The meaning of wide dynamic range and low illumination camera parameters
Wide dynamic technology is a technology used to allow the camera to see the characteristics of the image under very strong contrast.
When high-brightness areas illuminated by strong light sources (sunlight, lamps, reflections, etc.) and relatively low-brightness areas such as shadows and backlights coexist in the image, the image output by the camera will appear bright. Areas become white due to overexposure, while dark areas become black due to underexposure, seriously affecting image quality. The camera has limitations in its performance of the brightest areas and darker areas in the same scene. This limitation is commonly referred to as "dynamic range".
In a broad sense, "dynamic range" refers to the span in which a certain changing thing may change, that is, the area between the lowest extreme of its changing value and the highest extreme. This area is generally described as the highest The difference between the point and the lowest point. This is a very widely used concept. When talking about the image shooting indicators of camera products, the general "dynamic range" refers to the camera's ability to adapt to the light reflection of the scene in the shooting scene, specifically the brightness (contrast) and color temperature (contrast). ) range of change.
The wide dynamic range of cameras is dozens of times greater than traditional cameras with only a 3:1 dynamic range. Natural light ranges from 120,000Lux to 0.00035Lux on a starry night. When the camera looks out of the window from indoors, the indoor illumination is 100Lux, while the illumination of the outside scenery may be 10,000Lux. The comparison is 10,000/100=100:1. This contrast makes it easy for the human eye to see because the human eye can handle a contrast ratio of 1000:1. However, there will be big problems in handling it with traditional closed-circuit surveillance cameras. Traditional cameras only have a contrast performance of 3:1. They can only choose to use an electronic shutter of 1/60 second to obtain the correct exposure of indoor targets, but outdoor images will be cleared (completely white); or alternatively, the camera selects 1/6000 second to obtain perfect exposure for outdoor images, but indoor images will be cleared (completely black). This is a flaw that has been around since the invention of the video camera.
Modern transportation requires modern traffic management. In order to solve traffic congestion and blockage on major urban road sections and intersections, and reduce accidents and violations, it is very necessary to establish a modern intelligent traffic command and control system. At the same time, it is also of great significance for improving the image of the city and promoting the civilization and development of the city. The overall goal of the system design is to use road monitoring to implement traffic flow and traffic operation monitoring, implement real-time traffic control on key road sections, promptly discover various abnormalities and take emergency measures, ensure high-speed, safe and effective road operation, and improve the quality of modern life. Traffic levels. According to the actual needs of current traffic monitoring, controllable cameras or fixed cameras are generally installed at key locations such as traffic intersections, stations, commercial areas, and highway toll gates. After analyzing the special needs of road monitoring, this article mainly focuses on road monitoring cameras. Some suggestions were made for the selection and design.
Points of concern when purchasing road surveillance cameras
In the video control system, whether from the acquisition of front-end images of the system or the recording, display and control of back-end image signals, The performance of system equipment is a key factor in determining the success of system operation. There is no doubt that the quality of equipment selection directly affects the stability and reliability of the system, image quality, system service life and other issues related to the investment interests of the builder. Therefore, system equipment selection is an important link throughout the entire design process.
Road surveillance system camera demand analysis.
There are high requirements for the clarity and real-time performance of the image, and the license plate is required to be seen clearly. If the license plate number cannot be clearly confirmed, , then monitoring and capturing pictures will be meaningless. Since road monitoring requires 24-hour operation, high-quality images must be obtained even in extremely dark conditions. The dynamic range of light on outdoor roads changes greatly. The ambient illumination reaches 50,000Lx-100,000Lx under summer sunlight; it is only 0.1Lx under nighttime street lights, which shows a considerable change.
In this case, regardless of whether the camera has an automatic sensitivity adjustment function, it is impossible to adapt to such a wide range of illumination through the electronic shutter of the camera itself, and it is impossible to control the image effect. Therefore, the camera must have a wide dynamic range. When shooting under poor illumination conditions, the dynamic images captured will inevitably be interfered by noise, so wide dynamic cameras are required to have outstanding dynamic image noise elimination functions that can eliminate image shadows and tailing phenomena.
Basis for selecting road surveillance cameras
The core IC circuit in the performance of wide dynamic cameras is the CCD sensor chip. The working principle is that the CCD optical lens images the target scene on the CCD sensor. The sensor is a high-sensitivity CCD, and then the signal output by the CCD is passed through the CDS related sample and hold circuit, AGC and A/D at the speed of 50 fields and 25 frames per second (CCIR system 25 Frame/s; NTSC system 30 Frame/s). After processing by the conversion circuit, it is input into the memory, and then the high-speed computing chip and data processing function are used to read out the images stored in the memory line by line in a progressive scanning manner to form a full video signal. Therefore, in addition to selecting a CCD sensor with excellent performance, the data processing chip/processing circuit is also an important part of the quality of the camera output signal.
After summing up many years of practical experience, road traffic monitoring equipment integrators and engineering companies have selected cameras with the following requirements:
Cameras with high line count (500-540 TV lines) Industry standard camera. Low illumination (≤0.1lux), the minimum illumination reaches 0.0lux, and clear image effects can be obtained even in the dark night with low illumination. Using super-sensitive, large-size CCD (usually 1/2-inch CCD). Since the target size of the 1/2" camera is larger than that of the 1/3" camera, the imaging effect is better. (The imaging area is large; the luminous flux is large, and the illumination requirements are low.) It has ultra-wide dynamic shooting function, which can respond quickly and accurately in the case of high contrast and sudden changes in lighting, thereby obtaining high-quality, fully exposed images. picture. With super noise reduction technology, it can eliminate dynamic image noise, image shadows and smearing. Low smearing is particularly important when solving road traffic monitoring or parking lot monitoring problems caused by headlights. Fast (shutter speed cannot be slower than 1/1000 seconds) camera with high signal-to-noise ratio, automatic white balance adjustment and other functions. Using industrial-grade components, it has good all-weather working capabilities and long-term operation is stable and reliable. This article intends to explain the following key parameters.
1/2 EXVIEW HAD ?CCD
CCD products have been around for more than 30 years, developing from 200,000 pixels at that time to 5-8 million pixels currently. Regardless of its market size or Its application areas have achieved tremendous development, and it can be said to be gradually improving steadily. Especially in recent years, applications in the consumer field have developed faster.
The area of ??each pixel of the current CCD component has been reduced to less than 1/10 compared to the early stage of development. In the future, as application products tend to be miniaturized and high pixels are required, the unit area will be even smaller. While miniaturizing, various newly developed technologies are used to ensure that the sensitivity will not be affected by the reduction in unit area, and at the same time, its performance is required to be maintained or improved.
The following is a brief introduction to the CCD sensors developed by Sony Corporation by age:
1. Had sensor
The had (hole-accumulation diode) sensor is This is Sony's unique structure with a positive hole accumulation layer on the surface of the n-type substrate, p-type, and n 2-pole body. Due to the design of this positive hole accumulation layer, the dark current problem that often occurs on the sensor surface can be solved. In addition, designing a vertical tunnel that electrons can pass through on the n-type substrate increases the aperture ratio, in other words, the sensitivity is also improved.
In the early 1980s, Sony took the lead in using it in variable-speed electronic shutter products to obtain clear images when shooting fast-moving objects.
2. On-chip micro lens
In the late 1980s, due to the shrinkage of each pixel in the CCD, the light-receiving area will be reduced and the sensitivity will also be lowered. To improve this problem, Sony installs a tiny lens in front of each photodiode. After using the tiny lens, the photosensitive area is no longer determined by the opening area of ??the sensor, but by the surface area of ??the tiny lens. Therefore, the aperture ratio is increased in terms of specifications, and the brightness is greatly improved.
3. super had ccd
Since the late 1990s, the unit area of ??ccd has become smaller and smaller. The tiny lens technology developed in 1989 can no longer improve the brightness. If the amplification factor of the internal amplifier of the CCD component is increased, the noise will also be increased, and the image quality will be significantly affected. Sony has taken a step further in the research and development of CCD technology, improving the previous technology of using tiny lenses, improving light utilization, and developing technology to optimize the shape of the lenses, namely Sony super had CCD technology. Basically, it is designed to improve the brightness by improving light utilization efficiency, which also lays the foundation for the current basic CCD technology.
4. new structure ccd
As the aperture F value of the camera's optical lens continues to increase, more and more oblique light enters the camera, making it incident on the ccd component The light cannot be focused 100% on the sensor, and the sensitivity of the CCD sensor will be reduced. In 1998, Sony added an internal lens between the color filter and the light-shielding film to improve this problem. Adding this layer of lens can improve the internal light path, so that oblique light can be focused onto the photoreceptor. At the same time, the insulating layer between the silicon substrate and the electrode is thinned to prevent signals that would cause vertical CCD screen noise from entering, thereby improving the smear characteristics.
5. examine had ccd
Infrared light with a longer wavelength than visible light can also be used for photoelectric conversion in semiconductor silicon chips. However, so far, CCD cannot effectively collect these photoelectrically converted charges into the sensor. For this reason, Sony's newly developed "exview had ccd" technology in 1998 can effectively convert near-infrared light that has not been effectively utilized before into image data for use. The visible light range is expanded to infrared, which greatly improves the brightness. When using the "exview had ccd" component, high-brightness photos can be obtained in dark environments. Moreover, during the previous photoelectric conversion performed in the deep layer of the silicon crystal plate, the smear component that leaked into the vertical CCD part can also be collected into the sensor, so the noise that affects the image quality will also be greatly reduced.
Minimum illumination
Illuminance is a unit that reflects the intensity of light. Its physical meaning is the luminous flux illuminated on a unit area. The unit of illumination is lumens per square meter (Lm). Number, also called Lux: 1Lux=1Lm/square meter. In the above formula, Lm is the unit of luminous flux. Its definition is the surface area of ??1/60 square meters of pure platinum at the melting temperature (about 1770°C). The amount of light radiated within a solid angle of 1 steradian.
In order to have a perceptual understanding of the amount of illumination, let’s take an example for calculation. A 100W incandescent lamp emits a total luminous flux of about 1200Lm. If it is assumed that the luminous flux is evenly distributed on a hemispheric surface. above, then the illuminance values ??at 1m and 5m away from the light source can be obtained according to the following steps: The area of ??a hemisphere with a radius of 1m is 2π×12=6.28 square meters. The illuminance value at 1m away from the light source is: 1200Lm/6.28 square meters= 191Lux In the same way, the area of ??a hemisphere with a radius of 5m is: 2π×52=157 square meters. The illumination value at a distance of 5m from the light source is: 1200Lm/157 square meters=7.64Lux
It can be seen that the light emitted from the point light source Illumination follows the inverse square law. 1LUX is approximately equal to the illumination of 1 candle at a distance of 1 meter. It is the lowest illumination commonly seen in camera parameter specifications. It means that the camera can obtain clear images only under the marked LUX value. The smaller the value, the better. , indicating that the sensitivity of the CCD is higher. Under the same conditions, the illumination required by a black-and-white camera is far 10 times lower than a color camera that still has to deal with color density. The sensitivity of black and white cameras is about 0.02-0.5lux (lux), and the sensitivity of color cameras is mostly above 1lux. The illumination value is not only related to the aperture size (F value) of the lens, but also has a greater relationship with the surrounding environment during testing. In terms of aperture size (F value), the larger the aperture, the smaller the F value it represents. The lower the illumination required. 0.97lux/F0.75 is equivalent to 2.5lux/F1.2 is equivalent to 3.4lux/F1.0
Reference environment and illumination:
Refer to the approximate illumination of the environment
100000lux under summer sunshine
100lux indoor fluorescent lamp
10000lux outdoor on cloudy day
10lux indoors at dusk
TV studio? 1000lux
Candlelight 10-15lux at 20cm
60w table lamp 60cm away from desktop? 300lux
Street light at night? 0.1lux
The camera can be adjusted according to the degree Divided into
Ordinary type: the illumination required for normal work is 1~3lux
Moonlight type: the illumination required for normal work is about 0.1lux
Starlight type: normal work The required illumination is less than 0.01lux
Infrared type: Using infrared lighting, imaging can be performed even in the absence of light
The minimum illumination of the wide dynamic camera refers to the scene when the subject is being photographed The brightness of the light is low enough to cause the amplitude of the video signal output by the camera to drop to 50--33 of the standard amplitude of 700mV (the nominal video value is 1V, the standard value is 700mV): Another minimum illumination is on the CCD Illumination is the sensitivity of the CCD. The illumination value of the CCD is far lower than the minimum illumination value of the camera. Many unscrupulous businessmen label the minimum illumination value of the CCD as the minimum illumination value of the camera to deceive the uninformed. This is especially reflected in some domestic OEM products and some No-name, low-end camera products.
The evolution of low-light cameras in the market is simply divided into the following three steps: daytime color/nighttime black and white (COLOR/MONO); low-speed shutter (SLOW/SHUTTER) and super-sensitivity camera (EXVIEW HAD).
1. Color during the day / black and white at night (day and night camera COLOR/MONO)
This type of camera still has its own specific demand group in the market, day and night camera (COLOR/MONO) ) The camera takes advantage of the high sensitivity of black and white images to infrared rays. Under certain light source conditions, it uses line switching to convert the image from color to black and white to facilitate matching with infrared rays.
In the technological evolution of color/black-and-white line conversion, in the early days, 2 SENSORs (1 color, 1 black-and-white) were used to switch with a set of circuits. Currently, such cameras have adopted a single CCD (color) design. , it is a color camera during the day or when the light source is sufficient. When night falls or the light source is insufficient (usually between 1LUX and 3LUX), digital circuits are used to eliminate the color signal and become a black and white image. In order to match the infrared ray, the color camera is also removed. Indispensable infrared filter, although this method can achieve the purpose of "low illumination" at night, it has the disadvantages of blurred images and unnatural colors during the day, and the camera's shooting distance will be limited by the irradiation distance of the infrared light. However, whether the COLOR/MONO camera is a "low-light" camera is still quite controversial. Experts point out that the real "low-light camera" should refer to the functions that the camera itself (components and technology used) can achieve, while daytime color/ The black and white camera at night cannot be changed due to its limited CCD sensitivity. It only uses line switching and infrared light to improve its functions, so it cannot be regarded as a low-light camera.
2. Low-speed shutter (SLOW/SHUTTER)
This type of camera is also called a (picture) accumulation camera. It uses computer memory technology to continuously combine several images due to light. The insufficient and blurry pictures are accumulated and become a clear picture. The SLOW SHUTTER technology is used to reduce the camera illumination to 0.008LUX/F1.2 (×128), and the number of frames (128 frames) that the picture can accumulate is even Including imported brands and then the leading level. This type of low-illumination camera is suitable for monitoring in museums where red and ultraviolet damage is prohibited, observing biological activities at night, monitoring military coastlines at night, etc., and in relatively static places. Most of this type of low-light cameras are imported brands, which are expensive and have a low number of accumulated frames (32 frames).
3. Super-sensitivity camera (EXVIEW/HAD)
Super-sensitivity camera (EXVIEW/HAD), also known as 24-hour camera, was the most popular model in the world in 1998. Its color illumination can reach 0.05LUX, and black and white can reach 0.003-0.001LUX (it can also be combined with infrared to reach 0LUX). It can not only clearly identify images, but also provide real-time continuous images. This type of camera mainly uses the EXVIEW/HAD/CCD (super-sensing CCD) launched by SONY component factory in 1997. It uses patented technology to increase the aperture rate of each pixel of the CCD to achieve lower illumination requirements. This The emergence of the technology has been welcomed by the monitoring market, and it can show the best results in various lighting environments. Especially with special infrared lighting equipment, high-definition black and white images can be obtained, achieving zero illumination monitoring (complete without light). In the near-infrared region of 760mm-1100mm, if combined with infrared lighting of appropriate wavelengths, clear black and white images can be achieved.
Samsung TECHWIN (formerly Samsung Aviation, domestically known as Samsung Optoelectronics) relies on 30 years of production experience in the industry, and has always been at the forefront of continuous innovation in technology and providing customers with high-quality security products. front end. Its product series SHC-740, SHC-740, SHC-721, SDZ-330, SPD-3300, etc. all adopt 128 times frame accumulation technology, with a resolution of more than 520TVL, a signal-to-noise ratio of more than 50db, and day and night conversion functions.
In particular, SHC-740 (Figure 1) adopts EX-VIEW HAD CCD and Samsung SVⅢ DSP chip. It has a new breakthrough in low-light technology and high definition (up to 540TVL), allowing the camera to operate in almost completely dark conditions. You can also get high-quality pictures under low illumination. The lowest illumination color mode is 0.01LUX@F1.2, the black and white mode is 0.003lux@F1.2? The sens-up mode is 0.0003 LUX@F1.2, which is widely used in national defense, border, and military , highway
Wide dynamic range
In some situations where the contrast between light and dark is too large, due to the limitations of the photosensitive characteristics of the CCD, ordinary cameras often capture images with too bright background or too bright foreground. Dark situation. In response to this situation, wide dynamic technology emerged as the times require, which better solves this problem. Before this, traditional cameras generally used backlight compensation functions to adapt to situations with large light contrast.
When the objects in the field of view of a conventional camera need to see objects at the door or outside the window under a brighter background light, the central backlight compensation (BLC) mode is usually used, which mainly relies on improving the central part of the field of view. The brightness of the surrounding parts of the field of view is reduced to achieve the purpose of seeing objects in the center clearly.
Backlight compensation, also called backlight compensation, divides the picture into several different areas and exposes each area separately. In some applications, the field of view may contain a very bright area, and the included subject is surrounded by the bright field, making the picture dark and levelless. At this time, since the signal level detected by AGC is not low, the gain of the amplifier is very low and cannot improve the brightness of the main body of the picture. When backlight compensation is introduced, the camera only detects a sub-area of ??the entire field of view. By calculating The average signal level in this area determines the operating point of the AGC circuit. Since the average level of the sub-area is very low, the AGC amplifier will have a higher gain, which will increase the amplitude of the output signal, thereby making the subject picture on the monitor clearer and greatly reducing the subjective brightness difference between the background picture and the subject picture. The visibility of the field of view is improved. Although backlight compensation improves the brightness of the subject, the image quality will be more or less degraded.
The wide dynamic range technology is to expose twice at the same time, one fast and one slow, and then synthesize it so that bright and dark objects on the screen can be seen clearly at the same time. Although both are measures taken to overcome the problem of seeing the target clearly under strong backlight conditions, backlight compensation comes at the expense of the contrast of the picture, so in a sense, wide dynamic technology is backlight compensation. upgrade.
Due to its corporate background, Samsung TECHWIN (Samsung Optoelectronics) has rich experience in the national military industry. Its products focus more on industrial supplies, durability and stability. And has many years of experience in optical and semiconductor technology. It has good performance in products such as security surveillance cameras and embedded hard disk video recorders with built-in professional semiconductor chips. Its independently developed third-generation super image technology SVIII (Figure 5) is equipped with a dual-speed scanning CCD, which can capture images with wide dynamic effects. And using two 12-bit digital input terminals, the SVⅢ has a wide dynamic range of more than 80db. After a large amount of digital information is transmitted to the DSP through the 23-bit data bus, the DSP performs internal processing to ensure that there is no data loss. The wide dynamic range is then compressed into a 10-bit output through a non-linear adaptive WDR compression algorithm.
It adopts adaptive anti-mosaic color reproduction algorithm, reaching 540 lines of color and 570 lines of black and white.
Powerful sensitivity enhancement technology provides dynamic 3D filtered images and optimizes the signal-to-noise ratio. And enhance visibility under low-light conditions while maintaining full real-time effects.
Adopting advanced local area enhancement contrast technology, ideal contrast can be obtained even under poor illumination conditions.
The use of a unique color control algorithm can expand the range of white balance use, that is, the color can be accurately and truly reproduced within a wide color temperature range.
This also means that the color mode can be used in very low light conditions and the white balance can be well achieved.
Typical applications
At present, the focus of low-illumination, wide-dynamic cameras for road monitoring is highway toll monitoring systems, mainly for toll station lanes, toll plazas, and toll booths. Regarding the toll collection situation, observe and record the types of vehicles passing through the toll lane, the operation process of the toll collectors, and emergencies and special events during the toll collection process, and implement effective supervision. . Especially at night, toll station staff need to see the license plate clearly. Under normal circumstances, after the car lights are turned on, the ambient illumination on the road and the illumination of the license plate form a certain dynamic range. It is difficult for traditional cameras to "see clearly", so for low-frequency There is a demand for illumination and wide dynamic cameras.
The second is the electronic police system, which uses closed-circuit television monitoring and automatic recording of red lights to improve the command center’s intuitiveness, real-time mobilization capabilities and response capabilities to traffic accidents and accidents, as well as enhance Investigate and handle violations objectively, and conduct comprehensive coordinated control of the control area to improve vehicle traffic capacity. Since license plates need to be seen clearly and monitored 24 hours a day, there is a demand for low-illumination and wide-dynamic cameras.
In addition, there are also certain applications in urban commercial streets, which are used to understand the traffic conditions at some busy intersections, the operation conditions of vehicles around the road sections, the flow of pedestrians, and traffic security conditions.
Conclusion
As society becomes increasingly aware of traffic safety precautions, road infrastructure construction must cooperate with the construction of road monitoring systems, which undoubtedly indicates that there are huge business opportunities hidden in the traffic monitoring industry. . A camera with low illumination and wide dynamic range must have a wide space. Technology, Winner in Digital World, Samsung TECHWIN is willing to contribute its own strength to China's intelligent transportation industry with high-quality products and first-class services.
Some technical knowledge about wide dynamic range
Wide dynamic range is the ratio of the brightest brightness signal value that the image can resolve to the darkest bright light signal value that can be resolved.
The expression of wide dynamic range is expressed in "multiples" or "dB". When 100IRE is used as the standard, the conversion formula is: N dB=20log (V2/V1). The wide dynamic value of an ordinary camera (called V1) is 10dB. For example, the wide dynamic range is 48 dB, and the difference with an ordinary camera is 38 dB. V2/V1=80, which means that the wide dynamic difference with an ordinary camera is 80 times. Panasonic's third The third generation wide dynamic camera is 54 dB, V2/V1=160 times. The typical dynamic range of Ikegami ISD-A10 camera is 95 dB, V2/V1=17782 times, and the maximum wide dynamic range is 120dB, V2/V1=316227 times. From the perspective of "multiple", the wide dynamic range of cameras using Pixim DPS technology is greatly improved compared to the wide dynamic range of CCD.