Forensic image enhancement reveals blood-red stains

Blog, Digital Media Evidence, LBPD

In Dana’s case, Long Beach homicide detectives overlooked dark stains on the concrete floor of the master bathroom. These stains were made by Dana’s blood pulsing from her fatal head wound as she tried to escape from the bathroom, and as her husband dragged her into the shower, I believe. Huck tried to wash her blood away, but the concrete floor absorbed it, making it impossible for him to scrub the floor clean in the limited time he had. He tried to conceal the blood with decorative stain, I believe. (More here.)

When I studied the police photos of the bathroom, I saw reddish stains covered with a brownish wash. Other investigators dismissed my perceptions as biased and wrong. I wondered: Is there a way to objectively confirm my subjective impression that the dark stains look bloody red? Is there a way to separate the two different stain colors — brownish and reddish — to make blood in the police photos more obvious? The answer is: Yes, kind of.

I learned this forensic imaging technique a couple of years ago in a class taught by George Reis, an established expert in digital media forensics. It’s a relatively simple Channel Mixer adjustment applied in Adobe Photoshop CC 2018. This tool allows me to split the RGB (red, green, blue) in the images into three separate channels. Next, I can “turn up the volume” on the red channel, while “turning down the volume” on green and blue. This technique does not add red to the picture; rather, it emphasizes the red that’s already there. (Adobe explains.)

In his class notes, George Reis explained: “The Channel Mixer allows us to control the amount of data each separate color channel contributes to the image. With this control, if we need to de-emphasize a purple object, or isolate an orange object, we can get very precise control, which wasn╩╝t possible by splitting channels…. This tool has tremendous power for color isolation.”

To demonstrate, I applied Channel Mixer adjustments to police photo img_126. You can download the photo in full resolution here and try this yourself in Photoshop. Make sure that the image mode is RGB, 16 bit. In the gallery below, you can see a variety of “mixes” and settings that reveal the bloody redness of the stains. I also tried adding a levels adjustment to further reveal the red. Give it a try. (You can download additional police photos here.)

The stains are certainly red, but are they blood? Maybe a blood-stain-pattern expert might be able to draw conclusions. As far as I know, there’s no way to definitively confirm that a photograph depicts actual blood. Owing to the blunders and forensic malpractice of the Long Beach Police Department, the bathroom was not tested chemically for the presence of blood. But the LBPD did take photos, and that’s worth something, if only to me.

(George Reis is scheduled to teach a four-day workshop on Adobe Photoshop for Forensic Video Analysts in Huntington Beach in February 2019.)

Anger and disgust: emotion recognition demo

Blog, Digital Media Evidence

The police investigation into Dana’s death relied heavily on the assumptions and subjective judgments of the detectives in the case. They looked at surveillance videos and saw nothing that raised red flags for them. They assumed that all was well and, therefore, that no murder had occurred. When I voiced my subjective judgments — pointing out that there was a conflict between Dana and Huck on the night before Dana was found fatally injured — investigators dismissed my observations as biased.

Emotion recognition software probably wasn’t as widely available then as it is today. One app called Face++ scans photos for faces, reads cues in facial expressions, and — based on data — analyzes the emotions expressed in the faces. Granted, algorithms are not immune to bias. But when applied to images of Dana and Huck, perhaps the software is less biased than the LBPD or me.

According to Face++, Dana was angry that night. Huck’s facial expression revealed strong indicators of disgust when he looked at Dana. As I told investigators, all was not well, and Face++ seems to back me up. Here’s a short video in which I use Face++ to analyze images from that night: