Background

On October 22, 2016, this case was submitted to The Black Vault.  The photographer, submitted the following information:

“This picture was taken by my wife with a digital camera and tells me there was nobody around or behind when the photo was taken. She was also sat with 3 other people facing me that says the same.

I showed this pic to the hotel a few weeks after and they tell me of ghost activity in the hotel. I had the pic examined by a photographer that also says a ghost. I’m the one in the middle.

It was taken at the Holiday Inn, Leicester, built over a Roman garrison

Have any ideas on this?”

The case was turned over to TBV Investigations team member Jeremy Enfinger for analysis, who specializes in claims of ghost activity.

The Analysis

Metadata analysis indicates the image is the original file, however, EXIF metadata was not able to be extracted. The image was taken from a Samsung Model DV300, DV300F, or DV305F at the model’s maximum resolution of 4608 x 3456 pixels, using automatic exposure control with ISO 400, an exposure time of 1/8 sec and an F-stop of 2.5. With the jpeg quality being 94%, there is no indication that the image file has been compressed. The flash was fired during this exposure as well.

The time stamp on the metadata differs from the approximate date reported. According to the individual submitting the photo, it was taken around Christmas, 2015. The metadata indicates it was taken February 15, 2012 at 2:36 p.m. It should be noted, however, that if the camera owner did not set the time/date in the camera settings prior to this exposure, the metadata would not necessarily display accurate “modify date” information. It should also be noted that presence of Christmas decorations in the photo corroborates the individual’s report.

The Interview

After writing the initial report, on January 20, 2018, TBV Investigations team member Jeremy Enfinger conducted an interview with Shaun, the man who submitted the photograph.  He will be sending the actual camera to TBV Investigations for further testing, with hopes to try and reproduce the image using a multitude of settings and lighting situations, attempting to emulate the original setting.

Testing will be posted once it is completed.

General Observations

Aside from the Christmas decorations previously mentioned, ELA (error level analysis) shows similarities in texture of the anomaly with the background and close resemblance to some of the articles of clothing in the foreground. This similarity in texture implies there was no digital alteration of the photo in the area of interest around and including the anomaly. There is also stark contrast between the anomaly and light sources from within the field of view, and it blocks the light from the lamp (frame left), indicating the anomaly has mass and opacity through the clearly-visualized portions.

No notable indications of time/date found within the photograph. There’s a bottle of Whispering Hills Californian Chardonnay which began bottling in 2010, but no year is visible in the photo.

While the two men in the foreground (middle and left) have minimal red-eye effect from the flash, the man on right has moderate red-eye in the photo indicating his pupils were pointed directly toward the flash compared to the other two men in the foreground. The amount of red-eye for the anomaly is severe and more easily notable at a glance. This is inconsistent with the red-eye pattern of the men, and would typically indicate the subject would either be closer to the flash or looking more directly at the flash, neither of which seems to be the case give the direction it seems to be facing and physical location in the room, leaving some question about how this was accomplished. However, when examining the reflection of the flash in each of the subjects’ eyes, the angle of reflection is consistent with the men in the foreground and the anomaly, suggesting physical presence during the flash consistent with the angle/location of the photographer.

The anomaly does not possess the same RGB color values, and this section of the image appears to be desaturated compared to the rest of the photo, though some color is present. There is motion blur in a horizontal pattern surrounding the anomaly, but clear facial features are seen within the anomaly without motion. Initial observation was to suspect digital manipulation, but metadata contradicts this suspicion. It is possible to manipulate the metadata, but it would need to be intentionally done and knowledge of how to do this is not common.

Light reflection patterns are consistent in the anomaly compared to that of subjects in the foreground, however, there are shadows present at the edges of the anomaly that are not present with subjects in the foreground, indicating a discrepancy in how the room light interacts with the anomaly. Parts of the anatomy cannot be seen posterior to the facial anatomy, and appear transparent where the background is clearly visualized where anatomical features should be seen.

Conclusion

At first glance, my opinion was that this photo visibly appears to be digitally edited, however, there is evidence to suggest that the anomalous figure was in the room during the exposure. If it was hoaxed, someone would have had to go through great lengths to modify the metadata on the original image following photomanipulation, which generally would show on the ELA analysis and/or reduce the quality post-compression. Or someone found a way to take a photograph of a doctored photo/sample, which would possess accurate metadata, though I found no evidence of this.

Additionally, it’s possible that someone may have walked into the frame during exposure. With a 1/8 second exposure, a blur effect might be possible depending on how fast that person would be walking, but the blur should be consistent throughout the image of that person, not only in certain areas. It would also not explain the desaturation in color of the anomaly.

I would recommend further testing of the camera to include the features “split shot”, “motion photo” and “artistic brush”. It is possible the effects appreciated may have been added by mistake. I was not able to test this model of camera myself.

Given the red-eye appearance, the desaturation of the anomaly, the blurred edges (and clear facial features), the resemblance to Photoshop’s clone tool effect, and shadow discrepancies, there are enough indications to suspect some level of alteration. The problem is I can’t seem to find any evidence of manipulation given the data available. It is my opinion that this photo is authentic, however, I am unable to explain the anomaly or its features.

Currently, the photograph is with other photo analysts for evaluation, and results will be posted.

The Photograph

The Full Photo

Cropped and Zoomed:

Additional Analysis

Research image analyst Kenn Raymond, owner of VDI Spectral Art, decided to do his own independent analysis on the photograph above.  We were very intrigued by Kenn’s findings, because after extensive analysis, Kenn concluded the image to be an unknown, but did find a possible clue (as of 5/17/18 is still being researched) that shows the metadata lists the “number of images” as 2, indicating a possible double exposure.

https://www.theblackvault.com/casefiles/wp-content/uploads/2018/01/VDI-Spectral-Art.pdf

Additional details will be added, when available.

Replication?

Some visitors to The Black Vault are now submitting what they believe, are examples of this photo being replicated.

Below, you will find a few. Did they solve this case for you? Or does this prove the exact effect is still not achieved? Post your comments below.

Submission #1

Al Jacinto II on The Black Vault Fan Page on Facebook, posted the following photographs as “proof” that the image in question, is not a ghost (Image captions are quotes from Mr. Jacinto):

It should be pointed out that another comment made by Al Jacinto II stated:

“Thanks…but ghosts are unreal ha ha ha ha”

In other words – the agenda would simply be to debunk, and not offer a truly unbiased test to reproduce the result.

Submission #2

Reddit user extremesalmon submitted the following photograph, after seeing the above case:

Meta Data

The post Ghost Captured on Photograph, taken in Leicester, United Kingdom first appeared on The Black Vault Case Files.

The post Ghost Captured on Photograph, taken in Leicester, United Kingdom appeared first on The Black Vault Case Files.

Background

The following case was submitted to The Black Vault / TBV Investigations.  The witness submitted the following testimony:

Witness Testimony

“I was randomly shooting the moon with my Canon SX60HS which can zoom into craters on the moon. I had it on auto, no particular settings other than 3 shot burst.

I did not see the object when I clicked the shutter, it was only later when looking over the images of the day. What I captured I kept to myself, a few friends and family. Posted it to fb twice for some opinions later on.
It was only recently that I decided to throw it out there and get some real ideas on what, who and where this came from.

Thanks for your help,
NAME REMOVED

** I have added all of the pictured captured in the 3 burst. The times in exif data reflect this.
– image with object IMG_6440
– images without objects IMG_6441, IMG_6442

** I have added the photo to a video and blown it up here

Note from The Black Vault: The video was removed to withhold the witnesses identity. It only was a blow up of the below photographs, so all REAL evidence is maintained in this case file. For reference, here is a blow up of the object (with is a screen shot from the video).

The Evidence

The following photographs were submitted for analysis:

The Analysis

by Michael Kelly, TBV Investigations

After conducting a series of thorough photo forensic methods, a number of discrepancies were found in the photographs submitted by the witness.

We list only a few of the examples, though more can be found in the metadata using sophisticated photography software. These examples are sufficient to support our conclusion, and we feel that adding even more additional (and technical) evidence would only assist future digital manipulation of a “UFO” photograph.

Problem #1

The 3-shot burst mode is not indicated in the time stamp nor the metadata which indicates each shot was separated by a few seconds and not fractions of a second as would be the case in burst mode.

Image time stamps

• 19:15:09
• 19:15:12
• 19:15:20

It is noted that when the witness was asked about this, they submitted the following, and changed the original submission:

“Yep you are right in that they possibly weren’t the pictures I snapped in the burst, I have always assumed they were as I was going from memory out shooting. They were certainly part of the same group of images which I had shot in the burst, however they may have been just straight shots as well.”

Problem #2

The changing position of the moon in the photographs indicates that the camera was being held or pointed in different directions on a tripod. The witness confirmed in follow-up communications that the camera was being held.

When comparing all three images pixel for pixel, the moon in each image is identical.

With a change of position of the camera, the moon should have slight variations in each photograph. The fact that it does not suggests that the images were digitally modified, copying the same image of the moon into all three photographs.

Problem #3

Camera sensors create a high frequency noise pattern over an entire image.

In this photograph the moon and sky have this pattern, but the dot does not.

The fact that the dot does not have this pattern suggests that it was digitally inserted into the image.

Problem #4

Through investigation, the witness was found on Facebook using only publicly available information. Recent (public) posts include one of his own Photoshop works with a hidden “easter egg” that he inserted, wherein he invites his friends to attempt to find it.

Other posts highlight his Photoshop skills, along with his web  and graphics design career.

It is for this (and other) reasons we decided to strip the name of the “witness” off this report, and post it as Anonymous.

Conclusion

These photographs do not stand up to digital scrutiny and therefore must be labeled fraudulent.

There is additional information which TBV Investigations has left off this report.  We feel the above information is more than enough to support this conclusion, and we do not aim to enhance digital manipulation techniques.

Witness Responds to Digital Manipulation Claim

Note from The Black Vault: In the interest of fairness, the witness responded to our report prior to it being published. They felt the need to respond, and we offered to post their responses here, unedited.

This is what they submitted:

Let me address your points as we go.

TBV Investigations Final Response

In addition to the above statement, the witness submitted a video.  We assume by their description, they did not want it shared publicly, so we respect that and have not published it. However, we are going to respond, and ultimately leave it for all of the readers of this case to decide.

The following is from a third party consultant to TBV Investigations. Their ability for image analysis is highly credible.  We have withheld their identity (unless they give us permission later) to thwart future hoax submissions on who we call upon for help.  Their information below is very detailed, and can speak for itself.  Also, the analyst below did view the video (per the witnesses request to show to our analysts). The time codes referenced are irrelevant since the original response video by the witness is not show, but the analysis is self explanatory.

Regarding his video:
1. At 2:40, he correctly interprets the ELA result as being uniform. However, he draws an incorrect conclusion at 2:40 when he says that a resave would have removed the white noise. The white noise only gets removed if the picture is a low quality. His picture, as I previously noted, does not appear to be low quality.

2. At 5:15, he shows his camera roll.
This does not identify the pictures as being original — only that he has them in the same directory.

There’s also something interesting:
At 5:25, the camera roll shows many moon pictures: 6440-6448, but we only have 6440-6442.
Picture 6444, 6445, 6446, and 6447 were visibly altered:
6444 was cropped to a square (his canon doesn’t do that), recolored to a teal, and strong vignetting was added. 4665 had different color alterations. Similarly, 6446, 6447, and 6448 were all visibly altered. And this is easy to see from the thumbnail images.

This means that the known moon pictures (6440-6442) are sitting in a directory next to other pictures that were definitely altered.

3. At 5:38: Viewing metadata from Windows Properties literally proves nothing. In particular, Windows metadata viewer displays fields that do not exist in the file, and omits fields that do exist. See: http://www.hackerfactor.com/blog/index.php?%2Farchives%2F712-Metadata-Tools-of-the-Trade.html 

By the same means, at 6:35, he claims that the Microsoft property tab for “Previous Versions” says it was never resaved. However, that only applies to using Microsoft to edit pictures and retain revisioning. Again, it literally proves nothing.

4. At 6:55, he zooms in to show no photo manipulation. But again, this proves nothing.

5. At 7:20, he shows that the Microsoft properties for the created and modified dates match his claim and have not been edited. However, this just shows the metadata fields. (And it is my claim that the metadata was replaced.)

Now, let’s compare his counter-argument with my original findings:

Finding #1: Burst mode — alright, he admits he was wrong. We can look past this part.

Finding #2: The pictures show the same moon at the exact same orientation relative to the camera. Yet, the moon’s position changed relative to the camera. So he somehow managed to change the angle of the view of the moon, repositioning the moon in the frame, and still kept the exact same moon orientation.

This is a fun experiment to try at home: Hold the camera and take 3 pictures of a distant object — each 3 seconds apart. Then align the pictures on the object. The orientation should change. A very small object (only a few pixels in the frame) may appear to keep the orientation. A very large object, like the moon taking up 20% of the frame, should change orientation a little. Even though it is only a little different, this should be a visible difference when you overlay the two pictures and view the difference.

Then try the same experiment using a tripod — be sure the move the camera on the tripod a little so the object is not in the same relative location between frames. Again, it will change orientation.

And yet, he someone managed to take 3 pictures that all have the moon with the exact same orientation.

Here’s a difference-overlay of 6440 and 6442 (11 seconds apart): http://fotoforensics.com/analysis.php?id=4a3d92dcda18e048a36bea4871c2b34948246585.2346289&fmt=orig
– There is blue sky that exists in 6442 and not 6440 — visible due to aligning the moons.
– The UFO stands out because it doesn’t exist in 6442.
– There’s a tiny fraction of misalignment along one edge of the moon — likely due to scaling.
However, if the moon were rotated even one degree, then we should start seeing that reddish difference edge over most of the moon’s edge and most of the moon’s high-contrast features. We’re just not seeing that.

Even more interesting: Save the difference picture, bring it up in Gimp or Photoshop, and apply a histogram equalization to it. If you apply a histogram equalization to the difference picture, then you can see (1) moon features stand out along JPEG compression differences (expected), and (2) there is a dark swath through the moon that goes through both pictures. That dark swath should be a lighting artifact (like a lens flare), but it is in the same place in both pictures, even though the camera “moved” relative to the moon.

There’s one other thing that bothers me (and makes me chuckle):
In 6440, there is no motion blur. So the object is traveling slower than the shutter speed.
A few seconds later, in 6441, it is gone. So a UFO flew into view, sat still for a photo, and then vanished. In order to vanish, it would have to travel more than half the width of the frame in 3 seconds. Assuming it is far away and not in front of the lens, that would require a very fast speed.

Now, I’m not claiming that a real extra terrestrial UFO couldn’t travel really fast, but this seems like an amazing coincidence that he photographed it at the brief moment that it came to a stop.

Or maybe cloaking? Again — amazing coincidence that he photographed it right at the moment when it was about to put up the cloak.

Of course, if it were a blur from an insect that this close the the camera, then that would explain the really fast relative movement.

Regarding his written comments:

Accepted by MUFON? Good for him. MUFON didn’t contact me about this picture. Had they contacted me, I would have told them the same things I told you.

“#4 – The fact I post Graphics on social media is kind of moot, I am a web designer…”
Note: You hadn’t told me his name, but I got it from his youtube video.
This goes toward skillset. He clearly is capable of doing photo manipulation. This is evident in many of his pubic galleries. For example, his art gallery (LINK REMOVED TO WITHHOLD IDENTITY OF WITNESS) includes a collection of real, altered, and edited photos.

What do you think? Post your comments below! 

Special Thanks

TBV Investigations team members Jeremy Enfinger, Jim Kerr, and James Clarkson all contributed to this report and analysis.

Raw Metadata

Image 2

Image 3

The post UFO Appears Over Cornwall, Ontario Canada, While Witness Photographs Moon first appeared on The Black Vault Case Files.

The post UFO Appears Over Cornwall, Ontario Canada, While Witness Photographs Moon appeared first on The Black Vault Case Files.