CCD, English full name: charge coupled device, Chinese full name: charge coupled device. It can be called CCD image sensor or image controller. CCD is a semiconductor device, which can convert optical image into electrical signal. The tiny photosensitive material implanted in CCD is called pixel. The more pixels a CCD contains, the higher resolution it provides. The function of CCD is like film, but it converts light signal into charge signal. There are many photodiodes arranged in order on the CCD, which can sense light, convert the light signal into electrical signal, and convert it into digital image signal by external sampling amplification and analog-to-digital conversion circuit.
CCD image sensor can directly convert optical signal into analog current signal, and the current signal can be amplified and converted to realize image acquisition, storage, transmission, processing and reproduction. Its outstanding features are: 1. Small volume and light weight; 2. Low power consumption, low working voltage, shock and vibration resistance, stable performance and long service life; 3. High sensitivity, low noise and large dynamic range; 4. Fast response, self scanning function, small image distortion and no residual image; 5. Application of VLSI technology, high pixel integration, accurate size and commercial production The cost is low. Therefore, many instruments which use optical method to measure the outer diameter use CCD as photoelectric receiver.
CCD can be divided into linear array CCD and area array CCD. Linear array CCD usually divides the internal electrodes of CCD into arrays, each group is called a phase, and applies the same clock pulse. The required number of phases is determined by the internal structure of CCD chip. The CCD with different structure can meet the requirements of different occasions. Linear CCD can be divided into single channel and double channel, and its photosensitive region is MOS capacitor or photodiode structure, so the production process is relatively simple. It is composed of photosensitive area array and shift register scanning circuit, which is characterized by fast information processing speed, simple peripheral circuit and easy real-time control, but it can not process complex images due to small amount of information (linear CCD is shown in the right figure). The structure of area array CCD is much more complex. It consists of many photosensitive areas arranged into a square array and connected into a device in a certain form. It can obtain a large amount of information and process complex images.
Working principle of CCD
Characteristics of CCD
1. High resolution: the size of image points is μ m, which can sense and recognize fine objects and improve image quality. From the early 1 inch
From 1 / 2-inch, 2 / 3-inch, 1 / 4-inch to 1 / 9-inch, the number of pixels has increased from more than 100000 at the beginning to tens of millions of pixels, and there is a trend to continue to increase in the future.
2. Low noise and high sensitivity: CCD has very low readout noise and dark current noise, so it has relatively high signal-to-noise ratio (SNR). At the same time, it has high sensitivity of 0.0003-0.0005lux and even 0 lux low luminosity incident light can be detected, and its signal will not be covered by noise, so the application of CCD is not limited by climate;
3. High dynamic range: by
The dynamic range of the digital processed CCD signal can reach 400%, and the professional level can reach 600%. It can be applied to both strong light and weak light at the same time, so as to improve the application range of the system environment and avoid the signal contrast phenomenon caused by the large brightness difference;
4. Good linearity: the intensity of the incident light source is proportional to the output signal, which can reflect the detail level of the image and reduce the cost of signal compensation;
5. High quantum efficiency: very weak incident light can be recorded. With image intensifier and projector, even distant scenes in the dark can still be captured;
6. Large field of view: large area CCD chip can be manufactured by using semiconductor technology. At present, 35mm CCD, which is the same size as traditional film, has been applied in digital cameras and become the key component to replace professional optical cameras;
7. Broad spectral response: from 0.4 μ m to 1.1 μ m, it can detect light in a wide wavelength range, increase the flexibility of the system and expand the application field of the system; of course, according to different applications and requirements, it needs color filter or composite color filter;
8. Low image distortion: Using CCD sensor, the image processing will not be distorted, so that the surface information of the original object can be faithfully reflected;
9. Small size and light weight: CCD has the characteristics of small size and light weight, which is widely used
10. Low energy consumption, not affected by strong electromagnetic field;
11. Good charge transfer efficiency: the efficiency coefficient affects the signal-to-noise ratio and image resolution. If the charge transfer efficiency is not good, the image will become blurred;
12. Can be mass production, stable quality, strong, not easy to aging, easy to use and maintenance. Some burned CCD pixels can recover themselves after a period of live working.
Main CCD manufacturers:
The top seven manufacturers of CCD industry are all Japanese manufacturers, accounting for 98.5% of the global market share. In terms of technology development, the main manufacturers should be Sony, Philips, NEC and Kodak.
Working principle of CCD
1. CCD (charge coupled device) is a kind of special semiconductor. Now we are used to using CCD as the synonym of image sensor. The current CCD image sensor consists of three layers: the first layer: Micro transparent lens; the second layer: color separation lens; the third layer: photosensitive, storage, charge transfer (CCD) layer. As shown in the figure below:
2. The basic working principle of CCD: the photosensitive surface of CCD is a collection of several independent lithography units, which can store the signal charge generated by light or electric excitation. When the pulse with specific time sequence is applied to it, the stored signal charge can be transmitted in the CCD directionally and output the electric signal. The figure below shows the process of interline transfer CCD (it)
Make the schematic diagram:
3. There are two basic types of CCD: surface (channel) CCD (SCCD) and buried channel CCD (BCCD). Firstly, a semiconductor substrate is divided into a matrix of rows and columns by photolithography, as shown in the following figure (4 & times; 8 = 32 pixel units)
Each cell is a pixel cell. In each pixel cell, a light sensing region (light sensing diode), a charge storage region, a charge transfer region, a drain groove and an electrode are fabricated. As shown in the figure below:
It can be seen from the above diagram (front view) that each unit corresponds to a pixel, including a light sensing diode and the switch field effect transistor working with it, a transfer memory, and a flow channel and a transfer screen between the pixels. Therefore, the photosensitive surface accounts for about 1 / 2 of the area of each pixel. It is important to increase the photosensitive area, because the larger the photosensitive area is, the more sensitive the light is The more light the diode collects, the higher the image quality will be. However, the photosensitive area can not be made into an effective photosensitive part facing the light entrance. The area of the real photosensitive part is only 60% – 90% of the photosensitive area facing the light entrance. This is the so-called aperture ratio. But the area of each pixel is limited. At present, the solution is to add an optical lens in front of each sensitive area (the solution that Sony first came up with) to increase the light receiving area. This is the first layer of micro lens on CCD, so the sensitive area is determined by the micro lens, and the effect is very good.
4. Structure diagram of charge transfer mode of three typical image CCD
5. The following is an example of interline transfer CCD
The working principle of CCD is explained.
CCD structure is shown in the figure below, including photo diode, vertical register, row signal output register, control gate, charge detection, digital to analog converter amplifier and other units
Work flow chart of transfer CCD
The following is a face CCD image sensor. Now we will explain its working process in chronological order
1. During the positive scanning of the field signal, the photodiode in the photosensitive area converts the intensity of the light received into electrons and stores them in the photosensitive area for accumulation;
2. In the reverse scanning period (field blanking period) of field scanning, the transfer control gate changes from low potential to high potential, the photosensitive area and the vertical storage area (CCD) are turned on, and the charge stored in the photosensitive area is transferred to the vertical storage area. After the transfer, the transfer control gate potential becomes low potential, and the charge transfer is completed in the reverse scanning period (field blanking period);
3. During the positive scan of the field, the charge stored in the vertical storage cell is in the V ф 1
Under the action of V ф 2, V ф 3 and V ф 4 pulse sequence voltage, the stored charge is transferred down to the horizontal storage cell in turn. Each transfer process is completed only in the reverse scanning period of each row scan, and each vertical storage cell transfers only one pixel charge in the reverse scanning period of each row scan, so that only one row of pixels are stored in the horizontal output CCD after each transfer;
4. During the forward scan of line scan, the charge transferred to the horizontal storage cell is transferred to the left in turn under the action of H ф 1 and H ф 2 sequential pulse voltage, and the output (scanning signal of one line of image) is detected. After one line of output is completed, all the stored charges in the horizontal storage cell are transferred out, and the horizontal storage cell is emptied, waiting for the next transfer from the vertical storage cell
During the positive scanning period, the transfer output of an image is completed.
Working principle of color CCD
The pixel sensitivity of CCD only has a good proportional relationship to the intensity of light, and the sensitivity range exceeds the visible light range, but CCD can not recognize the color of light, it can only recognize the intensity of light, so CCD is color blind, it can only reflect the black-and-white image (gray level) of the image. How to recognize the color image?
1. Working principle of three primary color scheme
According to the principle of three primary colors, an image is separated into three primary color images red (R), green (g) and blue (b) (of course, it can also be separated into three complementary color monochrome images), which are composed of three CCD
The three primary color images are separately sensitized, and then the three primary color image signals are synthesized into full color image information signals for processing or output according to the needs. This is the imaging principle of three CCD, as shown in the figure below:
The advantages of this working mode are: high definition, good color restoration and convenient follow-up processing. The disadvantages are: high cost, three CCD work at the same time and debug well before leaving the factory; high maintenance cost, because of general maintenance, three CCD can not be optical overlap adjustment, even if one of the CCD is broken, three CCD should be replaced at the same time; high optical design requirements, in the optical path, three primary color spectroscope should be set to separate the panchromatic image into R, G, B three primary color images, respectively There should be three CCDs, and the images of the three CCDs must coincide in the physical vision. The manufacturer must make the three CCDs and the color separation mirror together, so the production cost is high.
2. Working principle of color single CCD
Different sensor manufacturers have different solutions. The most common method is to cover the RGB (red, green and blue) tricolor filter with a ratio of 1:2:1 (as shown in the figure)
That is to say, four image points constitute a color pixel (that is, red and blue filters cover one image point respectively, and the remaining two image points are covered with green filters). The reason for this ratio is that human eyes are more sensitive to green. Sony’s four-color CCD technology replaces one of the green filters with emerald green (some media call it e-channel), thus forming a new R, G, B, e four-color scheme. In the above scheme, the definition of color image is only 1 / 4 of that of monochrome CCD
After the CCD output, the RGB signal of each pixel in this group (four pixels) is recovered by calculation, so as to improve the clarity of the image.
Calculation method and recovery method, each company’s algorithm is not the same, the same image, through different recovery method (algorithm) to get the recovery image, there are also differences, for example, some images will be yellow, some will be red, and even a certain color will produce errors, such as red is restored to purple, in short, no matter how its algorithm, it is impossible to completely recover In addition, because the gamut of the three primary colors is smaller than the natural gamut, even if the most perfect algorithm is used, the restored image will produce color distortion, just as close to the original image as possible. In addition, why some algorithms (also known as color matrix) are better than others? This is because each developed algorithm has applied for copyright, and other companies must pay for it before they can use it. Of course, each developed algorithm will have its own characteristics, and it can also show the development ability of its own company