Talking about the front-end problem of high-definition cameras, a & s cited several existing obstacles and problems in many articles on IP and HD-SDI, from the introduction to the current general application. The top ten problems of front-end high-definition cameras have caused a lot of confusion in the selection and use of camera applications by engineers. Because these ten problems are almost related to the components of high-definition cameras, but for more professional With the knowledge of components, the engineering company really does not know how to see the clues from the specifications and conditions of the HD camera, so as to solve the corresponding problems.
Ten common problems of IP and HD-SDI front end:
Differences in sensor frame number conditions cause differences in high-definition effects;
The DSP / ISP performance difference and the output part are not high-definition BT.1120 / LVDS pure high-definition signals, but fake high-definition in the form of BT.656 ParallelScaleup;
The IP video signal flow packet is too large, and the delay is too large when the bandwidth is insufficient. The high-definition picture is not smooth;
Insufficient IPCODECCPU performance can not meet the high-definition frame rendering;
Front-end intelligent embedding, excessive energy consumption, easy to overheat and crash;
The number of IP video streams is insufficient to cope with the back-end demand and cause the screen to drop frames;
The lens must be matched with a suitable one-megapixel aspheric lens to have high-definition effect;
SDI image 720P / 1080i / 1080P high-definition formatless standard conformance;
The standardization of the SDK part of the private agreement is not uniform, it is not easy to integrate, and integration is difficult;
Most of the projects without ABF (Auto Focus) installation and adjustment are troublesome.
In the evaluation of the HD IP Megapixel and HD-SDI at the 2012 Secutech Taipei International Security Expo, we locked the HD selection project on how to solve these ten issues, and conducted initial testing and re-evaluation. Fortunately, from the 20 sets of IP Megapixel and 11 sets of HD-SDI cameras participating in the evaluation, of a total of 31 high-definition cameras, we see that different brands of high-definition cameras use different element components to solve the above ten problems. The resulting progress and differentiation.
The mainstream trend of CMOS development is clear
As one of the core components of high-definition surveillance cameras, the difference between CCD and CMOS should be clear. But to distinguish by the existing performance, the difference between the two in the evaluation is embodied in the aspects of sensitivity, noise processing and power consumption. In the past, under the same pixel conditions, CMOS was significantly worse than the sensitivity of CCD; and under the same sensor size, CCD will be far superior to CMOS in light input, which is usually the biggest difference between CCD and CMOS in imaging and brightness. . However, in the 1-2Lux low-light test environment in this evaluation, we saw that the performance of CMOS is completely inferior to that of CCD (refer to Figure 1 for comparison of results). Therefore, for common low-light high-definition CCD technology, in From the perspective of clarity, CMOS has been the first choice for HD cameras.
For IP high-definition surveillance, because the video needs to be compressed by the encoder before transmission, the compressed video pixels are only about 400,000. Therefore, it is undoubtedly a better choice for practical and cheap CMOS devices when it is difficult for CCD to obtain good high-definition effects. This has been confirmed by 20 IP Megapixel cameras. In addition to the theory; from the perspective of the effect of clarity, CCD devices of the same size will be better than CMOS sensors in resolution, but if you ignore the size problem, CMOS sensors will have a better advantage, this It can also be seen from the fact that the cameras participating in the test used CMOS with a size of 1 / 2.7 â€and 1 / 2.8†accounting for more than half. This also allows chip manufacturers to increase production capacity in mass production of large-size photosensitive components. Furthermore, from the perspective of mass production, CMOS chips have established an absolute successor advantage over CCD sensors in the case of high-definition resolution.
There is another important discovery that must be mentioned. Compared to the CCD image response speed, the CCD sensor and the CMOS chip jointly participated in this time. From the image transmission response and the capture action, you can see the CMOS sensor The response processing speed is much faster. For SensorCapture 30fps or 60fps, this evaluation also confirmed that the smoother condition is 60fps, but this has nothing to do with resolution and image quality. It should be correct for the wrong ideas in the past. It is also seen on high-definition IP and SDI cameras. To 720P @ 60fps, 1080P @ 30fps or 1080P @ 60fps is what this Sensor needs. This is an important parameter point for real-time video, especially in the application of remote transmission or alarm screen. The relatively stable image and real-time performance of the Sensor sensor have seen mature and reliable performance from this evaluation.
Multitasking of DSP / ISP image processing components
As CMOS manufacturers continue to improve in technology. In effect, some CMOSSensors today are getting closer to the effect of CCD. However, a good image capture must also be matched with a good image processing DSP / ISP processing component in order to play an appropriate function, especially after the DSP / ISP chip overcomes the front-end power consumption problem, whether it is in image control or signal noise The functions of the wider dynamic part than the backlight have received obvious attention from the manufacturers. Whether ISP is used alone or packaged with Sensor or IPCodec, many manufacturers have worked hard to develop multi-tasking DSPs with higher wide dynamics, better glare suppression and more stable color control on IP and SDI cameras / ISP chip, this situation is in the wide dynamic project of this Secutech evaluation project (refer to Figure 2 WDR wide dynamic comparison), you can also see the R & D capabilities and image processing components of various manufacturers in component adoption and API / SDK development The added value to the camera is also undoubtedly bringing ISP a broader application prospect in the future.
Through the ISP components in the circuit control of image sharpening and color tone and white balance, from the evaluation, we also see that the ISP control, especially the main control of white balance, is presented in DSP / ISP of different brands The performance of the color reproduction power and the white balance control power under each default value (refer to Figure 3AWB color difference white balance comparison), from the basic color complementary rendering and grayscale color rendering and white complete stability, DSP The / ISP image processing chip plays a very important role in decision-making. That is to say, the reproduction capability of the analog or digital signals from the sensor in the DSP / ISP is the key to determine the final color stability and tone deviation of the IP and SDI HD cameras . This is also a very important reference factor for engineering companies in product selection.
What does the Backend component of IP / SDI bring?
The IP compression chip and SDI transmitter chip are the core parts of the camera assembly. For the transmission result of the last one in IP and SDI, the IP compression chip and SDI transmitter chip play a key role. For IP Megapixel HD, the compression chip CODEC dominates some of the conditions for image output and post-transmission, including image traffic, stream number provision, processing CPU speed, screen delay difference, image number speed and heat consumption, etc. Some of the SoCs used by various manufacturers do not seem to have changed much from the content adopted in the past year, but from the perspective of the entire compressed SoC, the image packet flow setting and the number of streams provided and applications, CODEC CPU processing speed and The number of images adjusted is very different from the past. Take the setting of an image packet flow, because the general network bandwidth is limited, the high packet flow brought by high-definition video has become the biggest promotion bottleneck of high-definition. How to reduce the code flow as much as possible to transmit and maintain high-quality images is the primary problem that HD surveillance needs to solve.
At the same time, due to the heterogeneity of the network, different networks have different channel characteristics, different users enjoy different network bandwidth, and even the bandwidth of the same user may change at any time. With this development, we can see from the evaluation that various manufacturers are improving the compression rate of variable code streams and fixed code traffic. In the past, 4Mbps 720P and 8Mbps 1080P were required. With the efforts of chip manufacturers and camera manufacturers, the current IP Million HD can already be compressed between 2-3Mbps at 720P @ 30fps, and 1080P @ 30fps can also be controlled by bandwidth flow control below 6Mbps, which can be said to greatly improve the transmission burden of IP million HD and it is not changing For such compression results, the compression ratio will not cause thermal effects, so that the compression chip can operate stably and reliably.
Because the performance of CODECSoC has been greatly improved, and manufacturers have gradually developed consensus on the application specifications of streaming types, such as the main stream is supplied to the recording server in H.264, 720P or 1080P @ 30fps, and the monitoring is in MPEG4 Or H.264D1 or CIF @ 30 / 60fps to send monitoring images, while the mobile remote can provide 3-5fps frames for mobile monitoring by MJPEG or H.264, such a reasonable control that does not affect the flow and number of sheets The way is forming an IP high-definition monitoring standard, and the result of this application is to greatly reduce the excessive generation of image delay. From the results of this Secutech evaluation, it is more than 500ms that the human eye can distinguish the delay. The image delay has disappeared in the IP million high-definition participating models (refer to the picture lag comparison in Figure 4). In the same period last year, there were still many delays exceeding 500ms under the same detection conditions. This series of product component application improvements from Sensor to ISP to CODECSoC, the end result is a smooth and quality monitoring screen presentation. In addition, such a low compression ratio is a relatively favorable situation for storage, because the improvement of image resolution will inevitably consume more storage space. Taking 1920 × 1080 @ 30 frame video as an example, using the H264 encoding algorithm, in order to ensure clarity, the code stream is at least 6Mbps or more, which is about 4-8 times that of D1 standard definition video. Therefore, under the reduced bitstream, the recording space required for storage will no longer be shortened, which will bring down the storage cost.
The above is the observation result of the component parts in the evaluation of the IPMegapixel and HD-SDI main sample components in the evaluation of SecutechAward. Of course, the problem of a camera cannot be only in these aspects. For example, there are HD displays. Compared with standard-definition video, the amount of information in high-definition video is much richer, and the corresponding display performance requirements are also greatly improved, but the camera is only the front end of high-definition monitoring after all. Whether it is IP or SDI, we will see from the evaluation Until it is showing a stable and moving forward.
Anti-Uv Positve Ps Plate,Ps Printing Plate For Print,Ps Printing Plate,Green Positive Ps Plates
SUZHOU HUAGUANGBAOLI PRINTING PLATE CO.,LTD , https://www.huabaoiguang.com