1.3. Sensors and Image Quality

How Image Sensors Work

Inside every digital camera there is an image sensor, a rectangular chip that converts light into an electronic signal. Instead of film, the sensor is what actually records the scene.

The sensor is covered with millions of tiny light-sensitive sites called photosites. Each photosite corresponds to one pixel in the final image. When you press the shutter, the camera opens the shutter curtains for a set time. During this time, light from the lens hits the sensor. Each photosite gathers photons and stores a small electrical charge that is proportional to the amount of light received.

On their own, photosites only measure brightness, not color. To record color, almost all cameras use a color filter array on top of the sensor, usually a Bayer pattern. This pattern arranges red, green, and blue filters in a repeating grid. Each photosite only sees one color of light. After capture, the camera uses a process called demosaicing to combine the data from neighboring red, green, and blue filtered sites to create full-color pixels. This is part of what turns the raw sensor data into a usable image.

The sensor then converts the analog charge at each photosite into a digital value using an analog to digital converter (ADC). Brighter areas of the scene produce larger charges and therefore higher digital values. Dark areas produce smaller charges and lower values. These digital values form the basis of the image file that the camera saves as RAW or JPEG.

The quality of this capture depends on several key sensor characteristics. Two of the most important are the size of the sensor and the size of each photosite. Larger photosites can collect more light before they fill up, which generally means better performance in low light and less noise. The layout of the color filter array, the efficiency of the sensor at turning light into electrical charge, and the quality of the electronics also affect image quality, but you do not control those directly as a beginner. What you can understand and choose is sensor size, and how it affects your lenses, depth of field, and noise.

Sensor Sizes

Not all sensors are the same size. Sensor size is the physical dimensions of the chip, measured in millimeters. Larger sensors collect more total light at a given exposure, which influences image quality, depth of field, and field of view.

Here are some common sensor sizes:

Full-frame is based on the size of 35 mm film, so many traditional lens focal length numbers make the most intuitive sense on this size. APS-C and Micro Four Thirds are smaller, so they see a narrower field of view with the same lens. This is described more precisely with the concept of crop factor in the next section.

Larger sensors have several practical effects. At the same framing and f-number, they produce a shallower depth of field, which can make background blur easier to achieve. They also tend to have better performance at high ISO, because for the same resolution they often have larger photosites that gather more light. However, larger sensors require larger lenses to cover them, which makes the whole system bigger, heavier, and usually more expensive.

Smaller sensors make cameras and lenses more compact and affordable. They increase the effective reach of telephoto lenses, which is helpful for wildlife or sports. On the other hand, they make it harder to get very blurred backgrounds at the same framing and aperture, and they usually show more noise at high ISO compared to a larger sensor with similar technology.

There is no single best sensor size. The choice is a trade-off between image quality, depth of field control, size, weight, and cost.

Crop Factor

Crop factor is a way to compare the field of view of different sensor sizes to full-frame. It tells you how much smaller or larger a sensor is relative to full-frame in terms of how much of the scene it captures.

Most popular systems with smaller sensors quote a crop factor like this:

Crop factor is defined as:

$$\text{Crop Factor} = \frac{\text{Diagonal of Full-Frame}}{\text{Diagonal of Smaller Sensor}}$$

You do not need to calculate it yourself, but it is useful to understand what it does. To find the full-frame equivalent focal length of a lens on a smaller sensor, you multiply by the crop factor.

For example, a 50 mm lens on a 1.5x APS-C camera gives a field of view similar to a 75 mm lens on full-frame, because $50 \times 1.5 = 75$. A 25 mm lens on a Micro Four Thirds camera (crop factor 2x) behaves like a 50 mm lens on full-frame in terms of field of view.

Crop factor does not physically change the lens. The focal length printed on the lens barrel is always the true focal length. The smaller sensor simply captures a smaller portion of the image circle projected by the lens, like cropping the edges off a full-frame image. That is why it is called crop factor.

In practice, crop factor affects your choice of lenses. If you want a wide field of view on a small sensor, you need a shorter focal length lens. If you want more reach for wildlife, a smaller sensor effectively makes your telephoto lenses feel longer in terms of field of view. Crop factor also influences depth of field when you match framing between different sensor sizes, but the core idea to remember here is that crop factor is about field of view, not physical magnification.

Megapixels and Resolution

Megapixels describe how many pixels an image has. One megapixel is one million pixels. Resolution is the amount of detail that can be recorded and displayed. For a given sensor, more megapixels mean more pixels across the width and height of the image, which can potentially show more fine detail.

If a camera has an image size of 6000 pixels by 4000 pixels, the total megapixels are:

$$\text{Megapixels} = \frac{6000 \times 4000}{1{,}000{,}000} = 24 \text{ MP}$$

Higher resolution lets you print larger images at a given print quality and gives more room for cropping while still having enough pixels for sharing or printing. For example, 24 MP is usually plenty for large prints and gives more than enough resolution for web use. Modern cameras often range from about 16 MP to 60 MP or more.

However, more megapixels are not always better in real-world use. Increasing megapixels on the same sensor size means each photosite becomes smaller. Smaller photosites can collect less light before they fill up, which can increase noise at high ISO and may slightly reduce dynamic range, depending on sensor design. Also, very high resolution files are larger, which uses more storage and can slow down your editing.

Lens quality also matters. A low quality lens may not resolve enough detail to take full advantage of a very high resolution sensor. In that case, extra megapixels provide less visible benefit.

For most beginners, a camera with 16 to 24 MP is more than enough. It gives good flexibility for cropping and printing without overwhelming your storage. Very high megapixel cameras are useful for large commercial prints or heavy cropping, but they are not required for excellent images.

Dynamic Range

Dynamic range describes how wide a range of brightness levels a camera can record from the darkest shadows to the brightest highlights in a single exposure. A sensor with high dynamic range can capture detail in very bright areas and very dark areas at the same time.

In technical terms, dynamic range is measured in stops. One stop is a doubling or halving of light. If a camera can distinguish detail across 12 stops between pure black and pure white, it has more dynamic range than a camera that can only manage 8 stops.

Dynamic range is most important in high contrast scenes, for example a landscape with a bright sky and dark foreground, or a portrait with strong backlighting. If the scene brightness exceeds the sensor’s dynamic range, something must clip. Either highlights will blow out to pure white with no detail, or shadows will block up to pure black.

You see the effect of dynamic range when you try to recover details in editing. With a file from a sensor with high dynamic range, you can often lift shadows or pull back highlights without the image falling apart. If dynamic range is lower, shadow recovery introduces more noise and highlights may already be clipped with no detail to recover.

Sensor size, pixel size, and sensor design all affect dynamic range, but you cannot change them on a given camera. What you can control is how you expose the image and what exposure mode you use. Shooting RAW generally preserves more dynamic range than shooting JPEG, because RAW files store more tonal information. Some cameras also have features like highlight priority or dynamic range optimization that adjust exposure and processing to make better use of the available range.

As a beginner, you do not need to worry about the exact number of stops. Just be aware that dynamic range is about how forgiving your files are when dealing with bright skies and deep shadows in one shot and that shooting RAW and exposing carefully help you get the most from your sensor.

Digital Noise

Digital noise is the random, grainy or speckled variation that appears in images, especially in dark areas or when you use high ISO. Noise is similar to film grain, but it often looks less pleasant and can hide fine details or create colored specks.

Noise comes from several sources. The main ones are the random nature of light itself and the electronic noise of the sensor and its circuitry. When light levels are high, the signal from the scene is strong compared to the noise, so the image looks clean. In low light, the signal is weak and the noise becomes more visible.

ISO plays a major role in how obvious noise is. When you raise ISO, the camera amplifies the sensor’s signal to make the image brighter. However, it also amplifies the noise. This is why very high ISO settings often result in grainy images.

Sensor size and pixel size affect noise as well. Larger sensors with larger photosites can collect more light at the same exposure, which usually leads to less noise for the same ISO and exposure settings. Modern sensors have improved so much that many cameras handle high ISO much better than older models.

There are two main types of noise you will often see discussed in editing:

You can reduce visible noise in several ways. First, use the lowest ISO that still allows a sharp image. Second, expose properly. Very underexposed images that you brighten heavily in editing often look much noisier than images that were exposed well in-camera. Third, use noise reduction in editing software, which can smooth out noise. Strong noise reduction, however, can also blur fine detail, so it is a balance.

Noise is normal and sometimes unavoidable, especially in low light. A slightly noisy but sharp and well-timed image is usually better than a clean but blurred or missed shot. Learning how your camera behaves at different ISOs helps you decide what trade-offs you are comfortable with in different situations.

Views: 10