As user requirements continue to increase, traditional CCD technology is no longer able to meet current users' needs for digital images. In order to cater to user needs and occupy the market, some manufacturers have introduced several new CCD technologies in recent years:
(1) and INTERLINE TRANSFER CCD
Typical consumer-grade digital Cameras generally use INTERLINE TRANSFER CCD. Its structure is as shown below. Integrate a semiconductor into a photosensitive device: a photodiode, and some circuits. Each unit is arranged in a neat matrix, with as many rows as there are columns. The number of rows multiplied by the number of columns is the number of pixels of this CCD. In each pixel unit (small picture in the lower left corner), about 30% of the area is used to manufacture the photodiode (red part). In the remaining available area, a SHIFT REGISTER (purple part, shift register) will be placed. After receiving a command, the light intensity felt by the photodiode will be placed in this SHIFT REGISTER and maintained. This is an analog quantity.
The next step is to convert the light intensity value in each pixel into a digital quantity, and then combine it into a digital image by the processor in the camera. First, the parallel clock starts the first row; the serial clock starts the 1st, 2nd, 3rd... columns in sequence. In this way, each pixel in the first row is sent out of the CCD in sequence and enters the A/D CONVERTER (analog/digital converter, this device is specially used to convert analog quantities into digital quantities). Then the parallel clock starts the second row; the serial clock starts the 1st, 2nd, 3rd... columns in sequence. In this way, each pixel in the first row is sent out of the CCD in order and enters the A/D CONVERTER. In this way, the pixels in each row and column are converted into digital signals in an orderly manner. The camera's processor then combines these digitized pixels into a digital image.
The SHIFT REGISTER in each pixel unit is neatly arranged in rows, sandwiching the photodiode that actually acts as a light sensor. So this device is called: INTERLINE TRANSFER CCD. Since the actual photosensitive area of ??each pixel unit only accounts for about 30%, its photosensitive efficiency is relatively low. Therefore, in the real finished product, a MICROLENSES (micromirror) will be built on top of each pixel unit. The schematic diagram of MICROLENSES is in the lower left corner of the picture. The optical lens is directly above the photodiode and has a relatively large area, so that more incident light can be concentrated on the photodiode, making the equivalent photosensitive area reach about 70% of the pixel area.
Due to the existence of SHIFT REGISTER, INTERLINE TRANSFER CCD does not require a mechanical shutter. Use an electrical signal to instruct SHIFT REGISTER to hold the output signal of the photodiode, and the sampling process has been completed. This is the electronic shutter. The existence of SHIFT REGISTER also enables INTERLINE TRANSFER CCD to output video signals. The fact that we can see moving images on the color LCD viewfinder is also a result of SHIFT REGISTER.
The CCD used in KODAK professional products is FULL FRAME TRANSFER. In each pixel unit, 70% of the area is used to manufacture photodiodes. The entire pixel frame is almost entirely photosensitive area. There is no need and no way to place a larger area of ??MICROLENSES to increase its lighting. Its reading sequence is the same as INTERLINE TRANSFER CCD. The advantage of this structure is that the largest possible photodiode can be obtained to achieve better imaging quality. It can be said that with the same CCD area, FULL FRAME will definitely have better performance. Disadvantages: This CCD cannot input VIDEO images. You cannot use the LCD screen as a viewfinder. Must work with a mechanical shutter. And the mechanical shutter limits its maximum shutter speed.
NIKON D100 uses a full frame (FULL FRAME TRANSFER) CCD. Compared with an intermediate column transmission (INTERLINE TRANSFER) CCD, the full frame transmission CCD has a smaller effective pixel of each photodiode in the photosensitive device. The area is larger, allowing more image data to be captured. Generally speaking, the effective image data that a full-frame transmission CCD can capture is about twice that of a middle column transmission CCD, thus having the advantages of greater dynamic range, lower noise, and higher gray-level sensitivity, thereby improving Detailed performance in shadows and highlights.
(2) SUPER CCD
From the above articles, we can understand that the arrangement of CCD photosensitive points is the key to affecting the photosensitive range and dynamic capabilities of CCD. Early CCDs were all in an orderly "farming" shape. When CCD technology came into the hands of Japan's Fujifilm, engineers began to think about whether the CCD must be arranged like this? In order to combine the low-cost design of the INTERLINE TRANSFER CCD with the large photosensitive area of ??the FULL FRAME CCD, Fujifilm proposed a compromise solution, the SUPER CCD, that fell below the expert glasses. SUPER CCD is currently the only CCD on the market that uses a honeycomb structure. It relies on octagonal geometric structure and discontinuous arrangement, and is based on the INTERLINE TRANSFER CCD approach to maximize the effective area utilization of the CCD. However, the earlier technology made the channels too crowded and produced undesirable noise. Today, SUPER CCD has developed into the third generation, and almost all the undesirable shortcomings have been improved.
In early 2002, Fujifilm released the third generation Super CCD. In early 2003, Fujifilm released the fourth generation Super CCD. (See picture below). The new generation of SuperCCD has two specifications: Super CCD HR and SR. Super CCD HR (High Resolution) emphasizes Fuji's patented technology to further improve the resolution on a fixed-area CCD chip. HR technology can create 6.63 million pixel photosensitive elements on a 1/1.7-inch CCD. Digital cameras equipped with a new generation of HR photoreceptors will be able to output photos with 12.3 million recording pixels (just like the old 3 million pixel SuperCCD can output The output effect of this HR CCD will be comparable to the image quality effect of Fujifilm's current flagship S2PRO.
The other Super CCD SR is a new CCD structure. Like the HR, the CCD SR uses new miniaturization technology and can create 6.7 million pixel elements on a 1/1.7-inch CCD. (HR is 6.63 million). The difference is that SR emphasizes a higher dynamic range (Dynamic Range), which is claimed to be more than 4 times that of previous products. The main key to this difference is that CCD SR adopts a new structure that is different from the past: SR integrates S pixels responsible for high sensitivity (see picture: larger area) and R pixels (see figure: larger area) that can operate outside the general dynamic range. smaller area). By integrating the calculations of these two different pixels, SuperCCD SR will achieve higher sensitivity and wider dynamic range than previous CCDs with a single photosensitive structure. In the past, single-architecture photosensitive originals were sensitive to areas outside the dynamic range, that is, the highlights and dark areas. Because the sensitivity cannot be adjusted to adapt (the display quality in the middle range must be taken into account), it is painful to lose this part of the details. Traditional negative films can overcome this problem by coating finer color-sensitive photosensitive particles. Therefore, when digital images are compared with traditional images, dynamic range is often the key to traditional victory. Fuji's new technology obviously overcomes the noise interference caused when the originals are denser. SR technology uses 3.35 million S pixels and 3.35 million R pixels to integrate into 6.7 million performance. This method of division of labor and cooperation is still used in the industry. First case.
In February 2002, the American company Foveon released multi-layer color-sensitive CCD technology.
Before Foveon released the X3 technology, the structure of a CCD was generally similar to a honeycomb-shaped color filter (see the picture below), with a photoreceptor underneath to determine which of the three RGB primary colors the incident light was.
However, the disadvantages of cellular technology (also known as mosaic technology in the United States) are: the resolution cannot be improved, the color discrimination ability is poor and the production cost is high. For this reason, the production of high-end CCDs has been monopolized by Japan over the years. The new X3 technology allows electronic technology to successfully imitate the color sensing principle of "real film" (see the picture below), sensing color layer by layer according to the absorption wavelength of light. Corresponding to the shortcomings of cellular technology that one pixel can only sense one color, X3 has the same One pixel can sense three different colors, greatly improving image quality and color performance.
X3 also has another feature, that is, it supports VPS (Variable Pixel Aize), a more powerful CCD computing technology. Through the combination of "group pixels" (see the picture below). X3 can achieve ultra-high ISO values ??(resolution must be reduced) and high-speed VGA animation recording. What's more powerful than Super CCD is that each pixel of X3 can sense three color values. In theory, X3's animation shooting may be more refined than SuperCCD III under the same speed conditions. The characteristic of this invention is that traditional digital cameras mainly use 3 primary color filter matrices to produce 3 different color intensities for each light point (or pixel PIXEL): red (R), green (G) and blue (B) ) color data, which are then integrated with a color TV or monitor to form the image we see. However, according to experiments, it is pointed out that the human visual system is more sensitive to green than other red and blue colors, which also makes the traditional CCD matrix adopt a color ratio of 25% red and blue and 50% green. However, color differences cannot be corrected in such a ratio. The reason is that human vision is closer to analog effects rather than cut into digital levels. In order to make the colors of the scenery more realistic, this technology of SONY effectively guides and samples dark green and light green respectively! It is of great help to the faithful regeneration of green.