The working principle and application of anti shake cameras
Anti shake cameras are mainly used in some sports scenes to prevent the captured images from being unreal. At present, image stabilization technology is divided into two categories: mechanical stabilization and optical stabilization. Mechanical image stabilization is the process of reducing the illusionary images generated during lens shaking through mechanical means, while optical image stabilization is the process of correcting the illusionary images generated during shaking through electronic or software algorithms
This article mainly introduces optical image stabilization, and mechanical image stabilization will be introduced separately
Optical Image Stabilization OIS Optical Image Stabilization
In the photography industry, there is a concept of “safety shutter”, which means that the maximum shutter time is less than=1/equivalent focal length (seconds). For example, the iPhone 7 Plus has an equivalent focal length of 31mm and a safety shutter speed of 1/31s. However, in order to ensure nighttime imaging quality, the shutter speed of the iPhone without optical image stabilization is set at 1/17s, which is greater than the safety shutter speed. Therefore, electronic image stabilization compensation can only be relied upon during night photography, which can easily cause shaking.
With optical stabilization, the “optical axis offset” is corrected through the floating lens of the lens. The principle is to detect small movements through the gyroscope inside the lens, and then transmit the signal to the microprocessor. The processor immediately calculates the displacement that needs to be compensated, and then compensates the lens group according to the direction and displacement of the lens shake; Thus effectively overcoming the image blur caused by camera vibration.
With the help of this technology, the impact of “hand tremors” can be effectively offset. Therefore, manufacturers can extend the “safety shutter” time, for example, Apple and Samsung have extended the shutter speed to 1/4s, ensuring brightness and frame rate for night photography.
Of course, the addition of optical image stabilization also brings some side effects. On the one hand, the resolution of the image edges decreases significantly, and on the other hand, the floating lens group occupies more space. Therefore, many OIS equipped mobile phone cameras are very moving due to their “protrusion”.
OIS module control
The basic principle of optical image stabilization:
When the gyroscope is subjected to external vibrations, the vibration signal is fed back to the OIS control to drive the OIS motor for corresponding compensation of X&Y, thereby offsetting the image offset caused by shaking and ensuring that the camera can still maintain imaging stability in a shaking environment.
1) Gyroscope
Function: Sensing jitter
Application: Measure the tilt angle caused by shaking, and the system will predict the image offset caused by tilt based on this angle.
2） OIS driver IC
Function: Calculate compensation angle, command voice coil motor to adjust lens angle
Application: Convert the tilt angle information caused by gyroscope vibration into electrical signals to control the compensation of X and Y axes, ensuring stable imaging.
3) Hall sensor
Function: Electromagnetic induction
Application: Feedback the position information of the lens back to the OIS driver IC, forming a closed-loop control to accurately move the lens to the desired position.
4) Hall effect
By passing a controlled current I through both ends of a semiconductor wafer and applying a uniform magnetic field with a magnetic induction intensity of B in the direction perpendicular to the semiconductor wafer, a potential difference UH is generated in the direction perpendicular to the current or magnetic field. This voltage is called the Hall voltage, and this effect is called the Hall effect.
OIS calibration process
The calibration process of OIS is divided into:
The OIS testing process consists of three steps: Hall calibration, OIS calibration, and OIS verify
Electrical Image Stabilization EIS Electronic Image Stabilization
If you are careful, you may find that the top ranked Google Pixel phones are not equipped with OIS optical image stabilization technology. So, how do these phones rely on image stabilization?
Relying on EIS image optimization, this is a new term that was hyped up in OPPO R9s and later developed by Xiaomi Note2. Although the English name sounds high-end, it actually translates to electronic image stabilization.
Electronic image stabilization, which relies on software algorithms to achieve image stabilization, has always existed and was first applied in the recording mode of electronic cameras: by cutting the surrounding four sides of the image to a certain extent as a “buffer zone”, the image shake is correspondingly compensated in reverse.
In terms of camera imaging, with the mature application of mobile phone multi frame synthesis imaging algorithms, it is now possible to use the correlation between frames in the image sequence to replace and compensate for shaky pixels through software algorithms. Flagship phones like Pixel can use the motion of the gyroscope to estimate and compensate for image sequence jitter caused by phone shaking.
That is to say, EIS is achieved by cropping the image. Generally, the image is taken from the center position of the photosensitive chip, and when the phone shakes, the cropping position is adjusted on the photosensitive chip to compensate for the image and achieve image stability.
AIS anti shake
AIS anti shake is a fusion of AI algorithms in traditional anti shake algorithms, integrating the advantages of EIS and OIS anti shake. Through AI algorithms, camera motion is estimated to obtain smoother and more accurate positions, thereby achieving better video stabilization effects.
AIS is digital image stabilization, where the CPU works in conjunction with the acceleration sensor and direction sensor of the phone to analyze the shaking of the phone during photography, the direction and speed of the phone’s movement in the hand, and calculate the degree of image shaking. After correction by the image processing engine, a JPG file is generated to achieve image stabilization effect. AIS greatly tests the processing power of CPUs, which ordinary mobile phone CPUs do not have