Samsung Explains PDAF Pixel Masking in its 108MP Nonacell Sensors

Electronic Imaging publishes Samsung paper "A new PDAF correction method of CMOS image sensor with Nonacell and Super PD to improve image quality in binning mode" by Yeongheup Jang, Hyungwook Kim, Kundong Kim, Sungsu Kim, Sungyong Lee, and Joonseo Yim.

"This paper presents a new PDAF correction method to improve the binning mode image quality in the world’s first 0.8um 108 mega pixel CMOS Image Sensor with Samsung Nonacell and Super PD technology.

The conventional PDAF correction method is based on bad pixel-correction (BPC), replacing AF pixel to adjacent normal pixels. Recently, in order to suppress AF artifact, sensor provide an advanced algorithm that detect the directionality of the image pattern near the PDAF pixel area, and determine which area to be referred for the correction. For example, when the algorithm detects no directionality in the pattern, the PDAF pixel will be replaced by referring to all of the same color-pixels within the kernel of 7x7 or 9x9 pixels. If Slash pattern is detected, PDAF correction will refer to the pixels which are located in only slash direction as shown in Table 2. In spite of the advanced algorithm, the AF artifact still occur because of a higher AF density, directional misrecognition or exceptional case of direction.

In order to overcome the limitation of conventional method, new correction method, named Dilution mode have been introduced. In Dilution mode, a Nonacell, which contains AF pixels, outputs its own seed value to a binning output and deliver AF information by embedded data, which is a PDAF Tail mode."



0 Response to "Samsung Explains PDAF Pixel Masking in its 108MP Nonacell Sensors"

Post a Comment