In image detection of the actual work, we download/extraction of video image quality is not high, often appear unable to clearly identify the image of the people, license, or cannot see video recording process of events. In fact, some of the reasons for poor image quality, as figure mapping workers, no matter how video systems engineering and equipment quality, the image acquisition, conversion, and processing of image enhancement process, can't again artificially caused a drop in the quality of video image and the image quality loss.
After a period of exploration and research, the writer summarizes in figure investigation work need to handle and solve the following questions:
Frist,A process, enhancing the image to identify the cause of the video image fuzzy production before
There are many factors that can affect the image quality monitoring, this paper analyzed and discussed from the Angle of the graph mapping application.
1, have different quality standard: due to the unbalanced economic development, various units, departments, the performance of the installed video monitoring system, technical standards, the relevant provisions of the state is not very perfect, lead to the unit, department, the application of video surveillance system also exist large quality difference.
2, different video purposes: the construction unit and the person in charge of monitoring video do not match the requirement and purpose, caused the picture quality is uneven due to different emphasis. Therefore, in the use of surveillance video images as evidence or clues also appears a series of problems.
3, video image compression/transport/storage/unzip the quality is different: while current monitoring technology and the corresponding equipment has been widespread used in our country, however, due to the manufacturers in order to compress the storage space, reduce cost and improve the compression ratio. Simulation of the video image signal input to the converted into digital signal after approved by compression, transport, storage, and then the extract several links, image quality has been more or less affected, image clarity, grayscale, color reduction, real-time performance attenuation in different degrees.
4, the video acquisition is affected by the external environment and equipment inner noise: there are many factors which affect the image clearly degree, such as outdoor light enough even can cause the video image grey too concentrated; Images obtained by CCD (camera) after A/D (number/mode conversion, this feature in the image system implemented by digital acquisition card) conversion, transfer line will produce noise pollution, and so on. So inevitably reduces the image quality, light person performance for image not clear, hard to see the details; The person that weigh is blurred, even general picture look not to come out.
5, man-made video image quality of the secondary losses: many vendors for commercial purpose and man-made their need to use the exclusive video image browser can play, and the figure of all image processing software is used in the investigation work conditions are not fully compatible with the market all the video format, it can make video download/extraction on the post-processing need transformation format or download mode, to record the screen directly and transformation format or record the screen to download itself is a process of loss.
Second, choose the right method and video download transformation format
The root cause of the image format is affect image quality. Because under the influence of network bandwidth and storage costs, all equipment manufacturers of video surveillance equipment will take video compression coding technology, while it is possible to ensure that the application requirement of the majority of users, but for the case investigation, the majority of the video image quality far cannot meet the requirements. Figure investigation work, therefore, the first level is to maximize the avoid quality monitoring image resources from secondary damage, is also in copy, download or take to minimize the losses. In practice, a figure investigation personnel should grasp the following several aspects of knowledge and technology to reduce the quality losses on this link:
1, familiar with the types of image resources
Video image resources roughly divided into three categories: image is a person in the city network system resources, the second is the unit of video system based on DVR stored image resources, three is all kinds of relatively backward monitoring system based on computer + the card image of hard disk storage resources.
2, to master the most practical method for download
The first category belong to the metropolitan area network system, can easily take the network client obtained according to the IP address to download the video data; The second type of system has been tested, use IE browser downloads faster and reliable; The third class with no RJ45 interface, can only adopt the method of hard disk to kao or record the screen.
3, conversion, video image format is an important part of the image enhancement
After groping in our practical work experience, in the table 1 equipment storage formats are all kinds of basic software compatibility. If you met with various software system is not compatible with the video format, generally use fomatfactory, raccoon dog kennel or universal format converter basic can also be transformed. Whether what kind of software to convert, must be based on h. 264 AVI video file format enclosed. As for the software are not compatible format, or ask vendors for conversion tool, use either had to record screen, using professional collection as far as possible when recording equipment, even so, recorded video quality is not good.
Third, understand several key parameters of the image enhancement technology and their internal relations
Figure investigation work, after the video image capture image processing is a very important work, strengthen links in image processing is staff must master the basics of budgeting. Image processing, image processing index words commonly. Digital image refers to use digital cameras and other equipment after sampling and digital a large two dimensional array, the array elements, called pixels, its value is an integer, is called the grey value. The main contents of image processing, including image compression, enhancement and restoration and matching, describe and identify three parts.
1, the content of the image enhancement
Image enhancement is, among other things, to adjust the brightness, chroma, saturation and using histogram or curve adjustment levels as well as the image sharpening, noise reduction, nondestructive amplification processing, etc.
2, image increases fresh to processing parameters
In spite of the high flair of the human eye can identify thousands of colors, but in many cases, the image is blurred even invisible to the human eye, through the image enhancement technology, can make the fuzzy even invisible image is clear and bright. To improve image quality and image enhancement. Image enhancement without considering the decrease of the quality of the image of the reason, will only in the image features of interest are selectively enhanced, while weakening characteristics, you don't need the purpose of this work is to improve the intelligibility of the image and improve the viewer's useful information. In practice, we mainly use a variety of image processing software for image parameter adjustment, maximum limit for his useful information. Mainly includes: image enhancement to deal with the parameter levels, curves, color, brightness, contrast, saturation, etc. These parameters we can use on the market more saw the light and shadow magic hand, such as photoshop software can easily deal with.
3, flexible use of software image color parameters, avoid the loss of pixels
Although there are a lot of software can adjust the parameters of the image, but after processing the results vary from person to person. Investigate its reason, didn't find the lack of image parameters. Once right after find out why it easy to use software tools to enhance processing, restore the true colors of the image. After a period of groping, coloring order to adjust the brightness, contrast and saturation of the image loss is small, very intuitive to decrease the loss of pixels and to retain the original color details, and directly to adjust the brightness and contrast of image quality is relatively serious, pixels are lost. Adjust the levels of methods we summed up as "image enhancement levels execution of work".
