Mount bayonet software Chen days HD bayonet system --- high performance, high recognition rate, high capture rate, easy to use! The video track vehicles can be used intersections and sections of the bayonet illegal capture can be detected through a red light, compaction line, pressure double yellow lane change, turn left, straight, turn right, retrograde, illegal parking. Has a very powerful function. HD video SupIVS-HV bayonet system Description: HD mount system is a high-performance, all-weather passing vehicles automatically capture and license plate number recognition system, using high-resolution camera as a capture device, the use of state-of-the-art image processing and artificial intelligence technology, combined with neural network technology, pattern recognition technology that can automatically adapt to all kinds of weather, day and night lighting changes, without requiring the user to perform complex parameter adjustment. Two trigger mode for coil and video systems can be used. The system can be installed on major trunk roads, highways, bridges, tunnels and other places, to increase traffic monitoring efforts to curb speeding or accident escape, eliminating traffic safety hazard range of applications such as: HD bayonet system can be applied to intelligent transportation systems, road security mount intelligent vehicle monitoring, toll station control, highway load limit, speed limit control, parking management, and so on. Trigger mode: loop detectors and video detection is optional. The video detection largest supports 8-lane detection direction. Capture device: each lane at least more than 1.4 million pixel high-definition camera: the sense coil detection method: Direct to sense with high-definition cameras to capture, front-end without any IPC device. 2: video detection method: A: high-definition video stream mode: can be embedded machine direct detection, capture, identify high-definition video streaming. The machine can identify the maximum support 500 million pixels, can support four lanes;: HD video streaming DSP license plate recognition: to use direct detection, capture, identify high-definition video streaming. The machine can identify the maximum support 5 million pixels, maximum support four lanes, currently recommends 2-megapixel one, the basic guarantee will not be missed shot; C: & Poor clear trigger HD: 1 machine maximum support 4-way 16-lane detection, capture, and identification. The general definition detection, high-definition camera to capture the way. Features: A. all-weather vehicle after real-time capture and monitoring B. can monitor up to capture direction for each of the eight lanes (running red lights bayonet type, maximum support 4-way lane 16) C. system through the detection coil vehicle image capture, photo 1660 × 1024 × 24-bit resolution, up to 1.4 million pixels (industrial grade high-definition network camera, Gigabit Ethernet port, 1.4 million pixels, 2-megapixel and 4.0 megapixel, day and night can clearly others cab.), each time a picture, able to clearly identify the license plate, and live to see the cab. Image capture, the system is able to detect the vehicle information, such as date, time, location, direction, license plate number, vehicle speed, vehicle type (suspects, speeding, normal) superimposed on the image, easy artificial discriminant and after the evidence . Due to the photo large amount of information to be able to clear the the others driver and vice cab circumstances provide important evidence for law and order in this case. The D. plate numbers automatically identify, capture and identify the pictures stored on the license plate number and license plate color identification, vehicle identification and vehicle body color recognition E., and can be stored for three months of historical data in real time; F. uploaded to the server, and dedicated mass storage on the server, and real-time alarm in the center. HD the bayonet system performance indicators: Capture rate: video detection weather> 95%, coil detection> 99% recognition rate: 98% recognition error rate: <1% recognition accuracy: more than approximately 90% of the weather, the speed error : within ± 5%. System configuration: operating system WindowXP, CPU Pentium 4 3.0, 1GB of RAM and hard disk 80G