Scene Documentation

Introduction

HemoVision allows you to document and analyse bloodstains in complex environments using just photographs. To do this, HemoVision relies on markers that are placed on the crime scene. This documentation page explains how to correctly place and photograph them.

Marker Types

HemoVision uses two types of markers:

The smaller detail marker is used to indicate regions of interest. These markers should be placed in places that you wish to photograph in close-up later on. Each detail marker has a number, which has to be unique. In other words, never use markers that have the same number within the same scene.
The larger overview marker is used to position the scene in the world coordinate space. To this end, it has a reference point (cross) on the top left. By default, the world origin runs through this reference point, as illustrated in the figure below. Of course, you would typically want the origin of the world space to be somewhere else (e.g., the corner of the room), but more on that later.

Detail Marker Placement

When analysing (part of) a scene, HemoVision requires one overview marker and multiple detail markers to be placed. As explained before, detail markers indicate regions that you want to photograph from close by, with more detail. You can place as many detail markers as you like, making sure that each marker has a unique number. Detail markers can be placed on any flat surface, in any orientation. The image below shows two scenes with detail markers. In the left scene, all stains (and hence all markers) are placed on the same vertical wall. In the right scene, the bloodstains are spread across different surfaces, with different orientations. Hence, we placed markers on each surface containing interesting bloodstains.

Overview Marker Placement

Next, we need to position the overview marker. As explained before, the reference cross on the overview marker defines the world origin. This means that, by default, the height of the reference point with respect to the floor is zero. When loaded into HemoVision, a reconstruction will therefore look something like this:

Notice how most part of the images and stains lie below the floor. Obviously, we do not want this, but instead translate our whole scene so that the reference point is at the right height. If you have a 3D scan of the scene, you can simply rotate and / or translate your scene so that it matches up with your 3D scan, as shown below:

If you do not have a 3D scanner, however, you still want to be able express coordinates with respect to some predefined origin (e.g, the corner of the room). To make this easy, HemoVision offers three options for placing the overview marker. Each of these is discussed next. Note that in the below examples, the position and orientation of the detail markers is irrelevant.

Overview Marker on a Vertical Wall

The overview marker is placed horizontally on a vertical wall. By placing the marker horizontally, we can easily measure the height of the overview marker, and use this height to offset the origin (y-offset). This is illustrated in the figure below (a). Note how our scene has been moved up with respect to the origin. The origin now lies straight below the reference point of the overview marker.

If you want to further offset the scene parallel to the wall (e.g, to the corner of the room), you can measure the horizontal distance from the reference point to the wall, as shown by the red line in the figure below (b).

To Do:

Place marker horizontally on wall
Y-offset: Measure distance from floor to reference point (a)
X-offset (optional): Measure distance from reference point to corner (b)

Overview Marker on the Floor

The overview marker is placed flat on the floor. HemoVision implicitly rotates the marker from its standard upright position to a horizontal position, with its front facing up. The height of the marker (y-offset) is automatically set to zero. The origin falls in the overview marker reference point. No measurements are needed, as shown in figure (a) below.

If we want to express our results with respect to the wall (we want to have the origin in the wall), we measure the shortest distance from the reference point to the wall. In this case, make sure that the overview marker's long side is parallel to the wall. By specifying the z-offset, the origin now starts at the wall (figure (b)).

We can further move the origin towards the corner of the room, by measuring along the wall (x-offset). Again, make sure that the marker's long side is parallel to the wall. This is illustrated in figure (c) below.

To Do:

Place marker flat on floor (a)
Z-offset (optional): Place marker on floor parallel to wall, and measure shortest distance (b)
X-offset (optional): Place marker on floor parallel to wall, and measure to corner (c)

Overview Marker on the Ceiling

The overview marker is placed flat on the ceiling. HemoVision implicitly rotates the marker from its standard upright position to a horizontal position, with its front facing down. The height of the marker with respect to the floor need to be measured as y-offset (the height of the ceiling, in other words). X and z-offsets can optionally be measured, similar to when the overview marker is place on the floor.

Photography

This subsection will explain how and how many photographs should be taken to document and analyse bloodstains in HemoVision. For HemoVision, two types of images are required:

Overview Images: photographs taken from a distance, that encompass all markers (or as much markers as possible in the case of multi-surface documentation).
Detail Images: close-up photographs that capture detail on the bloodstains, and include at least one detail marker.

Single Surface

When all your bloodstains (and hence your markers) are on a single surface, you need to take:

One overview image that encompasses all markers, indicated by the blue camera in the image below. Try to move as close to the pattern as possible, while making sure that all markers are in the image.
As many detail images are you like. Each detail image is a close-up that includes (at least) one detail marker. Of course, you can make multiple photographs from the same detail marker. The green rectangles in the image below indicate possible detail images.

Multi Surface

When your bloodstains (and hence your markers) are on multiple surfaces, you need to take:

Many overview images that are taken from different positions and orientations. Each overview image should encompass as much markers as possible. Of course, it is not always possible to get all markers in each image. Just make sure that each marker is present in at least five images. As a general rule of thumb, take a minimum of 15 photographs from varying positions and orientations, as indicated by the blue cameras in the image below.
As many detail images are you like. Each detail image is a close-up that includes (at least) one detail marker. Of course, you can make multiple photographs from the same detail marker. The green rectangles in the image below indicate possible detail images.

Requirements

Photographs can be taken using any digital camera or smartphone camera. There are no strict requirements, though the following should be taken into account as much as possible:

We recommend a minimum resolution of 10MP. When photographing markers from far away, however, higher resolutions may be required.
Avoid distortion (e.g., fish-eye lens or wide-angle lens).
When using a flash, use a bounce flash as opposed to a direct flash to avoid unwanted glare and shadows.