How to read infrared sensor footage, and the eight-pointed star

Mikey · 21 May 2026

THE SOURCE

DOW-UAP-PR38, “Unresolved UAP Report, Middle East 2013”, a 106-second infrared video clip from the U.S. Department of War’s PURSUE Release 01, published at war.gov/ufo on 8 May 2026. It is one of 27 infrared sensor clips in the release and the one press coverage singled out as showing an “eight-pointed star with arms of alternating length”.

What this briefing is

This is the first video briefing in the series, and it does two jobs at once.

The first job is to introduce PR38, the single most visually distinctive clip in PURSUE Release 01. The second, and the more useful in the long run, is to teach the reader how to read a military infrared targeting feed in general. Release 01 contains 27 of these clips. Every later video briefing on this site will refer back to this one rather than re-explain the basics. If you only read one video briefing closely, read this one.

The short version is that PR38 is genuine sensor data, it is official, and it is easy to over-read. This briefing is about the difference between those statements.

TL;DR

PR38 is a 106-second clip of black-and-white infrared footage. It looks like a targeting-pod or surveillance-camera feed: a wide grey field, a small tracking reticle near the centre, corner brackets framing the tracked area, and a compass marker. Almost the entire telemetry layer, the rows of numbers that would normally tell you altitude, range, time and sensor mode, is hidden under heavy black redaction blocks.

For roughly the first third of the clip the framed area is empty grey sky. Then a bright object enters from the right of frame, appearing in the footage as a dark spot with sharp spikes radiating from it, the shape the press described as an “eight-pointed star”. A long, thin bright streak runs across the lower part of the frame for much of the clip. Over the remaining footage the camera slews to bring the object towards the centre reticle and tracks it there.

What the footage genuinely shows is a real, heat-emitting object captured by a real military sensor, and a real tracking effort by the operator. What it does not show, and cannot show on its own, is what the object was. The “eight-pointed star” shape is almost certainly an artefact of the camera, not the geometry of the object. This briefing explains why.

This briefing is evidence tier 2: primary footage. Genuine sensor data from an official source. Not, by itself, proof of anything anomalous.

What is on the screen, part by part

Before anything else, learn the furniture of the image. A targeting feed is not a photograph. It is an instrument display, and most of what you see is the instrument talking, not the object.

The grey field. The whole frame is infrared, also called thermal imaging. The camera is not seeing visible light. It is seeing heat. Pale areas are one temperature, dark areas another. This is the single most important thing to understand about the clip, and it gets its own section below.

The reticle. Near the centre sits a small cross or crosshair shape. This is the tracking gate, the point the sensor is aimed at or trying to lock onto. When the object and the reticle are not on top of each other, the operator is still slewing the camera to catch up. A reticle is an aiming mark. It is not evidence that the thing inside it is unusual.

The corner brackets. Four small right-angle marks frame a rectangle in the middle of the screen. This is the field-of-view or track box. Things drift across it as the camera moves and as the object moves, and you cannot tell those two motions apart from the footage alone.

The compass marker. A small letter “N” sits left of centre, a heading reference. It tells you which way is north in the sensor’s frame. It does not tell you the object’s bearing or speed.

The black blocks. Large black rectangles cover the top, sides and bottom of the frame. These are redactions, applied after the fact, over the telemetry readouts. On an un-redacted feed those areas would carry numbers: time, date, sensor mode, magnification, aircraft or platform altitude, slant range to target, and coordinates. In PR38 nearly all of it is hidden.

This matters in a specific way. The redaction blocks are the reason this clip cannot, by itself, tell you how big the object was, how far away it was, how fast it moved, or how high it sat. All of those depend on the telemetry, and the telemetry is blacked out. As Briefing 1 noted, the redaction codes on this release point at operational and personnel secrecy, the kind applied to any military sensor product, not at hiding the object. But the practical effect is the same either way: without the numbers, the clip is a picture, not a measurement.

Why infrared inverts hot and cold

Here is the part that trips up almost everyone watching this footage for the first time.

In PR38 the object of interest appears dark. The sky around it is pale grey. An untrained eye reads “dark object” as “dark-coloured object”, and that is wrong.

Infrared cameras have two display modes, often called white-hot and black-hot. In white-hot, heat shows as bright and cold shows as dark. In black-hot, it is reversed: heat shows as dark and cold shows as bright. PR38 is in black-hot. The bright pale field is the cooler sky. The dark spot is the warmer object. The long pale streak across the lower frame is, in the same logic, a cooler feature, consistent with a contrail or a band of cloud rather than an exhaust plume.

So the object in PR38 is not a black thing. It is a hot thing, displayed dark. A military jet engine, a flare, a hot balloon skin in sunlight, an aircraft, a drone, any of these would also read dark in this mode. The display tells you the object is warmer than its background. It does not tell you what the object is.

The eight-pointed star

This is the detail that gave the clip its press coverage, and it is the clearest lesson in the whole briefing.

In the footage the object genuinely does look like a star or asterisk: a central blob with sharp spikes radiating outward, and across the clip the spikes are not all the same length. The press description, “an eight-pointed star with arms of alternating length”, is a fair description of what the screen shows.

But that shape is almost certainly not the shape of the object. It is an optical artefact. When a bright, compact light source is captured by a camera, the camera’s own optics smear it into a star pattern. Anyone who has photographed a streetlamp at night has seen this. The technical name is a diffraction spike or lens flare; the spikes are produced by the geometry of the lens, the aperture and the sensor, not by the source. The number of points and their relative lengths are a fingerprint of the camera, and they stay roughly consistent whatever the bright source is. Across the frames of PR38 the star sits at different positions and at different focus states, and the spike pattern shifts and softens with it, exactly as an optical artefact would, and not as a solid eight-armed structure would.

The honest reading is therefore: PR38 shows a bright, compact, hot object, and the camera has drawn a star around it. The “eight-pointed star” is real on the screen and almost certainly not real in the sky. A genuine UAP can produce a flare artefact just as easily as a conventional aircraft can. The artefact tells you the source was bright and point-like. It tells you nothing else.

Why a single frame proves very little

If you pause PR38 on its most dramatic frame, you have a bright star-shaped object inside a military reticle, surrounded by redacted data. It looks like a smoking gun. It is not, and it is worth being clear about why.

A still frame strips out the one thing video has and a photograph does not: motion over time. The interesting questions about any UAP clip, did the object accelerate impossibly, did it change direction without banking, did it move faster than a known aircraft could, can only be answered by watching how the object moves relative to the background across many seconds, and ideally by reading the telemetry while it does. A frozen frame answers none of them.

A still frame also cannot separate the object’s motion from the camera’s. In PR38 the camera is slewing to track. Some of the apparent movement on screen is the object; some is the sensor turning. Only the moving footage, read carefully, can begin to tell them apart, and even then the redacted telemetry limits how far you can get.

This is the core caution for every video briefing on this site. Genuine sensor footage is strong evidence that something was there and was tracked. It is weak evidence, on its own, for what that something was, and a single frame is weaker still.

What PR38 reliably establishes

It establishes that a U.S. military sensor in the Middle East theatre, in 2013, captured a real, compact, heat-emitting object against the sky, and that an operator slewed the camera and tracked it. It establishes that the encounter was recorded, retained, classified, and has now been released through an official, traceable channel as part of PURSUE Release 01. It establishes that the object was bright and point-like enough to produce a strong optical flare in the sensor optics. And it establishes that the event was logged as unresolved, meaning the military’s own analysts did not, at least at the time, attach a confident identification to it.

That is a genuine and useful set of facts. It is the floor of what footage like this can give you.

What it does not establish

It does not establish what the object was. PR38 is fully consistent with a range of conventional explanations, an aircraft, a drone, a flare, a balloon, even a distant celestial source, and the footage alone does not let anyone choose between them. “Unresolved” means not yet identified. It does not mean unidentifiable, and it certainly does not mean anomalous.

It does not establish the object’s size, speed, altitude or distance, because the telemetry that would carry those numbers is under the redaction blocks. Any claim about how fast or how high the object was, drawn from this clip alone, is a guess.

It does not establish an eight-armed or star-shaped craft. The star is an artefact of the camera optics, as the section above sets out. Reading the flare as the literal shape of the object is the single most common error made with footage like this.

And it does not, by itself, rise above tier 2. PR38 is authentic primary footage from an official source, which is a strong place to start. It is not proof of anything anomalous, because a single sensor clip, however genuine, records that an object was tracked, not what the object was.

Related wiki entries

The wiki entries below give background on the people, programmes and document types behind this briefing.

PURSUE · Department of War · AARO

References and further reading

• DOW-UAP-PR38, “Unresolved UAP Report, Middle East 2013”, U.S. Department of War, PURSUE Release 01, hosted at war.gov/ufo
• Presidential Unsealing and Reporting System for UAP Encounters, U.S. Department of War, war.gov/ufo
• AARO UAP Records, All-domain Anomaly Resolution Office, aaro.mil/UAP-Records
• Background on infrared imaging modes and optical flare artefacts in targeting-pod footage, AARO technical assessments, aaro.mil
• Signals from the Periphery, Briefing 1, “Release one, what is actually inside the Department of War’s first UAP document drop”