Flexible video surveillance / security system implementation solution
The video surveillance and security industries are undergoing a tremendous transformation, and are developing from traditional analog CCTV cameras to logic-based digital cameras. The trend of higher video resolution, image signal processing, advanced video analysis, multi-camera systems and digital video compression has driven this shift. As a result, new challenges have emerged in the design of video surveillance cameras and digital video recorders (DVRs), including the increased use of CMOS sensors, network IP camera technology, increased system integration, smaller form factor, and the use of advanced Codec and increase DSP function.
Due to the diversity and competitiveness of the video surveillance / security market, it is often difficult for manufacturers to add differentiated features to their own products to differentiate their products from competitors ’products. Therefore, similar cameras and DVR products are very common. However, the trends mentioned above have created new opportunities for video security / surveillance original product manufacturers to achieve product differentiation.
Meeting the requirements of these applications requires additional processing power. Current digital video cameras use DSP chips, specialized ASSPs or ASICs to provide processing functions, but these solutions have some challenges. Using DSP chips often has performance bottlenecks because these chips usually process image signal processing tasks in a serial manner. ASSP may provide more performance, but often at the cost of design flexibility. ASIC has more performance optimization, but the output is not large enough to verify the cost and time required for ASIC development. Designers need a flexible method to provide the computing power needed from low-end camera / DVR to high-end applications for different market segments.
Implementation options
Video surveillance systems can be divided into two categories: camera systems and DVR systems. The camera system can integrate a single camera or multiple cameras. Some systems integrate DVR with a single camera or multiple cameras.
Figure 1: Block diagram of a multi-camera DVR video surveillance system.
Figure 1 is a general multi-camera DVR video surveillance system. The video source comes from CMOS video sensor, CCD video sensor or analog video source. The proprietary video interface converts the input video stream into a common format, and then multiplexes different video streams (and audio) and performs preprocessing in the image signal processing unit. The purpose of video preprocessing is to reduce noise and eliminate pixel defects.
Video analysis technology is applied to detect motion in predefined pictures. This motion detection output can reduce the required storage capacity. In some video surveillance applications, video analysis is necessary, such as people / car statistics, car license plate number recognition and / or face recognition.
A typical video surveillance system with multiple camera video sources will always generate a large amount of data, so reducing the required storage capacity is very important. MPEG-4, H.264, and MJPEG are used to reduce the required storage capacity. H.264 is a popular compression algorithm for video surveillance applications because it can provide good video quality at a fairly low bit rate (only MPEG-2 or MPEG-4 Part 2 half or lower bit rate) Features.
The compressed data is stored to the video storage server through the hard drive interface, or sent via the Ethernet network. Video data is usually taken from the hard drive before being transferred to the display, and undergoes decoding and some image post-processing, such as scaling, color space conversion, or overlay applications. Usually, a memory interface (such as DDR2) is used to store video frames. In addition, some systems require real-time encryption of video content to ensure security and source / user authentication. Finally, various tasks need to be controlled and coordinated / scheduled by the processor.
Figure 2: Selection steps based on throughput, flexibility, batch size, and form factor requirements.
Figure 2 illustrates the requirements of the system, such as throughput, flexibility, expected batch size and form factor, how to drive different implementation technologies and devices. For low data throughput requirements, the DSP processor is the most cost-effective device. However, the expected volume during the life of the product and the flexibility of the hardware will greatly influence the designer to choose between ASIC / ASSP and FPGA devices. If it is a system requirement with low hardware flexibility and a large volume is expected, the designer prefers the ASIC / ASSP solution; and for a system requirement with high hardware flexibility and a small expected volume, the designer prefers the FPGA solution. Common small video surveillance cameras have small size requirements, and designers prefer to use non-volatile FPGAs because they do not require additional external non-volatile configuration storage devices.
System integration combines the advantages of different devices: DSP processors sometimes integrate ASSP IP modules, ASSP / ASIC sometimes integrates processors, and FPGAs sometimes integrate hard processors and hard IP cores.
For the video surveillance system (Figure 1), due to the timing nature of these tasks, the DSP processor can effectively implement video analysis functions and processor blocks. However, in some cases, the FPGA's acceleration function helps to perform some video analysis functions, such as motion detection and facial recognition.
The H.264 encoder shown in Figure 1 is a functional block that requires high throughput, low hardware flexibility, large and complex, and is usually implemented with ASSP or ASIC to obtain a small size solution.
But ASIC and ASSP bring high non-recurring engineering costs, unless the volume is very large. If only a small percentage of users use special function modules, then FPGA solutions can provide better returns. For ASICs and ASSPs, the limited functionality and inherent inflexibility of these devices does not allow for the improvement or addition of new functions based on market requirements after the design is completed. FPGA can provide a wide range of functions and a high degree of flexibility.
Video surveillance system modules with the following requirements will benefit from the flexibility of FPGAs: (1) This video and audio interface and multiplexer require hardware flexibility to support different cameras and different numbers of cameras. (2) The DDR interface must support different memory bus widths and different DDR standards. (3) Support various image signal processing algorithms (two-dimensional FIR filter, two-dimensional median filter, scaling, edge detection, gamma correction, alpha blending, white balance, lens shading correction, defective pixel correction, demosaicing, Progressive scanning, color space correction, etc.), in a particular video surveillance camera implementation may require a subset of various algorithms. The effective implementation of these algorithm structures depends on the throughput requirements, which in turn depend on the number of cameras in the system and the video standard sampling rate. (4) The hard disk drive interface and Ethernet interface are not always required in the video surveillance camera system, but can be used as an alternative. (5) The display interface is not always required. When it is required, there are many different display interfaces that can be implemented. (6) Encryption and authentication are only required in certain systems. Depending on the different throughput requirements of different video formats and the number of different cameras in a particular surveillance system, the optimal encryption and authentication system may require different sizes and optimal implementation architectures.
Many network IP cameras are constrained by their very compact form factor, power consumption, and cost, especially embedded camera / DVR systems. Using non-volatile FPGA as a solution can solve these problems and gain the traditional advantages of FPGA.
Geared Stepper Motor,Planet Gearbox,Spur Gearbox,Nema23 Geared Electric Motor
Changzhou Sherry International Trading Co., Ltd. , https://www.sherry-motor.com