the past two or three years.

Read more...