Can video cameras perceive lights in a way that they will totally change the color of molten metals?

Question

Is here anyone with experience of camera physics that could help me with this issue or that knows someone I could contact at a university maybe?. I am trying to find out how regular video camera perceives lights and colors of molten metals in daylight, my understanding being that a regular video camera can't change a color of something in a molten state with silvery color into orange/yellow color just as it should not change the color of a burning flame into a different and wrong color.

The only camera effect I know of is that the brightness levels on a video could be right on the upper maximum threshold for the camera, and going on to exceed it, so that it produces a purple/magenta color.

That very bright objects (such as the Sun, pieces of paper catching the sun, white-hot metal or flames, bright lights, etc), filmed with a CMOS camera, send too many photons onto an area of the sensor. The sensor doesn't handle this well. This causes electrons to overspill onto an adjacent area of the sensor, but usually in the same direction for each bright spot. The electrons cause the voltage on these areas of the sensor to increase, and that causes a problem. The voltage is meant to be 0, so that when light hits it, the sensor can measure the difference between 0 volts and some higher voltage caused by the light hitting it. But now 0 isn't 0 anymore, it's higher. Put simply, the sensor would normally measure the light level by measuring the difference between 0 and, say, 5 (maximum brightness). So black would be 0 and 5 would be white. But now it's measuring between a minimum of 7 (due to electron overflow) and 5 (maximum). So the result is always negative (even bright light would be -2), which the camera treats as 0 (black). Hence, you see a black spot slightly off-centre from the very bright light sources. But why purple? Well, it's simply because the sensor's red, green, and blue detectors have different sensitivities. They tend to be more sensitive to green, because our eyes are more sensitive in green, and we want the image to look right. Conversely, they tend to be less sensitive in red and blue, because our eyes are less sensitive to those colours too. As areas of the sensor start to become overloaded due to the electron overflow, it's the green that get overloaded first, as that's what it's most sensitive to. So these green areas become black (see above), and you're left with just red and blue. If you mix red and blue you get purple/magenta, and this is hence what you see when the sensor starts to get overloaded. Eventually it would become black, but it's not always quite bright enough to do that in every case.

Answer 1

For starters,

You need to research Infrared light pollution caused by the high temperatures of the materials you're trying to image. Most cameras have an IR filter on the sensor, which blocks incoming wavelengths which are imperceptible to the human eye. These filters often have different characteristics depending on the camera make/model/manufacturer, which is why I say you need to do some research in that area.

It's possible to remove these filters, and photographers who are interested in IR photography often do; but because the glass of the lens bends visible light differently than IR light, there's effectively different focal planes for the two, and you can't focus them both at the same time. So, IR photographers will typically remove the sensor filter, and then attach an IR-pass filter to the front of the lens, which blocks the visible spectrum, and only allows IR light through, allowing the IR image to be properly focused. These IR pass filters are available in different wavelengths. IR pass filters won't be particularly relevant to your situation. However, IR cut filters may be.

When IR photographers remove the IR filter from their sensors, they effectively destroy the camera's ability to take "normal" pictures. This is one reason IR cut filters exist. They screw on to the front of a lens, and cut the incoming IR light, allowing a modified camera to behave normally again. But also (and here is where you might be interested) they supplement the IR-blocking capabilities of normal cameras, which may be suffering from IR pollution for various other reasons. Pointing a camera at hot metal isn't the typical scenario; more commonly, cheap Neutral Density filters and long exposures introduce IR light that the sensor filter wasn't designed to handle. However, ensuring that only the spectrum you want to capture is striking your sensor should help in your situation.

Other than that, make sure your white balance is set, because that can also tint the image, and get a good quality color chip chart for reference; most of them are made from plastic and wouldn't stand up to high temperatures, but there are some metal ones. You'd still need to keep it from melting, and be sure the dyes on the color chips were stable at whatever temperature you needed them to be.

Since you're pointing a camera at something very bright, you should be aware of the phenomenon known as aperture diffraction, which will affect the image's sharpness, but not so much overall color (although the colors of edges might be affected slightly; see halation, fringing, and chromatic aberration).

You might also be interested in a more full-spectrum camera, designed to image light invisible to humans, depending on your budget. I think Flir might make some, maybe? They're more commonly used in manufacturing and agriculture.