With the increase of extreme weather in recent years, there has been a long period of heavy fog over the world. The fog has caused inconvenience to people and the safety factor of driving has also been greatly reduced. This article explains how to use part of the light to achieve fog monitoring.
Fog technology development prospects In ships, ships, airplanes, etc., the sighting system has a very important role in sensing the surrounding situation. The sighting system is generally composed of a CCD camera and an infrared imaging system. The severe marine meteorological environment such as fog, water vapor, rain and snow can seriously affect the image quality of CCD and infrared imaging systems, mainly reflected in the decrease of image contrast, distant targets blurred and difficult to distinguish. This affects the ability to sense the surrounding situation.
Through the image processing algorithm to enhance the contrast of the image, that is, the video anti-reflection technology, the sighting system in foreign countries, especially in the United States, has been widely used. The contrast effect before and after the image processing is as follows:
From this contrasting effect picture, we obviously feel that the contrast of the image has been greatly improved by the video-enhancement processing. The original blurred ships have become more visible, thus improving the concept. The sighting distance of the sighting system improves the system's ability to perceive the surrounding situation. Therefore, the video anti-reflection technology has a good application prospect in the sighting systems such as warships and aircrafts. The application of this video anti-reflection technology, due to the limitations of algorithms and hardware implementation technologies, has just begun in China, and commercialized mature products are rarely seen.
The marine environment is extremely harsh, and weather such as fog, rain, and moisture is common, and the sighting system needs to be able to observe long-range, tiny, high-speed moving targets in a timely manner. If you cannot find the target in time, you may make your partner passive. Therefore, it is necessary to equip this kind of video anti-reflection equipment to enhance the observation ability of the sighting system.
Fog penetration technology In recent years, the use of video surveillance equipment to protect security has become a necessary means for all walks of life. However, the traditional video surveillance equipment, without exception, has a drawback, that is, the monitoring effect at night and foggy days is very unsatisfactory, and night and foggy days are multiple cases of time. In addition, for a slightly longer distance monitoring, it is almost a blank space.
The fogging principle is such that in the range of invisible light, light with a frequency can penetrate the fog, but due to the different wavelengths, it needs to be processed on the camera to achieve the purpose of focusing, and it also needs to be in the camera. A new design is used to image this frequency of invisible light. Since this invisible light has no corresponding visible light color map, the image presented on the monitor is black and white. Photographing objects through clouds and water vapor is equivalent to passing through a double lens (water droplets and actual lens), except that the R light can be correctly focused on the CCD image plane, and the GB in the RGB light cannot be projected onto the CCD image normally. On the surface, this results in the normal mode lens being unable to obtain normal and clear images of clouds and water vapor.
In the past when the CCTV lens was still below 300mm, the observation distance was generally limited to within 1km. This application has a lower requirement for weather visibility, but the focal length has been developed to 750mm today. The influence of fog on the monitoring image has to be caused. We value it. This situation is particularly important in long-distance monitoring such as highways, forest fire prevention, oilfield monitoring, and port terminals close to the sea. This kind of environment is often more prone to fogging, leaving 24 hours of uninterrupted surveillance with new challenges.
In response to this situation, a small number of manufacturers with design and R&D capabilities have worked hard to develop lenses with a fog-transmitting function, and have succeeded in achieving the listing of finished products. The emergence of this technology has greatly broadened the scope of application of video surveillance and is another classic example of human beings relying on smart intelligence to overcome the natural environment. A few manufacturers in the market do not have the ability to produce fog lens products, use ordinary products to sell as fog lenses, and claim to have a fog-transmitting function. This is an extremely irresponsible behavior. Of course, in the actual test can not be sneaked through, and ultimately can not get rid of the destiny of being eliminated, but to users of this function have to create a lot of obstacles in the product selection and waste a lot of time.
Video fog penetration technology Video enhancement technology, generally refers to the image due to fog and water dust, etc., resulting in ambiguous images become clear, emphasizing some interesting features in the image, suppressing features that are not of interest, making the image Improved quality and richer information. The enhanced image provides good conditions for the next application of the image. In general there are two types of anti-reflection technology: airspace and frequency domain methods. However, these methods have some drawbacks in their adaptability to different images. In the 1970s, American physicist Land et al. proposed the Retinex image enhancement method, which is an image processing model based on human visual perception. It can compress the dynamic range of the image and show the details that are obscured in the image. However, the algorithm is complex and it is difficult to implement the project, especially for the real-time video real-time enhancement, because of the large amount of calculation, it is difficult to practical application. With the improvement of hardware performance, we can finally turn this universally adaptable image enhancement algorithm into an actual engineering product. This is the hardware product of the Retinex algorithm implemented for the first time in the industry.
The Retinex algorithm is a model based on the human visual system to perceive and adjust the color and brightness of an object. This model explains the phenomenon that the wavelength and brightness of the color are not particularly corresponding to each other in human eyes which cannot be explained by the general color theory. Land through a large number of experiments to prove that I mention the surface color will not change due to changes in lighting conditions that color constancy. In simple terms, the constant color means that the color of the same object that humans feel is the same whether it is in the midday sun, incandescent light, or in dim light conditions. Because of this, it is necessary to remove some indefinite and non-essential influences such as light intensity and uneven illumination when performing image operations, and only retain the reflective properties of the object such as reflectivity and other information. Image processing based on this method can make the image have a good effect on edge sharpening, dynamic range compression, and constant color.
The basic idea of ​​Retinex's theory is that the original image is considered to be composed of the illumination image and the object reflection properties. The illumination light image directly determines the dynamic range of the pixels in an image. The reflection property of the object determines the intrinsic properties of the image. Therefore, the It is the basic idea of ​​the Retinex theory to remove or reduce the influence of the illuminated image in the original image to preserve the nature of the reflective property. Compared with other image enhancement methods, the Retinex algorithm has features such as sharpening, color constancy, large dynamic range compression, and high color fidelity.
Now the multi-part video enhancer only uses the global Retinex image enhancement algorithm to calculate the ratio between adjacent pixel gray values ​​in the logarithmic domain to obtain the relative light-dark relationship between adjacent pixel points, and then through the light and dark The relationship corrects the gray value of the original pixel, and finally linearly stretches the gray value of the corrected pixel to obtain an enhanced image. So the contrast of the resulting enhanced image is not high. The CASEVision VE9901 video enhancer adopts advanced multi-scale Retinex image enhancement algorithm, which has strong universality. It also provides optimized logarithmic histogram equalization and multiple noise filtering algorithms. It is based on the embedded hardware structure of DSP and has the advantages of small size, low power consumption and high performance. Real-time image processing automatically adapts to PAL and NTSC video images. And has very low delay, the delay time does not exceed one frame, namely PAL system video delay 40ms, NTSC system video delay 33ms. At the same time, it also supports full-screen enhancements and local window enhancements. Locally enhanced window sizes and positions can be dynamically adjusted.
Fog technology development prospects In ships, ships, airplanes, etc., the sighting system has a very important role in sensing the surrounding situation. The sighting system is generally composed of a CCD camera and an infrared imaging system. The severe marine meteorological environment such as fog, water vapor, rain and snow can seriously affect the image quality of CCD and infrared imaging systems, mainly reflected in the decrease of image contrast, distant targets blurred and difficult to distinguish. This affects the ability to sense the surrounding situation.
Through the image processing algorithm to enhance the contrast of the image, that is, the video anti-reflection technology, the sighting system in foreign countries, especially in the United States, has been widely used. The contrast effect before and after the image processing is as follows:
From this contrasting effect picture, we obviously feel that the contrast of the image has been greatly improved by the video-enhancement processing. The original blurred ships have become more visible, thus improving the concept. The sighting distance of the sighting system improves the system's ability to perceive the surrounding situation. Therefore, the video anti-reflection technology has a good application prospect in the sighting systems such as warships and aircrafts. The application of this video anti-reflection technology, due to the limitations of algorithms and hardware implementation technologies, has just begun in China, and commercialized mature products are rarely seen.
The marine environment is extremely harsh, and weather such as fog, rain, and moisture is common, and the sighting system needs to be able to observe long-range, tiny, high-speed moving targets in a timely manner. If you cannot find the target in time, you may make your partner passive. Therefore, it is necessary to equip this kind of video anti-reflection equipment to enhance the observation ability of the sighting system.
Fog penetration technology In recent years, the use of video surveillance equipment to protect security has become a necessary means for all walks of life. However, the traditional video surveillance equipment, without exception, has a drawback, that is, the monitoring effect at night and foggy days is very unsatisfactory, and night and foggy days are multiple cases of time. In addition, for a slightly longer distance monitoring, it is almost a blank space.
The fogging principle is such that in the range of invisible light, light with a frequency can penetrate the fog, but due to the different wavelengths, it needs to be processed on the camera to achieve the purpose of focusing, and it also needs to be in the camera. A new design is used to image this frequency of invisible light. Since this invisible light has no corresponding visible light color map, the image presented on the monitor is black and white. Photographing objects through clouds and water vapor is equivalent to passing through a double lens (water droplets and actual lens), except that the R light can be correctly focused on the CCD image plane, and the GB in the RGB light cannot be projected onto the CCD image normally. On the surface, this results in the normal mode lens being unable to obtain normal and clear images of clouds and water vapor.
In the past when the CCTV lens was still below 300mm, the observation distance was generally limited to within 1km. This application has a lower requirement for weather visibility, but the focal length has been developed to 750mm today. The influence of fog on the monitoring image has to be caused. We value it. This situation is particularly important in long-distance monitoring such as highways, forest fire prevention, oilfield monitoring, and port terminals close to the sea. This kind of environment is often more prone to fogging, leaving 24 hours of uninterrupted surveillance with new challenges.
In response to this situation, a small number of manufacturers with design and R&D capabilities have worked hard to develop lenses with a fog-transmitting function, and have succeeded in achieving the listing of finished products. The emergence of this technology has greatly broadened the scope of application of video surveillance and is another classic example of human beings relying on smart intelligence to overcome the natural environment. A few manufacturers in the market do not have the ability to produce fog lens products, use ordinary products to sell as fog lenses, and claim to have a fog-transmitting function. This is an extremely irresponsible behavior. Of course, in the actual test can not be sneaked through, and ultimately can not get rid of the destiny of being eliminated, but to users of this function have to create a lot of obstacles in the product selection and waste a lot of time.
Video fog penetration technology Video enhancement technology, generally refers to the image due to fog and water dust, etc., resulting in ambiguous images become clear, emphasizing some interesting features in the image, suppressing features that are not of interest, making the image Improved quality and richer information. The enhanced image provides good conditions for the next application of the image. In general there are two types of anti-reflection technology: airspace and frequency domain methods. However, these methods have some drawbacks in their adaptability to different images. In the 1970s, American physicist Land et al. proposed the Retinex image enhancement method, which is an image processing model based on human visual perception. It can compress the dynamic range of the image and show the details that are obscured in the image. However, the algorithm is complex and it is difficult to implement the project, especially for the real-time video real-time enhancement, because of the large amount of calculation, it is difficult to practical application. With the improvement of hardware performance, we can finally turn this universally adaptable image enhancement algorithm into an actual engineering product. This is the hardware product of the Retinex algorithm implemented for the first time in the industry.
The Retinex algorithm is a model based on the human visual system to perceive and adjust the color and brightness of an object. This model explains the phenomenon that the wavelength and brightness of the color are not particularly corresponding to each other in human eyes which cannot be explained by the general color theory. Land through a large number of experiments to prove that I mention the surface color will not change due to changes in lighting conditions that color constancy. In simple terms, the constant color means that the color of the same object that humans feel is the same whether it is in the midday sun, incandescent light, or in dim light conditions. Because of this, it is necessary to remove some indefinite and non-essential influences such as light intensity and uneven illumination when performing image operations, and only retain the reflective properties of the object such as reflectivity and other information. Image processing based on this method can make the image have a good effect on edge sharpening, dynamic range compression, and constant color.
The basic idea of ​​Retinex's theory is that the original image is considered to be composed of the illumination image and the object reflection properties. The illumination light image directly determines the dynamic range of the pixels in an image. The reflection property of the object determines the intrinsic properties of the image. Therefore, the It is the basic idea of ​​the Retinex theory to remove or reduce the influence of the illuminated image in the original image to preserve the nature of the reflective property. Compared with other image enhancement methods, the Retinex algorithm has features such as sharpening, color constancy, large dynamic range compression, and high color fidelity.
Now the multi-part video enhancer only uses the global Retinex image enhancement algorithm to calculate the ratio between adjacent pixel gray values ​​in the logarithmic domain to obtain the relative light-dark relationship between adjacent pixel points, and then through the light and dark The relationship corrects the gray value of the original pixel, and finally linearly stretches the gray value of the corrected pixel to obtain an enhanced image. So the contrast of the resulting enhanced image is not high. The CASEVision VE9901 video enhancer adopts advanced multi-scale Retinex image enhancement algorithm, which has strong universality. It also provides optimized logarithmic histogram equalization and multiple noise filtering algorithms. It is based on the embedded hardware structure of DSP and has the advantages of small size, low power consumption and high performance. Real-time image processing automatically adapts to PAL and NTSC video images. And has very low delay, the delay time does not exceed one frame, namely PAL system video delay 40ms, NTSC system video delay 33ms. At the same time, it also supports full-screen enhancements and local window enhancements. Locally enhanced window sizes and positions can be dynamically adjusted.
Kitchen Sink Water Tap,Brass Kitchen Tap,White Kitchen Faucet,Flexible Kitchen Faucet
Wenzhou Yili Sanitary Ware Co., Ltd. , http://www.cn-faucet.com