|
|
| A single Super Hi-Vision (SHV) video frame consists of approximately 33 million (7680 × 4320) pixels. Recently, we constructed a prototype 33M-pixel image sensor, and captured moving picture video with it. Before this development, there was no image sensor capable of shooting a moving picture at this resolution; the conventional method of shooting SHV video was through a camera using a four-imager pickup system with an off set of half a pixel pitch*1 in the diagonal direction. |
- The incorporation of CMOS technology made it feasible to integrate new functions in the image sensor, including high-speed and high-data rate signal output and on-chip analog-to-digital conversion.
- After considering all aspects of camera performance, including lensf resolution and size, image sensor sensitivity, and dynamic range, we determined that the pixel size should be 3.8μm square. LVDS*2 is used for the output signal scheme, because of its suitability for high-speed transmission and low power consumption.
|
| We will work to enhance the resolution of the imaging/display system, as well as its tone reproduction, to attain the full potential of SHV video. We will also conduct studies to examine the relationship between a broadcasting system conveying such strong sensations of reality and human sensory characteristics. |
| *1 : The four-imager pickup system with an offset of half a pixel pitch in the diagonal direction employs a total of four image sensors, two for green, and one each for red and blue. This scheme attains the equivalent of twice the resolution, both horizontally and vertically, by attaching two green imagers to the prism, offset at half a pixel pitch diagonally in relation to each other.
*2 : Low Voltage Differential Signaling (LVDS).
|
33M-pixel image sensor
 |
|
