Photosensitive element

Overview

The development path of digital cameras can be said to be the development path of photosensitive components. There are two core imaging components of digital cameras: one is the widely used CCD (charge coupled) device; the other is the CMOS (complementary metal oxide semiconductor) device. Compared with traditional cameras, traditional cameras use "film" as the carrier for recording information, and the "film" of digital cameras is its imaging photosensitive element.

The photosensitive element is the "film" of the digital camera that does not need to be replaced, and it is integrated with the camera, so it is exactly the heart of the digital camera.

The photosensitive element uses the photoelectric conversion function of the photoelectric device. Convert the light image on the photosensitive surface into an electrical signal proportional to the light image. Compared with the photosensitive element of the "point" light source such as photodiode and phototransistor, the photosensitive element is a functional device that divides the light image on the light-receiving surface into many small units and converts it into usable electrical signals.

There are two main types of photosensitive elements: CCD (charge coupled) and CMOS (complementary metal oxide semiconductor). As a new type of shooting function of mobile phones, the built-in digital camera function is the same as the low-end (100,000-1.3 million pixel) digital camera that is usually seen. The photosensitive elements of digital cameras in most mobile phones are basically CMOS. The photosensitive element is also called the image sensor.

Development history

CCD was developed in 1969 by Ball and Smith of Bell Research Laboratory in the United States. In the 1980s, although the CCD image sensor was defective, the difficulties were finally overcome due to continuous research, and a high-resolution and high-quality CCD was manufactured in the second half of the 1980s. In the 1990s, a megapixel high-resolution CCD was produced. At this time, the development of CCD was advancing by leaps and bounds. It has been more than 20 years since the development of CCD. After entering the mid-1990s, CCD technology has developed rapidly, and at the same time, the unit area of ​​CCD has become smaller and smaller. But in order to improve the image quality of the image while reducing the CCD area, SONY developed the SUPER HAD CCD in 1989. This new photosensitive element relies on the magnification of the internal amplifier of the CCD component when the CCD area is reduced. Improve image quality. After that, NEW STRUCTURE CCD, EXVIEW HAD CCD, four-color filter technology (applied exclusively for SONY F828) appeared one after another. The Fuji digital camera uses Super CCD (Super CCD) and Super CCD SR.

For CMOS, it is convenient for mass production, fast, and low cost, which will be the development direction of the key components of digital cameras. With the continuous efforts of CANON and other companies, new CMOS devices are constantly being introduced, and high dynamic range CMOS devices have appeared. This technology eliminates the need for shutter, iris, automatic gain control and gamma correction, and makes it close to the CCD. Image quality. In addition, due to the inherent plasticity of CMOS, a large-scale CMOS photoreceptor with high pixels can be made without increasing the cost. Compared with the stagnation of CCD, CMOS has shown vigorous vitality as a new thing. As the core component of digital cameras, CMOS photoreceptors have gradually replaced CCD photoreceptors, and hopefully they will become mainstream photoreceptors in the near future.

Exmor R CMOS backside lighting technology sensor improves the sensitivity of traditional CMOS sensor. Exmor R CMOS adopts a backside irradiation technology that irradiates light to the side without a wiring layer, contrary to the ordinary method. Since it is not hindered by metal lines and transistors, the aperture ratio (the area ratio of the photoelectric conversion part in a pixel) can be Increase to nearly 100%. Compared with the previous surface-illuminated products with 1.75μm intervals, the back-illuminated products have a great advantage in sensitivity (S/N). Sony’s new Cyber-shot products, WX1 and TX1, adopted a new Exmor R CMOS sensor for the first time in the field of digital cameras. The sensitivity of this Exmor RCMOS sensor is twice that of the same size sensor in the past, so shooting in a low-light environment can greatly reduce noise and obtain clearer images. In the actual tests since then, it is also shown that these two Cyber-shot digital cameras not only provide a high sensitivity of up to ISO 3200, but also have excellent noise suppression capabilities. At the same time, these two digital cameras also provide a series of advanced functions such as handheld night scene shooting and panoramic scanning, which are also technical extensions of the new generation of image sensors. In traditional CMOS sensors, each pixel must be equipped with a corresponding A/D converter and corresponding amplifying circuit. Therefore, this part of the circuit will take up more pixel area, which directly causes the actual photosensitive area of ​​the photodiode to become smaller. Ability becomes weak. The single pixel of CCD does not require A/D converter and amplifier circuit. The photodiode can obtain a larger actual photosensitive area and a larger aperture ratio. Therefore, in the field of small-size image sensors, CCD still occupies a certain advantage. In the field of image sensors, due to the large area of ​​a single pixel, the area occupied by the A/D converter and the amplifier circuit is only a small part of the entire pixel, which has little effect. Therefore, CMOS sensors have also been widely used.

And Exmor R CMOS "places" the photodiode on the top layer of the image sensor chip, and moves the A/D converter and amplifier circuit to the "back" of the image sensor chip instead of traditional CMOS. Like the sensor, the A/D converter and amplifier circuit are located on the upper layer of the photodiode, which "blocks" part of the light. In this way, the light coming in through the microlens and color filter can be utilized by the photodiode to the maximum, and the aperture ratio can be greatly improved. Even a small-sized image sensor can obtain excellent high sensitivity.

In comparison, the photodiode of the traditional surface-illuminated CMOS sensor is located at the bottom layer of the entire chip, while the A/D converter and amplifier circuit are located on the upper layer of the photodiode, so the distance between the photodiode and the lens Farther, the light is easier to lose. At the same time, these circuit connection layers will also block the light path from the color filter to the photodiode, which directly results in less light. The Exmor R back-illuminated CMOS sensor solves this problem.

Classification

CCD

Charge Coupled Device, which uses a high-sensitivity semiconductor material, composed of many photosensitive units, usually millions Pixels are the unit. When the CCD surface is illuminated by light, each photosensitive unit will reflect the charge on the component, that is, the light is converted into electric charge; the signals generated by all the photosensitive units are added together to form a complete picture. It is then converted into a digital signal, compressed and stored in the camera's internal flash memory or built-in hard disk card. Companies capable of producing CCDs are: Sony, Philips, Kodak, Panasonic, Fuji, Sharp, and most of them are Japanese manufacturers.

CMOS

Complementary Metal-Oxide Semiconductor, like CCD, is a semiconductor that can record light changes in digital cameras. The manufacturing technology of CMOS is no different from that of general computer chips. It mainly uses semiconductors made of silicon and germanium to make it coexist with N (charged-charged) and P (charged + charged) grades on CMOS. In semiconductors, the current generated by these two complementary effects can be recorded and interpreted into images by the processing chip. However, the disadvantage of CMOS is that it is too prone to noise. In addition to CCD and CMOS, there is also the exclusive SUPER CCD launched by Fuji. The SUPER CCD does not use conventional square diodes, but uses an octagonal diode, and the pixels are in a honeycomb shape. The form is arranged, and the area of ​​the unit pixel is larger than that of the traditional CCD. As a result of rotating the pixels by 45 degrees, the extra space that is useless for image shooting can be reduced, and the efficiency of light concentration is relatively high. After the efficiency is increased, the sensitivity, signal-to-noise ratio and dynamic range are improved. The arrangement structure of SUPER CCD is tighter than that of ordinary CCD. In addition, the utilization rate of pixels is higher. That is to say, under the same size, the photodiode of SUPER CCD absorbs light relatively high, which makes the sensitivity, signal-to-noise ratio and The dynamic range has been improved.

Difference

The difference between the two photosensitive elements

It can be seen from the working principle of the two photosensitive elements that the advantage of CCD lies in the good image quality. However, due to the complexity of the manufacturing process, only a few manufacturers can master it, resulting in high manufacturing costs, especially for large CCDs, which are very expensive. At the same time, in the past few years, CCD has started from 300,000 pixels and has been developed to 6 million. The improvement of pixels has reached a limit.

At the same resolution, CMOS is cheaper than CCD, but the image quality produced by CMOS devices is lower than that of CCD. Most consumer-level and high-end digital cameras on the market use CCD as sensors; CMOS sensors are used as low-end products in some cameras. If any camera manufacturer uses CCD sensors for cameras, the manufacturer will definitely It spared no effort to use it as a selling point to hype it, and even dubbed it "digital camera". For a time, whether to have a CCD sensor has become one of the criteria for people to judge the grade of a digital camera.

One of the advantages of CMOS image sensors is that the power consumption is lower than that of CCD. In order to provide excellent image quality, the price of CCD is higher power consumption. In order to make charge transfer smooth and reduce noise, The transmission effect needs to be improved by the high pressure difference. However, the CMOS image sensor converts the charge of each pixel into a voltage, amplifies it before reading, and can be driven by a 3.3V power supply, and the power consumption is lower than that of a CCD. Another advantage of CMOS image sensors is that they are highly integrated with peripheral circuits. ADC and signal processors can be integrated together to greatly reduce the size. For example, CMOS image sensors only need one set of power supplies, but CCDs need three or four. Since the manufacturing process of ADC and signal processor is different from CCD, it is difficult to reduce the size of CCD kit. However, the first problem that CMOS image sensors solve is to reduce the generation of noise. Whether future CMOS image sensors can change the fate that has been suppressed by CCD for a long time, the development of technology in the future is an important key.

Application functions

Compared with traditional cameras, traditional cameras use "film" as the carrier for recording information, and the "film" of digital cameras is its imaging photosensitive device, and it is Integrated with the camera. The photoreceptor is the core and the most critical technology of a digital camera. There are two core imaging components of digital cameras: one is the widely used CCD (charge coupled) device; the other is the CMOS (complementary metal oxide conductor) device. The photosensitive elements used in digital cameras in mobile phones are basically CMOS.

Each pixel in a traditional CCD consists of a diode, a control signal path, and a power transmission path. SUPER CCD uses honeycomb-shaped octagonal diodes, the original control signal path is cancelled, only one direction of power transmission path is needed, and the photosensitive diode has more space.

The output pixel of SUPER CCD will be higher than the effective pixel, because CCD is not very sensitive to green, so it is synthesized by G-B-R-G. In fact, some of the real pixels of each synthesized pixel are shared, so the image quality is somewhat different from the ideal state. This is why some high-end professional digital cameras use 3CCD to sense the RGB three-color light separately. The SUPER　CCD achieves the same R, G, and B pixels by changing the arrangement relationship between the pixels, and when synthesizing pixels, three are also used as a group. Therefore, the traditional CCD uses four to synthesize one pixel. In fact, only three are enough, and one is wasted. However, SUPER CCD has discovered this, and only three can be used to synthesize one pixel. In other words, CCD synthesizes a pixel every 4 points, and each point is calculated 4 times; SUPER CCD synthesizes a pixel every 3 points, and each point is also calculated 4 times, so the utilization rate of SUPER CCD pixels is higher than that of traditional CCDs. More pixels are generated.

The price of CMOS with the same resolution is cheaper than CCD, and the image quality produced by CMOS devices is lower than that of CCD. The main advantage of CMOS for CCDs is that they are very power-saving. Unlike CCDs composed of diodes, CMOS circuits have almost no static power consumption and only consume power when the circuit is turned on. This makes the power consumption of CMOS only about 1/3 of ordinary CCD, which helps to improve people's bad impression that digital cameras are electric tigers. The main problem of CMOS is that it overheats due to the frequent current changes when processing fast-changing images. If the dark current is well suppressed, it will not be a big problem, and if it is not well suppressed, it will be very easy to appear noise.

The image data scanning method of CMOS and CCD is very different. For example, if the resolution is 3 million pixels, then the CCD sensor can continuously scan 3 million charges. The scanning method is very simple. It is like passing a bucket from one person to another, and only after the last data scan is completed. Amplify the signal. Each pixel of a CMOS sensor has an amplifier that converts electrical charges into electrical signals. Therefore, the CMOS sensor can perform signal amplification on a per-pixel basis. This method can save any invalid transmission operations, so fast data scanning can be performed with only a small amount of energy consumption, and the noise is also reduced. This is Canon's complete charge transfer technology within the pixel.

Working principle

The charge coupled device image sensor CCD (Charge Coupled Device), which is made of a high-sensitivity semiconductor material, can convert light into electric charge. The analog-to-digital converter chip converts the digital signal into a digital signal. The digital signal is saved by the internal flash memory of the camera or the built-in hard disk card after compression, so the data can be easily transmitted to the computer, and with the help of computer processing methods, according to needs and imagination To modify the image. The CCD is composed of many photosensitive units, and the signals generated by all the photosensitive units are added together to form a complete picture.

Compared with traditional film, CCD is closer to the way the human eye works on vision. It's just that the retina of the human eye is composed of rod cells responsible for light intensity sensing and cone cells for color sensing, and they work together to form visual sensing. After 35 years of development, the general shape and mode of operation of the CCD have been finalized. The composition of the CCD is mainly composed of a mosaic-like grid, a condenser lens, and an electronic circuit matrix at the bottom. Companies capable of producing CCDs are: SONY, Philps, Kodak, Matsushita, Fujifilm, and Sharp. Most of them are Japanese manufacturers.

Complementary Metal-Oxide Semiconductor CMOS (Complementary Metal-Oxide Semiconductor), like CCD, is a semiconductor that can record light changes in digital cameras. However, the disadvantage of CMOS is that it is too prone to noise. This is mainly because the early design made CMOS overheat due to the frequent current changes when processing fast-changing images.

Specifications

Size: the area size of the photosensitive device

Pixels: the more the number of pixels, the larger the size of a single pixel, the more photons captured, and the The better the performance

Signal-to-noise ratio: The higher the signal-to-noise ratio, the clearer the collected image will be.

Size labeling

There are two labeling methods for the size of the photosensitive element, namely optical format (OF, Optical Format) and size type.

Optical format

The optical format is generally expressed by the ratio of the diagonal length of the photosensitive element, that is, OF=diagonal length/1 inch=diagonal length/16mm. It should be noted that 1 inch here is not equal to the usual 25.4mm, but 16mm, that is, when the photosensitive element is 12.8mm×9.6mm, it is a 1-inch photosensitive element. This marking method is mostly used in pocket digital cameras and consumer-grade digital cameras with an aspect ratio of 4:3. The size of the photosensitive element ranges from 1/5 inch to 2/3 inch.

Size type

The aspect ratio of the photosensitive element of digital SLR cameras is mostly 3:2, and the size marking method is different. Generally, the photosensitive element size type is used to mark. Mainly divided into full frame Full Frame (close to or equal to 135 frames, such as Canon 1Ds series, 5D Mark II 36.0mm×24.0mm, Nikon D3, D700 36.0mm×23.9mm, Nikon D3x, Sony α900 35.9mm×24mm , Canon 5D 35.8mm×23.9mm, etc.), APS-H size (Canon 1D series 28.1mm×18.7mm, lens focal length conversion factor is 1.3), APS-C size (such as 23.6mm×15.8mm, 22.2mm× 14.8mm, 20.7mm×13.8mm, etc., the lens focal length conversion coefficients are 1.5, 1.6 and 1.7 respectively). Olympus and Panasonic digital SLR cameras use photosensitive elements with a size of 17.3mm×13.0mm, an aspect ratio of 4:3, and a lens focal length conversion factor of 2.0. From the structure of the camera, there are two systems, called the 4/3 system and the micro 4/3 system.

In 1996, the APS system (Advanced Photo System), jointly developed by Nikon, Canon, Minolta, Fuji, and Kodak, came out. The APS system has been greatly improved on the basis of the original 135 film system, including comprehensive innovations in cameras, photosensitive materials, printing equipment, supporting products, etc., greatly reducing the size of the film, using a new smart cassette design, incorporating digital technology, and becoming able to record Intelligent film system for shooting data and auxiliary information. The APS system is a major change to the traditional photography system, and it should have a good development prospect. Unfortunately, it did not happen at the right time. Due to the advent and rapid development of digital cameras, the APS system was quickly eliminated.

APS system has three kinds of film frames to choose from, namely: APS-H, APS-C and APS-P. APS-H is 30.2mm×16.7mm, which is the largest frame that can be taken by APS film; APS-C is 25.5mm×16.7mm, with a part of the left and right parts blocked, and the aspect ratio is close to 3:2 of the 135 frame; APS-P It is a part of the upper and lower blocks, which is 30.2mm×9.5mm, which belongs to the ultra-wide screen format.

The method of marking the size of the photosensitive element of the digital SLR camera borrows the APS standard, and the size of the photosensitive element is close to the APS-C size of 20.7mm×13.8mm (for Sigma), 22.2mm×14.8mm, 22.3mm× 14.9mm (for Canon in the above two sizes), 23.0mm×15.5mm (for Fuji), 23.4mm×15.6mm (Pentax K20D), 23.5mm×15.6mm (Sony α700), 23.5mm×15.7mm (Sony α350, Pentax K200D, K10D, Km), 23.6mm×15.8mm (for Nikon, called DX format, and Sony α300, α200), etc. are all called APS-C format, while Canon EOS-1D series uses 28.1mm×18.7mm Called APS-H frame.

For a photosensitive element with the same effective pixel, generally the larger the size, the larger the unit area of ​​each pixel, the better the photosensitive performance, and more image details can be recorded.

Size comparison

42

Influencing factors

Overview

Factors affecting the photosensitive element:

For digital cameras, the imaging There are two main factors: one is the area of ​​the photosensitive element; the other is the color depth of the photosensitive element.

Area

The larger the area of ​​the photosensitive element, the larger the image. Under the same conditions, more image details can be recorded, the interference between pixels is also small, and the image quality is better. But with the development of digital cameras in the direction of fashion and compactness, the area of ​​the photosensitive element can only become smaller and smaller.

Color depth

In addition to the area, the photosensitive element has an important indicator, that is, the color depth, which is the color bit, which is how many binary numbers are used to record the three primary colors . The photosensitive element of non-professional digital cameras is generally 24-bit, the sampling of high-end points is 30-bit, and the recording is still 24-bit. The imaging device of professional digital cameras is at least 36-bit, and it is said that it has 48 bits. CCD. For a 24-bit device, the brightness value that the photosensitive unit can record is up to 2^8=256 levels, and each primary color is represented by an 8-bit binary number. The maximum color that can be recorded is 256x256x256, about 16,77 Ten thousand kinds. For a 36-bit device, the brightness value that the photosensitive unit can record is up to 2^12=4096 levels, and each primary color is represented by a 12-bit binary number. The maximum color that can be recorded is 4096x4096x4096, about 68.7 billion. . For example, if the brightness of the brightest part of a certain subject is 400 times the brightness of the darkest part, if you use a digital camera that uses a 24-bit sensor to shoot, if the low-light part is exposed, the brightness will usually be higher than that of the darkest part. 256 times the parts are all overexposed, the gradation is lost, and bright spots are formed. If the high-light part is exposed, the parts below a certain brightness are all underexposed. If you use a professional digital camera that uses a 36-bit photosensitive element, it will not There is such a problem.