CN205545633U - Night driver assistance system based on infrared thermal imaging - Google Patents
Night driver assistance system based on infrared thermal imaging Download PDFInfo
- Publication number
- CN205545633U CN205545633U CN201620281043.6U CN201620281043U CN205545633U CN 205545633 U CN205545633 U CN 205545633U CN 201620281043 U CN201620281043 U CN 201620281043U CN 205545633 U CN205545633 U CN 205545633U
- Authority
- CN
- China
- Prior art keywords
- input
- outfan
- electrically connected
- infrared
- video
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Expired - Fee Related
Links
Abstract
The utility model discloses a night driver assistance system based on infrared thermal imaging, including on -vehicle thermal infrared imager, pedestrian detecting system and car audio and video system, the output and pedestrian detecting system's the input of on -vehicle thermal infrared imager are connected, pedestrian detecting system's the output and the input of car audio and video system are connected, on -vehicle thermal infrared imager includes infrared camera lens, infrared detector and image processor, infrared camera lens and infrared detector connect, infrared detector's output and image processor's input electric connection, image processor's output and pedestrian detecting system's input signal connection. Having developed to this system creativeness nature infrared image's digital details reinforcing algorithm (DDE), having improved the details of image and the adaptability of scene effectively, the driver can clearly see the road surface condition through the display, has improved the security of traveling greatly.
Description
Technical field
This utility model belongs to automotive safety ancillary technique field, is specifically related to a kind of based on infrared thermal imaging
DAS (Driver Assistant System) at night.
Background technology
Traffic safety is one great focal issue of society, is paid close attention to by national governments and society,
Also be intelligent transportation system to endeavour solve significant problem.The most all bring because of vehicle accident every year
Massive losses, wherein accounts for more than 90% due to what driver error caused, therefore carries out based on intelligent transportation system
The Key Technology of Traffic Safety research of system becomes the most urgent.Research shows, if vehicle front peace
Equipped with the warning system of spacing alarm, radar velocity measurement etc, occur the probability of vehicle accident can reduce 62%.
If running into misty rain weather or night running, the probability of happening of traffic accidents also can be greatly improved.
By DAS (Driver Assistant System) at night, driver, at visibility extremely low night or misty rain weather, depends on
So it is observed that the road conditions in 300 meters of front, if vehicle front occur pedestrian (include by bike or
The pedestrian of motorcycle), system can send image and voice alarm, reminds driver pedestrian's location,
Vehicle accident prevents trouble before it happens.
Utility model content
The purpose of this utility model be to provide a kind of night of based on infrared thermal imaging DAS (Driver Assistant System),
With the problem solving to propose in above-mentioned background technology.
For achieving the above object, this utility model provides following technical scheme: a kind of based on infrared thermal imaging
DAS (Driver Assistant System) at night, including vehicle mounted infrared thermal imaging system, pedestrian detecting system and vehicle carried video system,
The described outfan of vehicle mounted infrared thermal imaging system is connected with the input of pedestrian detecting system, described pedestrian detection
The outfan of system and the input of vehicle carried video system connect, and described vehicle mounted infrared thermal imaging system includes infrared
Camera lens, Infrared Detectors and image processor, described infrared lens and Infrared Detectors connect, described red
The outfan of external detector and the input of image processor are electrically connected with, the output of described image processor
End is connected with the input end signal of pedestrian detecting system, and described pedestrian detecting system includes video decoding chip,
The input of described video decoding chip is connected with the output end signal of vehicle mounted infrared thermal imaging system, described video
The outfan of decoding chip is electrically connected with the input of Video processing front end, described Video processing front end
Outfan is electrically connected with by the input of pedestrian detection device with video frequency output array, described video frequency output
The outfan of array is electrically connected with the input of Video processing rear end, the output of described Video processing rear end
End is electrically connected with the input of video display units.
Preferably, described image processor includes FPGA processor and dsp processor, at described FPGA
The input of reason device is electrically connected with by the outfan of A/D converter with Infrared Detectors, described infrared spy
The outfan of the input and thermoelectric refrigerator of surveying device is electrically connected with, two groups of inputs of described thermoelectric refrigerator
End is electrically connected with the input of clock management circuits and D/A converter respectively, described FPGA processor and
Dsp processor is electrically connected with, and the outfan of described clock management circuits and the input of dsp processor are electrical
Connect, and the outfan of the input of described clock management circuits and FPGA processor is electrically connected with, described
The outfan of FPGA processor is electrically connected with by the input of D/A converter and video driver, described
The outfan of video driver is electrically connected with the input of video display units.
Preferably, the outfan of described dsp processor is electrically connected with RS232 interface.
Preferably, described FPGA processor and dsp processor are all electrically connected with flash memories.
Technique effect of the present utility model and advantage: should night of based on infrared thermal imaging DAS (Driver Assistant System),
After pedestrian detecting system is arranged on vehicle mounted infrared thermal imaging system, the used time can not be bypassed (closedown), to subtract
Few system power expense, if starting pedestrian detecting system, then the CVBS signal of vehicle mounted infrared thermal imaging system
Can be input in system, be converted into digital signal through video encoder and enter digital signal processor (DSP),
After the DSP video signal to collecting carries out the process such as image segmentation, template matching, machine learning, detection
To pedestrian and be marked (rectangle frame being added in video flowing, pedestrian is enclosed), finally by DSP
Interior video ADC be again decoded into the CVBS signal with pedestrian detection information export analog monitor or
Person's vehicle carried video system, the video signal of output is exactly the real-time infrared image with information of road surface, and
There is pedestrian detection function;Because pedestrian detection is based on infrared chart, therefore from thermal infrared imager out
Image must be clear fine and smooth, there is high contrast and signal to noise ratio, otherwise influence whether the segmentation of image
With the detection of pedestrian, the numerical details developing to native system creativeness infrared image strengthens algorithm
(DDE) details of image and the adaptability of scene, it are effectively improved.Pedestrian detection algorithm use double
Threshold Image Segmentation, with SVM classifier separate image remote, in, carry out template matching the most again, also
Applying many frame checks be effectively improved verification and measurement ratio and lower false alarm rate, these algorithms all receive good effect,
Driver can clearly see surface conditions by display, substantially increases the safety of traveling.
Accompanying drawing explanation
Fig. 1 is principle modules schematic diagram of the present utility model;
Fig. 2 is vehicle mounted infrared thermal imaging system principle modules schematic diagram of the present utility model;
Fig. 3 is pedestrian detecting system principle modules schematic diagram of the present utility model;
Fig. 4 is image processor principle modules schematic diagram of the present utility model.
Detailed description of the invention
Below in conjunction with the accompanying drawing in this utility model embodiment, to the technology in this utility model embodiment
Scheme is clearly and completely described, it is clear that described embodiment is only this utility model one
Divide embodiment rather than whole embodiments.Based on the embodiment in this utility model, this area is common
The every other embodiment that technical staff is obtained under not making creative work premise, broadly falls into this
The scope of utility model protection.
The auxiliary at a kind of based on infrared thermal imaging night that this utility model provides as Figure 1-3 is driven
System, including vehicle mounted infrared thermal imaging system, pedestrian detecting system and vehicle carried video system, described vehicle mounted infrared
The outfan of thermal imaging system is connected with the input of pedestrian detecting system, the outfan of described pedestrian detecting system
Connecting with the input of vehicle carried video system, described vehicle mounted infrared thermal imaging system includes infrared lens, infrared spy
Survey device and image processor, described infrared lens and Infrared Detectors to connect, described Infrared Detectors defeated
The input going out end and image processor is electrically connected with, the outfan of described image processor and pedestrian detection
The input end signal of system connects, and described pedestrian detecting system includes video decoding chip, described video solution
The code input of chip is connected with the output end signal of vehicle mounted infrared thermal imaging system, described video decoding chip
Outfan is electrically connected with the input of Video processing front end, and the outfan of described Video processing front end passes through
Pedestrian detection device is electrically connected with the input of video frequency output array, the output of described video frequency output array
End is electrically connected with the input of Video processing rear end, and the outfan of described Video processing rear end shows with video
Show that the input of device is electrically connected with.
Specifically, as shown in Figure 4, described image processor includes FPGA processor and dsp processor,
The outfan of described dsp processor is electrically connected with RS232 interface, at described FPGA processor and DSP
Reason device is all electrically connected with flash memories, and the input of described FPGA processor passes through A/D converter with infrared
The outfan of detector is electrically connected with, the input of described Infrared Detectors and the outfan of thermoelectric refrigerator
Be electrically connected with, two groups of inputs of described thermoelectric refrigerator respectively with clock management circuits and D/A converter
Input be electrically connected with, described FPGA processor and dsp processor are electrically connected with, described Clock management
The outfan of circuit and the input of dsp processor are electrically connected with, and the input of described clock management circuits
End is electrically connected with the outfan of FPGA processor, and the outfan of described FPGA processor is turned by D/A
The input of parallel operation and video driver is electrically connected with, and the outfan of described video driver and video show
The input of device is electrically connected with.
The work process of this image processor:
(1) power on beginning, and dsp processor will be stored in data (the DSP program sum inside flash memories
According to, and the gain after nonuniformity correction and side-play amount) it is loaded into DSP internal RAM and external SRAM respectively
(look-up table LUT), completes the initialization of necessity;
(2) system starts, the analog-signal transitions of IRFPA is become digital signal to deliver to FPGA by A/D
In processor, and the nonuniformity correction coefficient finding out correspondence in LUT for address wire with these data increases
Benefit and side-play amount, the D/A that the data after processing in FPGA processor deliver to rear end by FIFO is carried out
Analog digital conversion and Video Composition, and show the most in real time;
(3) when needs update correcting gain and side-play amount coefficient, sent to dsp processor by serial ports
Interrupt instruction, the control of external SRAM is given dsp processor by FPGA processor;
(4) dsp processor is under the control of outside serial communication, gathers the number of responses under high temperature and low temperature
According to, and calculate gain and side-play amount according to two point correction algorithm, there is external SRAM in result of calculation, write
Flash memories;
(5) giving FPGA processor by the control of external SRAM, FPGA processor starts real time correction
Work.For simplied system structure, it is achieved miniaturized system designs, and data buffering FIFO is at FPGA
Reason device realizes, and is greatly saved system space and cost, and decrease that dsp processor response interrupts time
Between.
The operation principle of this system:
This system uses 384*288 non-refrigeration focal surface Infrared Detectors, receives infrared band and is
8-14wM, analogue signal, with the analogue signal of image information, is carried out by detector output with high-speed ADC
Gathering, be quantized into 14B work T digital signal and send into FPGA processor, FPGA processor is at microcontroller
Under cooperation, digitized infrared image signal is carried out Nonuniformity Correction, rejects blind element, utilize science
Computational methods carry out image enhaucament, filtering, noise reduction, greyscale transformation, overlaying character etc., change into 4:2:2
BT.656 video standard signal export to external video decoder, change into composite video signal (CVBS)
Giving display to show, the image seen the most over the display is exactly infrared chart.
Pedestrian detecting system:
As it is shown on figure 3, the CUBS signal of vehicle-mounted type thermal infrared imager output decodes through TVP5146 video,
The BT656 signal being quantized into 8bit enters the Video processing front end (VPFE) of DSP, and collected regards
Frequently signal is carried out a series of process such as image segmentation, machine learning, template matching by pedestrian detection mastery routine,
Pedestrian labelling detected, be converted into analogue signal through the Video processing rear end (VPBE) of DSP and send aobvious.Cause
Complicated for method comparison, code after at utmost optimizing process per second can only achieve 18 frames, so per second also
The image having 7 frames does not process, and is directly sent out to video display units from thermal infrared imager, but not shadow
Ring Detection results.
Should night of based on infrared thermal imaging DAS (Driver Assistant System), pedestrian detecting system is arranged on vehicle mounted infrared
After thermal imaging system, the used time can not be bypassed (closedown), to reduce system power expense, if started
Pedestrian detecting system, then the CVBS signal of vehicle mounted infrared thermal imaging system can be input in system, through Video coding
Device is converted into digital signal and enters digital signal processor (DSP), the DSP video signal to collecting
Carry out after image segmentation, template matching, machine learning etc. process, detect pedestrian and be marked (
Rectangle frame is added in video flowing, and pedestrian is enclosed), finally again decoded by DSP interior video ADC
The CVBS signal with pedestrian detection information is become to export analog monitor or vehicle carried video system, output
Video signal be exactly the real-time infrared image with information of road surface, and have pedestrian detection function;Because
Pedestrian detection, therefore must be clear fine and smooth from thermal infrared imager image out based on infrared chart,
There is high contrast and signal to noise ratio, otherwise influence whether the segmentation of image and the detection of pedestrian, native system
The numerical details creatively developing infrared image strengthens algorithm (DDE), is effectively improved image
Details and the adaptability of scene.Pedestrian detection algorithm uses the segmentation of dual threshold image, uses SVM classifier
Separate image remote, in, carry out template matching the most again, also apply many frame checks to be effectively improved inspection
Survey rate and attenuating false alarm rate, these algorithms all receive good effect, and driver can be clear by display
See surface conditions, substantially increase the safety of traveling.
Last it is noted that the foregoing is only preferred embodiment of the present utility model, and need not
In limiting this utility model, although this utility model being described in detail with reference to previous embodiment,
For a person skilled in the art, it still can be to the technical scheme described in foregoing embodiments
Modify, or wherein portion of techniques feature carried out equivalent, all in spirit of the present utility model and
Within principle, any modification, equivalent substitution and improvement etc. made, should be included in of the present utility model
Within protection domain.
Claims (4)
- null1. one kind night of based on infrared thermal imaging DAS (Driver Assistant System),Including vehicle mounted infrared thermal imaging system、Pedestrian detecting system and vehicle carried video system,It is characterized in that: the described outfan of vehicle mounted infrared thermal imaging system is connected with the input of pedestrian detecting system,The outfan of described pedestrian detecting system and the input of vehicle carried video system connect,Described vehicle mounted infrared thermal imaging system includes infrared lens、Infrared Detectors and image processor,Described infrared lens and Infrared Detectors connect,The outfan of described Infrared Detectors and the input of image processor are electrically connected with,The outfan of described image processor is connected with the input end signal of pedestrian detecting system,Described pedestrian detecting system includes video decoding chip,The input of described video decoding chip is connected with the output end signal of vehicle mounted infrared thermal imaging system,The outfan of described video decoding chip is electrically connected with the input of Video processing front end,The outfan of described Video processing front end is electrically connected with by the input of pedestrian detection device with video frequency output array,The outfan of described video frequency output array is electrically connected with the input of Video processing rear end,The outfan of described Video processing rear end is electrically connected with the input of video display units.
- nullA kind of night of based on infrared thermal imaging the most according to claim 1 DAS (Driver Assistant System),It is characterized in that: described image processor includes FPGA processor and dsp processor,The input of described FPGA processor is electrically connected with by the outfan of A/D converter with Infrared Detectors,The input of described Infrared Detectors is electrically connected with the outfan of thermoelectric refrigerator,Two groups of inputs of described thermoelectric refrigerator are electrically connected with the input of clock management circuits and D/A converter respectively,Described FPGA processor and dsp processor are electrically connected with,The outfan of described clock management circuits and the input of dsp processor are electrically connected with,And the outfan of the input of described clock management circuits and FPGA processor is electrically connected with,The outfan of described FPGA processor is electrically connected with by the input of D/A converter and video driver,The outfan of described video driver is electrically connected with the input of video display units.
- A kind of night of based on infrared thermal imaging the most according to claim 2 DAS (Driver Assistant System), it is characterised in that: the outfan of described dsp processor is electrically connected with RS232 interface.
- A kind of night of based on infrared thermal imaging the most according to claim 2 DAS (Driver Assistant System), it is characterised in that: described FPGA processor and dsp processor are all electrically connected with flash memories.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201620281043.6U CN205545633U (en) | 2016-04-01 | 2016-04-01 | Night driver assistance system based on infrared thermal imaging |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201620281043.6U CN205545633U (en) | 2016-04-01 | 2016-04-01 | Night driver assistance system based on infrared thermal imaging |
Publications (1)
Publication Number | Publication Date |
---|---|
CN205545633U true CN205545633U (en) | 2016-08-31 |
Family
ID=57141765
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201620281043.6U Expired - Fee Related CN205545633U (en) | 2016-04-01 | 2016-04-01 | Night driver assistance system based on infrared thermal imaging |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN205545633U (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111175239A (en) * | 2020-01-19 | 2020-05-19 | 北京科技大学 | High-spectrum nondestructive testing and identifying system for imaging of colored drawing cultural relics under deep learning |
-
2016
- 2016-04-01 CN CN201620281043.6U patent/CN205545633U/en not_active Expired - Fee Related
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111175239A (en) * | 2020-01-19 | 2020-05-19 | 北京科技大学 | High-spectrum nondestructive testing and identifying system for imaging of colored drawing cultural relics under deep learning |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2019183751A1 (en) | Detection and warning method for snow and ice in front of vehicle, storage medium, and server | |
WO2019183752A1 (en) | Method for detecting and warning of accumulated snow and ice in front of vehicle, storage medium and server | |
CN205123883U (en) | Shortwave infrared imaging system based on inGaAs detector | |
TW200824948A (en) | Lane departure warning method and apparatus of using the same | |
CN102592332A (en) | Automobile driving recording system | |
CN201677853U (en) | Anti-theft system based on human face recognition for controlling automobile ignition | |
CN205545633U (en) | Night driver assistance system based on infrared thermal imaging | |
JP2002362302A (en) | Pedestrian detecting device | |
CN206162876U (en) | Road speed limit prison bat system based on visibility detection | |
CN106248635B (en) | A kind of visibility detecting method based on digital camera | |
CN104331949A (en) | Automobile data recorder | |
CN203984590U (en) | The vehicle-mounted safe examination system of a kind of bus based on mm-wave imaging technology | |
CN202923497U (en) | Device for enhancing visual field of driver in rainy and foggy weather | |
CN107682661A (en) | The system and method for panorama driving recording are realized using panorama function of parking | |
CN205050342U (en) | Car snapshot system that makes a dash across red light that whistles | |
CN105046980A (en) | Detection system for red light running of vehicles | |
CN101837758A (en) | Road surface deceleration strip vehicle-mounted early warning device | |
CN205621280U (en) | A safe speed with apart from warning system for highway | |
US20190217774A1 (en) | Image recognition-based intelligent alarm lamp | |
CN207637310U (en) | The detection warning device of vehicle periphery mobile object | |
CN206436833U (en) | A kind of automobile instrument panel | |
CN106627149A (en) | Vehicle instrument panel and control method | |
CN205212930U (en) | Traffic camera prevents frostbite | |
CN210742957U (en) | Bus stop with flow of people statistics function | |
CN103606207A (en) | Vehicle-mounted safety device based on image identification |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C14 | Grant of patent or utility model | ||
GR01 | Patent grant | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20160831 Termination date: 20170401 |