It becomes quite difficult to understand the perception of Phase Detection Autofocus in DSLR Technology. To understand the perception of Phase Detection Autofocus, it is important first to understand how DSLR camera works.
Whenever a picture is taken, the light rays tend to enter the lens and enter the camera. The camera has a mirror that is partially transparent, and this mirror is set at an angle of 45 degrees.
This is why the light tends to reflect vertically through the pentaprism. The vertical light is converted into horizontal light by the pentaprism and is reverted. Some light also enters the main mirror and is easily reflected using the secondary mirror.
Phase Detection Autofocus Sensor
Finally, the light can reach the Phase Detection Autofocus sensor which tends to direct the light to a group of sensors. This is the point when the camera tends to analyse the images, and if it appears that the images are not identical, and then the lens has to make adjustments to get the proper image.
Though this process appears to be pretty simple, there is one major issue. The lens has to make an adjustment to the image as per the Phase Detect sensor. The interesting part is that the image is captured using a completely different device. It is the sensor that is present at the back of the camera.
What happens is that when you take the picture, both the secondary and the main mirror tend to flip up and then the shutter opens. The light tends to enter through the lens and immediately hits the camera sensor.
Phase Detection Autofocus can only work the correct way if the distance between the camera sensor and the lens mount as well the distance between the phases detect sensor and the lens mount is identical. Even a slight deviation can cause autofocus to work in an inaccurate way.
The working process of Phase Detection Autofocus
When you look at the Phase Detection Autofocus system, it works more like the rangefinder cameras. Light is bounced off the secondary mirror. This light is then received by two small image sensors which have micro lens above them. Every focus point has about two sensors for the phase difference.
Light tends to enter the sensors, and if an object is in focus at that time, then the light rays from lens extreme sides converge in the centre of the sensor. When both the objects tend to have identical looking images, then this means the object is in a state of perfect focus.
The intricacies of Phase Detection Autofocus
The light will not be able to converge if the object is not in focus. The light will hit the different sides of the sensor in such a situation. The positive aspect about phase detection systems is that they can exactly identify if an object is back focused or front focused. It sends instructions to the camera lens regarding how much focus is required and how the lens should turn the focus.
This is what happens. Light enters the lens. The way the light reaches the sensors helps the AF system to determine whether an object is back focused or front focused. The Autofocus sends instructions to the lens for adjusting the focus. This process continues till the perfect focus is achieved.
If still the focus cannot be achieved then the lens resets itself and tends to reacquire the focus. When the perfect focus is attained, then a confirmation is sent by AF system that the object is now in focus. This literary takes a fraction of a second. This is why phase detection is termed to be a fast process.
Well, the conclusion is that Phase Detection Autofocus is a complex process. This is why phase detection autofocus systems need to be installed and aligned properly during manufacture.