Due to the need of safe city construction, the development of video surveillance has been further promoted, and thus the application of FPGA in this field has also been promoted.
Especially now that the requirements for multi-channel, high-definition, networking, high-speed communication interface, and intelligence have promoted the further development of the field of FPGA-based video wall controllers.
On the contrary, the advancement and renewal of FPGA chip technology, IP core, and reference design has promoted the development of video surveillance.
Now, it is difficult to meet the needs of high-performance systems by simply using DSP processors or off-the-shelf chips (ASSP).
However, due to the high integration and flexibility of current programmable devices, as well as low power consumption and wide operating range, their prices continue to drop. Therefore, the unique high performance and flexibility of programmable logic gate arrays (FPGAs) are used. , So that it can construct many video surveillance products.
1.What makes FPGAs remarkable?
FPGAs are programmable like GPUs or CPUs but are aimed at parallel, low-latency, high-throughput problems like inference and Deep Neural Networks.
FPGAs have a number of benefits, the most notable of which is speed.
While FPGAs run at a slow clock speed relative to modern CPUs, they are fundamentally concurrent, rather than running streams of sequential instructions, with data flowing optimally between these concurrent operations, resulting in a dramatic net increase in performance.
There is the potential for applications to run up to 100 times faster over the same code running on traditional CPUs.
FPGAs contain millions of reprogrammable logic blocks that can be used to perform many actions at the same time, delivering the benefits of parallelism and concurrency.
When writing code, engineers can take advantage of this parallel architecture by breaking problems down into well-structured, self-contained processes that can run concurrently.
For example, when an image is processed nonconcurrently, a single worker would process the whole image pixel by pixel. But when the same image is processed concurrently, it is broken down into pieces that are processed simultaneously by different workers, and then pieced back together.
This makes the process more complex but far quicker - the incoming data must be split apart in an optimal way, distributed efficiently to the workers, then the processed data collected and reassembled, ideally without blocking the pipeline of work.
In a normal CPU, this involves data being pushed and pulled from memory, and costly protocols for processes to agree on what is the current state of memory. Even the largest Intel CPUs have only 18
cores. In comparison, in an FPGA, data flow can be engineered so it never leaves the chip.
Tens of thousands of concurrent processes can occur, and the timing of the processing optimised so throughput
is always maximal.
2. The application of FPGA in intelligent video surveillance
At present, the resolution of IP cameras is gradually evolving from standard definition D1 to high definition (1920×1080), and local real-time compression must be performed, so hard compression can only be used. If multiple DSP processors are used, the system cost, integration, and power consumption will increase, which is unacceptable to users; if a single-chip low-cost FPGA device is used, the performance cannot meet the design requirements.
However, if a single-chip high-performance Stratix series FPGA device is used, the requirement can be met. Because this device has a corresponding structured ASIC-Hard-Copy series device, it can further reduce the cost to 1/10 and reduce the power consumption by 50%. Therefore, this FPGA device can be used as a single-channel high-definition IP camera
In order to monitor the multi-channel picture locally, it is usually necessary to multiplex the multi-channel video data and divide and scale the picture. Therefore, the standard CCIR656 format data must be sent to the video multiplexing scaling division part for processing.
The abundant memory resources in FPGA devices are more suitable for use as the line buffer necessary for the video multiplexing and scaling algorithm, so this part can quickly realize the screen multiplexing and scaling and segmentation functions.
Then it is sent to the multi-channel H.264 D1+CIF encoding part, and the powerful parallel processing capabilities inherent in FPGA can meet the processing speed requirements of the H.264 algorithm. Compared with multiple ASSP or DSP processor implementation schemes, a single-chip FPGA provides more stable system performance, lower cost and the best price/performance ratio.
3. Use FPGA to realize DSP real-time video processing function
Compared with ASSP and chipset solutions, FPGAs can provide different levels of flexibility according to the actual needs of design engineers and maintain significantly better performance than traditional DSPs.
Real-time video processing requires extremely high system performance, so almost all general-purpose DSPs with the simplest functions do not have this function.
The programmable logic device allows designers to use parallel processing technology to implement video signal processing algorithms, and only a single device can achieve the desired performance.
DSP-based solutions usually need to embed many DSPs on a single board to obtain the necessary processing capabilities, which will undoubtedly increase the overhead of program resources and data memory resources.
Because it is extremely difficult to send high-bandwidth video data and maintain appropriate quality of service (QoS) on extremely narrow transmission channels (such as wireless channels), designers are committed to improving error correction, compression, and image processing based on FPGA implementation. technology.
The core of the MPEG-4 algorithm is an operation called Discrete Cosine Transform (DCT). The DCT part has been standardized and can be effectively implemented in FPGA. Many dedicated MPEG decoders also use these parts (such as motion estimation modules). FPGA.
Because the FPGA can be reconfigured, the device can be easily refreshed and new algorithms can be integrated throughout the development phase (including after configuration).
Another important part of the video system is the color space conversion. The FPGA system architecture can adjust the algorithm of the application system to achieve the best performance and efficiency.
FPGA can provide the most practical and valuable high-efficiency and high-efficiency products through custom adjustments. Designers can compromise between the scope of application and speed, so as to realize the specified function at a much lower rate than the DSP clock.
For example, in the median filter application, the DSP processor needs 67 clock cycles to execute the algorithm, while the FPGA only needs to work at a frequency of 25MHz, because the FPGA can implement this function in parallel.
But the DSP that realizes the above-mentioned function must work under 1.5GHz frequency, it can be seen that in this particular application, the processing capacity of FPGA solution can reach 17 times of 100MHz DSP processor.
Many real-time image and video processing functions are suitable for implementation with FPGA devices, including: image rotation, image scaling, color correction and chroma correction, shadow enhancement, edge detection, histogram function, sharpening, median filter and speckle Analysis etc. Many functions are aimed at specific applications and systems, and are built on top of the core architecture (such as 2D-FIR filters).
4. Use FPGA to build image and video wall controllers for embedded systems
The use of FPGA devices to build video and image controllers is making image display technology enter more and more embedded applications. Due to the perfect combination of performance and flexibility, FPGA applications in the DSP field are becoming more and more common.
iSEMC has launched a new low-power field programmable gate array (FPGA) series of video wall controllers, further expanding its resources for a wide range of low-power programmable solutions for power-conscious designs.
The new FPGA devices provide the best power consumption, area, logic, and function ratios per I/O in programmable logic devices. This makes it an ideal choice for portable electronic devices in consumer electronics, industrial, communications, medical, and test applications, especially those that require I/O-intensive memory bus operations, general-purpose I/O expansion, sequencing, interface conversion, storage, and The application of man-machine interface touch screen and keyboard technology.
