Report itu-r bt. 2053-2 (11/2009) L
Download 1.02 Mb.
|
- Navigate this page:
- 2.4 8k 4k camera (CCD)
- 2.5 8k 4k camera (CMOS)
2.3 4k 2k cameraOlympus Optical Co., Ltd. has developed an experimental 4k 2k camera in May 2002. The HDTV camera employed four 2/3 inch FIT CCDs (R, G1, G2, B) with a 30 fps progressive scanning. In order to achieve 4k 2k resolution, the two green panels were arranged by the diagonal-pixel-offset method. CRL (Communications Research Laboratory) and JVC (Victor Company of Japan) have jointly developed a camera system with 2 000 scanning lines called Quadruple HDTV. The camera system employs three cmOS sensors of 3 888 2 192 pixels (2 inch equivalent) and outputs the video signals in four HDTV signal channels. The camera scanning is at 30p (30 fps with progressive scanning) or 60i (60 fps with interlaced scanning). The sensitivity of the camera was slightly worse than that of a normal HDTV camera. NHK has developed an experimental 4k 2k colour camera operating in a 2 250-line, 60 frame progressive system [Mitani, K. et al., February/March 2002]. Its main features are listed in Table 2. It is equipped with 8 million pixels CCD sensors (2.5 inch equivalent). Since the CCD has four times as many pixels as the 1 080p system, a very high-clock frequency of approximately 600 MHz would be required. To overcome such a difficulty, the image area of the CCD is divided into 16 sub-areas, each of which is independently scanned at the relatively low-clock frequency of 37.125 MHz. The signals from each sub-area are then combined to result in the full picture [Smith, C. et al., June 1999]. TABLE 2 Main features of a 4k 2k camera developed by NHK
2.4 8k 4k camera (CCD)In 2002, NHK realized a 4 000 scanning line image system [Mitani, K. et al., January 2003], i.e. with a number of scanning lines four times higher than that of present HDTV systems. Table 3 shows the 4 000-scanning line image system signal formats. A 4 000-scanning line camera system with a 16:9 aspect ratio would need an imaging device with a capability of about 32 million pixels at 60 fps, which is not yet within current technological feasibility. Therefore, the image is processed by dividing it into signals of the size of an HDTV signal, and the current HDTV signal processing circuit and peripherals are applied. The 4 000 scanning line imaging system uses four 8 million pixel CCDs (2.5 inch equivalent) for acquisition of the motion images. As shown in Fig. 1, the optical image passing through a lens is divided by a colour separation prism into three different spectra: two green (G1, G2), red (R), blue (B). Each forms an image on its own 8 million pixel CCD. The four CCDs are located in the prism with a half pixel offset, see Fig. 2. The result is a resolution equivalent to that of a 4 000-scanning line signal CCD colour camera that shoots in colour by using 32 million pixels (8 million pixels 4). The external appearance of the prototype camera head is shown in Fig. 3. The camera head weighs 76 kg, and it consumes approximately 600 W. It has a resolution limit of 2 700 TV lines or higher. TABLE 3
Signal format of a 4 000 scanning lines image system
2.5 8k 4k camera (CMOS)NHK has developed a new 8 megapixels CMOS sensor for a 1.25-inch optical format [Yamashita, T. et al.] in the framework of camera head miniaturization associated to the development of a zoom lens. A pixel number of 3 840 (horizontal) 2 160 (vertical), i.e. twice (in both directions) that of HDTV images was chosen. Figure 4 shows a block diagram of this sensor architecture. The sampled pixel values from the pairs of columns are transmitted to conventional 10-bit analogue-to-digital converters (ADCs) in each parallel column. Then the pixel values are stored in SRAM banks. Digital data are read out from this SRAM bank by 16 parallel ports. Optical black pixels are positioned around the periphery of the active area. The optical black pixels on the top and bottom of the sensor are used to compensate the difference between the offset levels of each of the columns, while those on the right and the left of the sensor are used to compensate the difference between the offset levels of each line. Table 4 shows a comparison between the specification of the CCD sensor on the first model of the 4k-scanning-line camera and that of this CMOS sensor. The pixel size of the CMOS sensor is 4.2 4.2 μm2, which is smaller than that of the CCD sensor. However, due improved photoelectric transfer characteristics, the sensitivities of these sensors are nearly the same. About 2 000 ADCs are built in, and 16 pixels can be output from the sensor in parallel as a digital signal during one clock period. The reasons for using a CMOS sensor are its low power consumption and the feasibility of column parallel output structure. A wide bandwidth sensor, as used in ultra-high-definition systems, should have a multiple-output structure. Since the CMOS sensor can output arbitrary pixels, it is hard to see the difference in gain or offset. On the other hand, the CCD sensor is better suited for the area-divided output. In this case, it is easy to see the difference in output for each region. Directory: dms pub -> itu-r -> opb -> rep rep -> Report itu-r m. 2243 (11/2011) rep -> Report itu-r m. 2038 Technology trends rep -> Report itu-r sa. 2312-0 (09/2014) rep -> Report itu-r rep -> Report itu-r sm. 2179 (09/2010) Short-range radiocommunication devices measurements rep -> Report itu-r rep -> Report itu-r rs. 2194 (10/2010) rep -> Report itu-r rs. 2165 (09/2009) rep -> Rep. Itu-r bs. 2104 Download 1.02 Mb. Share with your friends:
| ||||||||||||||||||||||||||||||||||