OIS, EIS and Other forms of Image stabilization extensively explained......Complete with comparison of the human eye
One particular feature we always notice when viewing the camera modules of Flagship devices is the term called OIS(Optical Image Stabilization). Another feature that we tend to see in mid-range and budget phones that are almost used as a replacement to the OIS is the next term called EIS(Electronic image Stabilization). There are other forms of Image Stabilization, like the AIS and Hybrid stabilization, which we will treat in todays article.
All these stabilization methods are out for one function, to literally stabilize. You might have noticed, when you want to take a picture of an object and you tilt your hand a little, the picture comes out blurry. Now, there are mechanical workings that occur on the inside when you're trying to take a photo. To learn the terms that tally with our present topic, let's quickly learn about Shutter speed and Aperture.
1.0: Shutter speed:
In photography, a shutter is a device that allows light to pass for a determined period, exposing photographic film or a photosensitive digital sensor to light in order to capture a permanent image of a scene. Think of it like a window that stays closed till you fire up your camera.
Shutter speed is the length of time your camera shutter is open, exposing light onto the camera sensor. Essentially, it’s how long your camera spends taking a photo. This has a few important effects in how your images will appear. It can actually range from half-seconds to minutes. The Higher the shutter speed, The lesser the light that will enter the lens and vice versa to this effect.
The problem with lower shutter speeds is that the highly sensitive sensor will move slightly when capturing the image or video without gimbals and tripods and causes blur and jitter in the pictures and videos respectively.
The problem with lower shutter speeds is that the highly sensitive sensor will move slightly when capturing the image or video without gimbals and tripods and causes blur and jitter in the pictures and videos respectively.
2.0: Aperture:
In optics, an aperture is a hole or an opening through which light travels. More specifically, the aperture and focal length of an optical system determine the cone angle of a bundle of rays that come to a focus in the image plane.
The aperture is measured in f-number which is usually designated as f/1.8, f/2.2, f/2.4 etc. The smaller the f-number, the more the light entering the lens and results in much better photos in low light conditions.
Fact: The Shutter speed, Aperture and the ISO form the three most important terms in photography. But ISO isn't concerned with Stabilization.
3.0: Image stabilization: How does it work?
Image stabilization broadly refers to a number of techniques employed to reduce blurs and jitters in the photos and videos caused by the motion of the camera during the exposure period of the sensor to the subject.
Shutter speed plays a significant role in the need for image stabilization. If the shutter speed is too low, the time taken for achieving a steady shot is reduced.
If the shutter speed is very high, there is more time for you to take a steady shot but a slight shake during the exposure will be amplified, resulting in a blurry image or a jittery video output.
Image stabilization is utilized to fix these problems by granting you a higher shutter speed without the shakes and jitters. Now, there are different types of image stabilization utilized in smartphones. Lets explore them:
3.1: Types of Image Stabilization:
Each of these stabilization types might be available on smartphones based on the budget and make of the mobile itself. These are:
- Optical Image Stabilization
- Electronic Image Stabilization
- Artificial Image Stabilization
- Hybrid Image Stabilization
3.1.1: Optical Image Stabilization (OIS):
OIS steadies the shakes and movements by the use of a sophisticated device inside the camera lens.
3.1.2: Elements/Setups of Optical image stabilization:
It is a mechanism used in a still camera or video camera that stabilizes the recorded image by varying the
optical path to the sensor.
The key element of all optical stabilization systems is that they stabilize the image projected on the sensor before the sensor converts the image into digital information. This is done by understanding the camera's movement using sensors such as gyroscopes, and speculating how the lens needs to move to correct the anomaly. The lens module is then generally moved sideways or up and down, normally by using electromagnet motors. All of this happens as the image is recorded to reduce camera movement blur. Definitely, this has to be done by using a physical contraption housed in the mobile phone that tries to make up for the shakes you make when you try to use the camera. This contraption is known as the VCM (Voice call motor). Usually employed in most phones released in the market today. Although it is gradually being replaced by the latest means, some of which is the new Mems, which promises to fix the issues faced with the former.
optical path to the sensor.
The key element of all optical stabilization systems is that they stabilize the image projected on the sensor before the sensor converts the image into digital information. This is done by understanding the camera's movement using sensors such as gyroscopes, and speculating how the lens needs to move to correct the anomaly. The lens module is then generally moved sideways or up and down, normally by using electromagnet motors. All of this happens as the image is recorded to reduce camera movement blur. Definitely, this has to be done by using a physical contraption housed in the mobile phone that tries to make up for the shakes you make when you try to use the camera. This contraption is known as the VCM (Voice call motor). Usually employed in most phones released in the market today. Although it is gradually being replaced by the latest means, some of which is the new Mems, which promises to fix the issues faced with the former.
As OIS is definitely a physical device, there are some components and constituents that make up the mechanism proper. Now these also double as types of OIS systems, so lets identify and explain them.
3.1.2.1: Lens-based or Floating lens set-up:
Floating lens setup is usually found on large camera systems such as the DSLR (Digital Single Lens Reflex) cameras. These cameras are large enough to employ interchangeable lens and there's enough distance between the modules to place the lens set-up. Of course, this fails in smartphones as phones are not large enough to use them. The main component lens is given enough allowance to move feely to and fro and up and down within the limited space for accurately capturing a stabilized frame. Now, this is an universal stabilizer but lenses with larger focal length tend to provide better stabilization.
3.1.2.2: Sensor-based or sensor shift set-up:
Previously, we explained that lens setup cant be used on smartphones. The sensor shift mechanism is the most common implementation of OIS and found mostly in smartphones where lenses are static and built into the smartphone chassis.
In sensor-shift image stabilization, it’s the sensor that moves to keep the subject in frame and not the lens. These can be called ''in body stabilizers'' and affect the angle of the image sensor in response to stimulus and can also adjust the focal length of the lens as well.
3.1.3: Gyroscope in Stabilization:
Gyroscopes are hardware important for collecting data about the orientation of your smartphone or digital camera and tracks movement of your hand. These are important for implementing correct measures for movements. The yaw, pitch and roll data from the gyroscope is then fed into the microcontroller which then gives the necessary information about how much the lens or the sensor should shift to compensate the movement, based on the intensity.
3.1.4: Axis in stabilization:
Axis is an imaginary line about which a body rotates, or a fixed reference line for the measurement of coordinates. These axis in this context are the X,Y and Z axis. A 2-Axis OIS would resist shakes in the (X,Y) axis while a 3-Axis OIS would resist shakes in the (X,Y) or (X,Z). Now, does this mean that 4-Axis and 5-Axis OIS are gimmicks? No, because the next point is a common(though seen mathematically) rotation around two axis at infinity. This is why it is called four axis. Think of a straight movements of a boat on the ocean, it is strictly not moves along a straight line but rotation around an axis through the centre of the earth.
3.1.2: Electronic image stabilization:
As the name implies, this stabilization technique is done using software means hence there is no need for a physical contraption as in OIS. The Electronic Image Stabilization (EIS) system reduces image shake and controls image stability by manipulating the image electronically, using the light-sensing chip, the Charge Coupled Device (CCD) of the camera. Once the image hits the chip, and if the system detects what it thinks is camera shake through its sensors, it responds by slightly moving the image so that it remains in the same place on the CCD. For example, if the camera shakes to the left, then the image moves to the right to compensate, thus eliminating the shake.
There are two ways the EIS system works to reduce the movement of the image. The first method is to increase the size of the image by digitally “zooming in on the image so that it is larger than the CCD. By making the image larger, the system can then scan within the image to counter the movement created by the shake. To test if your camera utilizes this system, simply turn it on and watch the image through the viewfinder. If the image zooms in slightly, then it is using this method of image stabilization. However, because this system zooms in on the image, the image resolution is somewhat decreased.
The second method of the EIS system uses of an oversized CCD. As a video image covers only about 90 percent of the chip’s area, there is system space available for which to move the image. For example, when the image is stable, the chip centers the image on the CCD. But if the camera shakes to the right, there is space for the image to roam back to the left to compensate for the shake, keeping it in exactly the same place on the CCD, thus eliminating the shake.
There are two ways the EIS system works to reduce the movement of the image. The first method is to increase the size of the image by digitally “zooming in on the image so that it is larger than the CCD. By making the image larger, the system can then scan within the image to counter the movement created by the shake. To test if your camera utilizes this system, simply turn it on and watch the image through the viewfinder. If the image zooms in slightly, then it is using this method of image stabilization. However, because this system zooms in on the image, the image resolution is somewhat decreased.
The second method of the EIS system uses of an oversized CCD. As a video image covers only about 90 percent of the chip’s area, there is system space available for which to move the image. For example, when the image is stable, the chip centers the image on the CCD. But if the camera shakes to the right, there is space for the image to roam back to the left to compensate for the shake, keeping it in exactly the same place on the CCD, thus eliminating the shake.
While capturing a shot or a video, the microcontroller keeps a reference frame in memory for comparison with the successive shots following the first one. This comparison with the preceding frames one after the other helps the microcontroller for recording the variations between them and makes horizontal and vertical arrangements to make the shots look smooth. Now, in the case of video stabilization, if the microcontroller finds that the variation data is towards a single direction, it predicts the succeeding frame path, generating a smooth video, or a still image.
3.1.3: Artificial intelligence stabilization:
Although these features are pretty exciting (at least on paper), a past tear down of the Huawei P20 Pro showed it had OIS on all Three cameras, when it clearly wasn't supposed to.
3.1.4: Hybrid image stabilization:
While not a common stabilization term, is used to describe smartphones that employ two or more of the stabilization techniques. Definitely Today's flagships are a good example.
The Optokinetic reflex is the term used to describe when the eyes automatically tracks moving objects. The Optokinetic reflex (or optokinetic nystagmus) stabilizes the image on the retina through visual feedback. It is induced when the entire visual scene drifts across the retina, eliciting eye rotation in the same direction and at a velocity that minimizes the motion of the image on the retina. When the gaze direction deviates too far from the forward heading, a compensatory saccade is induced to reset the gaze to the centre of the visual field.
For example, when looking out of the window at a moving train, the eyes can focus on a moving train for a short moment (by stabilizing it on the retina), until the train moves out of the field of vision. At this point, the eye is moved back to the point where it first saw the train (through a saccade). But what happens if you rotate your head, or shake it side to side? Unless you're deliberately trying to look at something off to one side, your eyes tend to move less than your head. This is because the Vestibule-Ocular Reflex (VOR) automatically compensates for those movements to stabilize the image on your retina. Without these compensatory eye movements, even small involuntary head movements could blur your vision. An example is that when there is an engine malfunction that causes vibrations, pilots complain of not being able to read the cockpit instruments because their eyes cannot stabilize the vibrations.
So that's the end of our topic covering the image stabilization types in smartphones. Any questions? feel free to drop a comment.
No comments