A Crash Course in Forensic Scanning
September 4, 2006
For many years, folding rulers, tape measures and black and white photography were the standard tools of the trade for collision reconstruction engineers. Armed with these tools, experts collected as much data as possible to evaluate a collision sequence. In those days, as it does today, the quality of the data collected directly affected the accuracy and the depth of the analysis. However, conventional data collection tools limited the quality of the recorded data. Typical problems encountered by collision reconstruction specialists included the difficulty associated with re-creating realistic representations of a collision area and vehicle(s), as well as performing diagnostic examinations of vehicles and measuring roadway curvature, elevations, roadway friction factors, illumination levels and vehicle dimensions. New technology, however, has provided forensic surveyors and engineers with the tools to address these and other reconstruction needs. Through the specific utilization of cutting-edge laser scanning technology, engineers and surveyors can now collect numerous measurements cost-effectively, with unprecedented accuracy, at the speed of (laser) light. This new technology has the ability to rapidly advance the field of forensics-not just in the area of collision reconstruction but in any discipline that requires precision measurements.
The Laser Scanning ProcessTraditionally, forensic engineers have utilized professional surveyors and/or the equipment professional surveyors use. Though traditional surveying equipment is still a valuable tool in the engineer's toolbox, its speed limitations and measurement recording techniques have been far surpassed by laser scanning technology.
Before the advent of laser scanning technology, the data collection process hindered forensic engineers from obtaining more than a couple of measurements per minute. Now they are able to record 100,000 measurements every minute. These measurements are recorded from laser light that reflects from objects after being projected from a series of rotating mirrors. The scanner is indiscriminate in terms of the features it measures.
If the scanner is directed toward an intersection, for example, it will record the roadway widths, as well as the lane lines, traffic signals, signs, utility pole lines and trees. If a building is scanned, the dimensions of the walls, as well as each individual brick and the mortar between the bricks will all be recorded. The quality and amount of data recorded is historically unprecedented. The mass quantity of recorded raw data (often referred to as a point cloud) is accurate to within a few millimeters (approximately the size of a pencil eraser). In addition, since the raw data cannot be altered or modified in its initially recorded state, the data validation process with other engineers and investigators has been greatly streamlined.
The point cloud for each object scanned is created and appears, in real-time, as a graphical three-dimensional image on the laptop computer connected to the scanner as the object is scanned. The point cloud is a dimensionally accurate representation of the object scanned and the image can be enhanced through software provided with the scanner. Software employed by the scanner provides a relatively simple process to "shrink-wrap" the point cloud image. The process begins with connecting the points together in a triangulated mesh. The program then applies edge detection algorithms to the triangulated mesh to identify the outlines of specific objects. Intensity mapping and rendering are applied to the mesh to establish the outlines of the objects. The result is a visual image that provides a clear delineation of the geometric elements of the object through the use of lines and colors. The point cloud data can be exported to other applications but there are caveats. The size of the data file is such that the volume of data sometimes overwhelms applications. In these cases, some points may need to be deleted. This can be accomplished so as to have little effect on the accuracy of the model.
Scanning BenefitsLaser scanning produces increased quantities of data, improved analyses and more detailed exhibits. It also captures data with unprecedented speed in areas where it would not typically have been possible to collect data. These benefits enhance the engineer's and surveyor's ability to analyze and present the data.
For example, in the past, forensic engineers and surveyors have faced costly and time-consuming data collection challenges that made it difficult to meet their client's needs. Engineers and surveyors have had difficulty measuring complex vehicular crush profiles, small variances in building facades multiple stories off the ground, and roadways in high-traffic areas. However, the enhanced speed, precision and operation of laser scanning have now made these issues easier to successfully address.
Since the laser scanner will record every object within its view, the necessity of returning to the site of an event is eliminated even if the focus shifts during the development of a case. Previously, during the life of a case, an engineer or surveyor might have been forced to re-examine a vehicle or return to the location of an event to record an object not initially considered to be an issue. Now, with the advent of indiscriminate data collecting, all objects are accurately measured regardless of their importance to the case at the time. When measuring collision scenes, the positions of sight obstructions such as trees, bushes, signs, etc., become part of the three-dimensional data stored as part of the point cloud.
As the quantity of data increases, so does a surveyor's or engineer's ability to perform more accurate analyses. Some computer applications have the ability to calculate the speed of a vehicle using supplied stiffness coefficients in conjunction with vehicle crush profiles. Enhanced detail and accuracy with crush profiles increases the accuracy of the computed speeds.
Some simulation packages utilize road and terrain contours to determine the motion of a vehicle. Laser scanning enables the surveyor or engineer to take detailed measurements of a road surface, producing three-dimensional models for analysis. Since the point cloud produces a three-dimensional image, the surveyor or engineer can now view the field data, including the site and vehicles, in a three-dimensional world. Not only is this a valuable tool for the purpose of analysis, but the ability to simulate the driver's view in that three-dimensional world is remarkably beneficial when showing a layperson the results of an analysis.
Some simulation packages can predict the movement of a body within a vehicle as that body strikes the interior contact surfaces. The laser scanner can be used to create a variety of complex contact surfaces by recording the location of every object, from the seat and the dashboard, to the steering wheel and the gear shift.
Once the data is collected and the analysis is complete, it may be necessary to create demonstrative evidence to help explain the analysis to those who do not have engineering or surveying backgrounds. Laser scanning is an excellent means of collecting three-dimensional data for computer animations. Engineering animators often struggle with a lack of sufficient detail when it becomes necessary to create an animation from a surveyor's measurements. Laser scanning technology, however, has provided animators with the ability to quickly and cost-effectively convert the millions of point cloud measurements into accurate three-dimensional models and animations.
Some real-life examples of laser scanning use in forensic applications follow.
Case Example: Comparison of Damage ProfilesA product liability suit was brought against a vehicle manufacturer for a single car collision involving a pole. A key element in the analysis was the quantification of the change in velocity of the vehicle at the time of the collision. Classic reconstruction techniques were utilized to perform the calculations. The profile of the damage to the vehicle was measured as part of the analysis. The manufacturer, in response to the plaintiff's analysis, performed a crash test of an exemplar vehicle. Laser scans of both vehicles were ultimately completed, which allowed the plaintiff's experts to compare the profiles of the damage to both the actual collision vehicle and the test vehicle. The comparisons were utilized during the direct testimony and cross-examination of the expert. They were also shown to the jury during testimony and made available to them as evidence by the court to use in deliberations. The ability to define and view these differences in three dimensions was critical in explaining to a jury the validity of the plaintiff's original analysis as well as the problems with the defendant's analysis.
Case Example: Forklift Loading AccidentA forklift operator loaded a large A/C unit onto a stake-body truck from a loading dock meant for larger trailers. The truck moved forward and the forklift, together with the operator, fell off the dock resulting in injury to the operator. The analysis included an attempt to reconstruct how the collision occurred, including why the truck moved forward. A laser scan was performed of the loading dock, the forklift and the truck in question.
Initial measurements were extracted from the scan including the relative heights of various parts of the loading dock, the gate and back of the truck. Utilizing this data, an initial analysis was completed. A three-dimensional model was quickly made using the scan of the dock, truck and forklift and the load that was being moved. By re-creating the forklift operator's actions in the modeled world, it was shown that the incorrect method utilized by the forklift operator when loading the A/C unit would have resulted in the load catching on the gate of the truck, then pushing it forward. This re-creation proved to be not only an invaluable reconstruction tool, but was then quickly turned into an animation for demonstrative purposes.
It is apparent that for all facets, from the initial investigation to the preparation of exhibits, laser scanning technology is poised and ready to reshape the field of forensic surveying.
The author wishes to acknowledge the contributions of Andrew Durian, BSME; James Schmidt, BSME; and Terry Myers for their invaluable contributions to the collection, analysis and presentation of the data referred to in this article.
Typical Uses of the Laser Scanner in Forensic and Non-Forensic AreasNon-forensic uses of the laser scanner are presently much more prevalent. Typical uses of the laser scanner include:
- Measuring buildings and their facades for historical restorations or documentation;
- Measuring complex mechanical operations like pipelines and the inner workings of chemical and nuclear power plants for the purpose of establishing as-built drawings;
- Measuring difficult to reach or complex surface structures like bridges, roadways and buildings for as-built drawings;
- Documenting disaster areas, including building collapses, roadway defects and structural failures, damaged vehicles; collision areas, including roadways, shipping docks, parking lots and stairways; the remnants of building fires; interference checks with new designs and as-built scan databases;
- Recording crime scenes;
- Comparing and inspecting work in progress with original plans;
- Performing condition assessments of bridges;
- Surveying high-traffic areas without shutdowns or risk to a survey crew;
- Conducting remote and accurate measurements of rock faces;
- Gathering volumetric measurements of cuts and fills; and
- Producing accurate 3D models of movie sets for the purpose of generating digital special effects.