With the rapid development of computer image processing technology, the increasing popularity of information, and the improvement of people’s security awareness, real-time video surveillance systems have gradually become an integral part of people’s lives and have been widely used in production, transportation, and society. Security and other aspects. However, due to bad weather (fog, rain, wind, light, etc.) and the limitations of the monitoring system's own technical conditions, video images often do not achieve the desired results. The image quality is degraded or unclear, which in turn causes difficulty in operations such as identification, forensics, and event analysis, and prevents the system from being used properly. Therefore, the research and application of fuzzy image processing technology have very important significance in the field of security.
Causes of fuzzy image analysis system self-factors
In an all-analog monitoring system, from the front-end to the back-end, it consists of image acquisition, image transmission, image storage, and image display. Loss of video information may occur at every link, that is, the quality of the image is degraded or blurred.
Lens: affects the accuracy of the flux and imaging of the camera, will directly lead to image blur; camera sensor: affect the collection of optical signals and photoelectric conversion effect, will directly lead to image blur; BNC connector at both ends of the video transmission cable: The signal shielding gap will cause signal loss. After the video transmission cable is transmitted over a long distance, problems such as resistance, shielding, and impedance matching of the transmission cable will cause signal attenuation, which will directly lead to deterioration of the image quality. There will also be some signal loss on the image presentation side.
In an all-digital video surveillance system, using network transmission, the transmission and storage of digitally encoded video signals is more effective than analog systems, which can more effectively avoid image damage due to signal attenuation. However, image signal loss during lens, image capture, and back-end presentation still cannot be avoided. In addition, in the digital video surveillance system, the video signal A/D conversion and video coding and compression are added, and these links still cause the loss of image information. The existing video compression encoding algorithms are lossy compression, which will directly lead to the loss of video information and affect video clarity.
Natural environment
In addition to the system itself, the natural environment has a great influence on the sharpness of video images. If you experience natural weather such as wind, rain, snow, and fog, you will experience a sharp drop or blurring of image quality. In addition, there are insufficient illumination, backlight, backlight, low temperature or high, will have an impact on the image restoration system, affecting image clarity. In poor light conditions, the camera's Sensor imaging produces a lot of noise that can affect the image sharpness and can significantly increase the coded code stream.
Human environment
The power supply system power supply is not "clean", that is, intrusion into a relatively strong interference signal, specifically refers to superimposed on the 50Hz sine wave interference signal, if there is a large power SCR frequency modulation speed control device, SCR Devices, thyristor AC-DC converters, etc. will all cause pollution to the power supply.
There is a strong electromagnetic interference source or electromagnetic radiation near the TV monitoring system. Electromagnetic interference sources such as electric welding, radio transmission, large motors, and large relay interference can also cause interference with video signals. Electromagnetic interference can cause the image to have vertical bars with the same spacing or the image has regular flashing stripes, which causes the image to be blurred. In addition, it is caused by man-made destruction, such as license plates that have been made difficult to recognize, etc., resulting in the camera being unable to capture the license plate number.
Development of fuzzy image processing technology
Since the 1990s, the security industry has experienced a period of rapid development. With the development of security monitoring systems from full analog to full digital, people are increasingly demanding image quality, and the image resolution has also undergone a transition from CIF to standard definition to high definition.
Early traditional monitoring systems used analog signal transmission methods, known as the first generation of full analog monitoring era, when video files were recorded and saved by video tape recorders. Because the storage of video images is based on analog devices, the image storage device suffers from loss of information as the usage time of the analog storage device increases. However, blurry image processing technology is based on digital image processing technology, so it has not been introduced into the field of security in the purely analog era.
Due to the inherent defects of analog monitoring systems, with the continuous development of computer technology, digital storage methods have gradually replaced analog tape storage. Followed by entering the second generation of digital surveillance system, digital video recording DVR-based, replacing the analog video recorder, the first step out of digital surveillance. After the digitization of video images, image processing technology has begun to be used. Fuzzy image processing technology has gradually been introduced into the field of security.
With the development of computer and network technology, the current video surveillance has developed into an all-digital monitoring era based on IP networks, and thus entered the era of third-generation all-digital network video surveillance. Representative products of this era are mainly IPC and NVR. After digitizing video images, blurry image processing technology has been widely used in the field of security.
Fuzzy image processing technology is a kind of image processing technology. It uses a computer to perform operations or processing on digital signals in image information in order to improve the image quality and achieve the desired results. Therefore, it is also called computer image processing technology. For example, it removes noise from noise-contaminated images, enhances images with weak information, and performs geometric correction on distorted images. In recent years, with the increasing awareness of people's safety precautions and the increasingly strong social security needs, surveillance image processing technology will continue to evolve in such market demands and laws, and will play an increasingly obvious role.
Security industry special requirements
Computer image processing converts image signals into digital signals and processes them with a computer. Because computer processing speed is extremely fast, and digital signals have the characteristics of small distortion, easy preservation, easy transmission, and strong anti-interference ability, the application of computer image processing is extensive, including aerospace, telemetry, medical equipment, and industrial automation. , security identification and other major areas. Every application field has its own specific requirements, and its application in the security monitoring industry also has its inherent specialities:
1, the higher the image clarity. At the scene of security surveillance, agencies often need to identify suspects, evidence, etc. through surveillance videos. Videos with low resolution generally do not meet this requirement. At the traffic monitoring site, * the need to identify the license plate, violations, drivers and other requirements through the monitoring images, fuzzy images can not be applied at this occasion.
2. The monitoring of different industries has different requirements for images. For example, medical monitoring requires relatively high color reproduction of images. Intelligent traffic monitoring requires a relatively high requirement for the camera's nighttime illumination and capture speed, and it requires clear identification of the license plate. Unattended monitoring requires long-term stable operation of the equipment under unsupervised conditions.
3, outdoor installation, unguarded. In the field of security, most of the equipment needs to be installed outdoors, and the equipment needs to withstand the year-round wind and sun. The aging of electronic devices can be relatively faster than in other areas. The aging of cameras, lenses, transmission lines and other facilities will cause the image to become increasingly blurred.
4, the requirements of massive video channels. In large-scale safe city monitoring projects, the number of video channels will reach tens of thousands of roads or even more. Therefore, in the field of video surveillance, video coding is expected to achieve the highest bit rate compression ratio, thereby reducing the bandwidth and storage capacity requirements. This results in more information being lost in the video coding process, resulting in blurred images.
These special application requirements in the field of security will lead to a decrease in image sharpness, which in turn will have very high requirements for image clarity, which will inevitably lead to a broad application prospect of fuzzy image processing technology.
The limitations of fuzzy image processing techniques
Currently, due to factors such as hardware technology level, transmission bandwidth, and application environment, image blurring cannot be completely resolved.
From image capture, transmission, storage to display, any one of the links is critical to the quality of the image. Any step that goes out of the way will affect image quality, and this effect is irreversible. So, to completely solve the problem of image blurring requires a full range of technical updates. For example, in the current context of digital image technology, coding technology is one of the bottlenecks affecting image quality. If an encoding algorithm with a high compression ratio and low image loss occurs, it will certainly solve the problem of image blur caused by compression. However, to achieve this algorithmic effect, a higher computational cost is usually required, so an update of the hardware technology is also required to satisfy such an algorithm.
Similar to the super-resolution reconstruction fuzzy image processing technology, due to the complexity of the algorithm, the current conventional devices are still difficult to achieve real-time processing of high-definition images, so the efficiency of the algorithm is still one of the reasons for the current lack of blurred image processing. It is necessary to start with both algorithms and hardware to improve the efficiency of the algorithm and also to improve the hardware performance.
In addition, all kinds of algorithms for fuzzy image processing are based on the application of a specific scene. The locality and limitations of various algorithms have caused obstacles to the application of the algorithm. So for many years to come, there are still a long way to go for image processing algorithms and models.
Development Trends and Prospects
In a broad sense, an image that does not have enough information can be called a blurred image, so the image blurring problem will have a new appearance as people's needs increase. For example, the current monitoring field may only need an image with a certain resolution to meet the needs of face recognition. With the development of society, we may need to analyze someone’s mouth movement through a video to analyze him. Whatever you say, you will need higher resolution and clearer images. This will be an eternal pursuit, so the problem of fuzzy image processing will always be studied.
As part of the Internet of Things, video surveillance will eventually develop in the direction of intelligence as the Internet of Things continues to develop and apply. Auto-analysis of image blurring will become one of the techniques for system self-checking applications. Through intelligent analysis, the system will It automatically recognizes the image's geometric shape, color, noise, blur, fusion, and super-resolution image effects for automatic analysis and processing. With the wider application of fuzzy image processing technology, the perfect combination of fuzzy image processing technology and intelligent analysis will inevitably become a trend.
Causes of fuzzy image analysis system self-factors
In an all-analog monitoring system, from the front-end to the back-end, it consists of image acquisition, image transmission, image storage, and image display. Loss of video information may occur at every link, that is, the quality of the image is degraded or blurred.
Lens: affects the accuracy of the flux and imaging of the camera, will directly lead to image blur; camera sensor: affect the collection of optical signals and photoelectric conversion effect, will directly lead to image blur; BNC connector at both ends of the video transmission cable: The signal shielding gap will cause signal loss. After the video transmission cable is transmitted over a long distance, problems such as resistance, shielding, and impedance matching of the transmission cable will cause signal attenuation, which will directly lead to deterioration of the image quality. There will also be some signal loss on the image presentation side.
In an all-digital video surveillance system, using network transmission, the transmission and storage of digitally encoded video signals is more effective than analog systems, which can more effectively avoid image damage due to signal attenuation. However, image signal loss during lens, image capture, and back-end presentation still cannot be avoided. In addition, in the digital video surveillance system, the video signal A/D conversion and video coding and compression are added, and these links still cause the loss of image information. The existing video compression encoding algorithms are lossy compression, which will directly lead to the loss of video information and affect video clarity.
Natural environment
In addition to the system itself, the natural environment has a great influence on the sharpness of video images. If you experience natural weather such as wind, rain, snow, and fog, you will experience a sharp drop or blurring of image quality. In addition, there are insufficient illumination, backlight, backlight, low temperature or high, will have an impact on the image restoration system, affecting image clarity. In poor light conditions, the camera's Sensor imaging produces a lot of noise that can affect the image sharpness and can significantly increase the coded code stream.
Human environment
The power supply system power supply is not "clean", that is, intrusion into a relatively strong interference signal, specifically refers to superimposed on the 50Hz sine wave interference signal, if there is a large power SCR frequency modulation speed control device, SCR Devices, thyristor AC-DC converters, etc. will all cause pollution to the power supply.
There is a strong electromagnetic interference source or electromagnetic radiation near the TV monitoring system. Electromagnetic interference sources such as electric welding, radio transmission, large motors, and large relay interference can also cause interference with video signals. Electromagnetic interference can cause the image to have vertical bars with the same spacing or the image has regular flashing stripes, which causes the image to be blurred. In addition, it is caused by man-made destruction, such as license plates that have been made difficult to recognize, etc., resulting in the camera being unable to capture the license plate number.
Development of fuzzy image processing technology
Since the 1990s, the security industry has experienced a period of rapid development. With the development of security monitoring systems from full analog to full digital, people are increasingly demanding image quality, and the image resolution has also undergone a transition from CIF to standard definition to high definition.
Early traditional monitoring systems used analog signal transmission methods, known as the first generation of full analog monitoring era, when video files were recorded and saved by video tape recorders. Because the storage of video images is based on analog devices, the image storage device suffers from loss of information as the usage time of the analog storage device increases. However, blurry image processing technology is based on digital image processing technology, so it has not been introduced into the field of security in the purely analog era.
Due to the inherent defects of analog monitoring systems, with the continuous development of computer technology, digital storage methods have gradually replaced analog tape storage. Followed by entering the second generation of digital surveillance system, digital video recording DVR-based, replacing the analog video recorder, the first step out of digital surveillance. After the digitization of video images, image processing technology has begun to be used. Fuzzy image processing technology has gradually been introduced into the field of security.
With the development of computer and network technology, the current video surveillance has developed into an all-digital monitoring era based on IP networks, and thus entered the era of third-generation all-digital network video surveillance. Representative products of this era are mainly IPC and NVR. After digitizing video images, blurry image processing technology has been widely used in the field of security.
Fuzzy image processing technology is a kind of image processing technology. It uses a computer to perform operations or processing on digital signals in image information in order to improve the image quality and achieve the desired results. Therefore, it is also called computer image processing technology. For example, it removes noise from noise-contaminated images, enhances images with weak information, and performs geometric correction on distorted images. In recent years, with the increasing awareness of people's safety precautions and the increasingly strong social security needs, surveillance image processing technology will continue to evolve in such market demands and laws, and will play an increasingly obvious role.
Security industry special requirements
Computer image processing converts image signals into digital signals and processes them with a computer. Because computer processing speed is extremely fast, and digital signals have the characteristics of small distortion, easy preservation, easy transmission, and strong anti-interference ability, the application of computer image processing is extensive, including aerospace, telemetry, medical equipment, and industrial automation. , security identification and other major areas. Every application field has its own specific requirements, and its application in the security monitoring industry also has its inherent specialities:
1, the higher the image clarity. At the scene of security surveillance, agencies often need to identify suspects, evidence, etc. through surveillance videos. Videos with low resolution generally do not meet this requirement. At the traffic monitoring site, * the need to identify the license plate, violations, drivers and other requirements through the monitoring images, fuzzy images can not be applied at this occasion.
2. The monitoring of different industries has different requirements for images. For example, medical monitoring requires relatively high color reproduction of images. Intelligent traffic monitoring requires a relatively high requirement for the camera's nighttime illumination and capture speed, and it requires clear identification of the license plate. Unattended monitoring requires long-term stable operation of the equipment under unsupervised conditions.
3, outdoor installation, unguarded. In the field of security, most of the equipment needs to be installed outdoors, and the equipment needs to withstand the year-round wind and sun. The aging of electronic devices can be relatively faster than in other areas. The aging of cameras, lenses, transmission lines and other facilities will cause the image to become increasingly blurred.
4, the requirements of massive video channels. In large-scale safe city monitoring projects, the number of video channels will reach tens of thousands of roads or even more. Therefore, in the field of video surveillance, video coding is expected to achieve the highest bit rate compression ratio, thereby reducing the bandwidth and storage capacity requirements. This results in more information being lost in the video coding process, resulting in blurred images.
These special application requirements in the field of security will lead to a decrease in image sharpness, which in turn will have very high requirements for image clarity, which will inevitably lead to a broad application prospect of fuzzy image processing technology.
The limitations of fuzzy image processing techniques
Currently, due to factors such as hardware technology level, transmission bandwidth, and application environment, image blurring cannot be completely resolved.
From image capture, transmission, storage to display, any one of the links is critical to the quality of the image. Any step that goes out of the way will affect image quality, and this effect is irreversible. So, to completely solve the problem of image blurring requires a full range of technical updates. For example, in the current context of digital image technology, coding technology is one of the bottlenecks affecting image quality. If an encoding algorithm with a high compression ratio and low image loss occurs, it will certainly solve the problem of image blur caused by compression. However, to achieve this algorithmic effect, a higher computational cost is usually required, so an update of the hardware technology is also required to satisfy such an algorithm.
Similar to the super-resolution reconstruction fuzzy image processing technology, due to the complexity of the algorithm, the current conventional devices are still difficult to achieve real-time processing of high-definition images, so the efficiency of the algorithm is still one of the reasons for the current lack of blurred image processing. It is necessary to start with both algorithms and hardware to improve the efficiency of the algorithm and also to improve the hardware performance.
In addition, all kinds of algorithms for fuzzy image processing are based on the application of a specific scene. The locality and limitations of various algorithms have caused obstacles to the application of the algorithm. So for many years to come, there are still a long way to go for image processing algorithms and models.
Development Trends and Prospects
In a broad sense, an image that does not have enough information can be called a blurred image, so the image blurring problem will have a new appearance as people's needs increase. For example, the current monitoring field may only need an image with a certain resolution to meet the needs of face recognition. With the development of society, we may need to analyze someone’s mouth movement through a video to analyze him. Whatever you say, you will need higher resolution and clearer images. This will be an eternal pursuit, so the problem of fuzzy image processing will always be studied.
As part of the Internet of Things, video surveillance will eventually develop in the direction of intelligence as the Internet of Things continues to develop and apply. Auto-analysis of image blurring will become one of the techniques for system self-checking applications. Through intelligent analysis, the system will It automatically recognizes the image's geometric shape, color, noise, blur, fusion, and super-resolution image effects for automatic analysis and processing. With the wider application of fuzzy image processing technology, the perfect combination of fuzzy image processing technology and intelligent analysis will inevitably become a trend.
Pu Adhesive Sealant For Automobile
Polyurethane Sealant,Windsreen Sesalant,Adhesive Sealant
Neutral Silicone Sealant,Professional Silicone Adhesive Co., Ltd. , http://www.chsiliconesealant.com