CN104751138B - A kind of vehicle mounted infrared image colorization DAS (Driver Assistant System) - Google Patents
A kind of vehicle mounted infrared image colorization DAS (Driver Assistant System) Download PDFInfo
- Publication number
- CN104751138B CN104751138B CN201510141721.9A CN201510141721A CN104751138B CN 104751138 B CN104751138 B CN 104751138B CN 201510141721 A CN201510141721 A CN 201510141721A CN 104751138 B CN104751138 B CN 104751138B
- Authority
- CN
- China
- Prior art keywords
- infrared image
- infrared
- image
- digital processing
- processing element
- 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
Landscapes
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The present invention relates to a kind of vehicle mounted infrared image colorization DAS (Driver Assistant System), including the infrared thermal imaging device being loaded on vehicle body, the infrared thermal imaging device is connected with the digital processing element being loaded on vehicle body, digital processing element is transferred to after gathering infrared image in real time by infrared thermal imaging device, the infrared image after digital processing element is handled is shown by the display unit on vehicle body.The invention provides it is a kind of under overnight sight by the scheme of infrared image colorization, simple infrared hybrid optical system is difficult to the difficulty distinguished before compensate for, and enhances the ability that driver obtains road information at night, is the important reference of nighttime driving.The incidence of vehicle at night accident can greatly be reduced.
Description
Technical field
The present invention relates to a kind of vehicle mounted infrared night vision DAS (Driver Assistant System), the particularly colorization of vehicle mounted infrared night vision image
System.
Background technology
With the increasingly reinforcing of fast development and the security protection consciousness of automobile market, people want to automotive safety safeguards technique
More and more higher is sought, vehicle mounted infrared auxiliary drive system also slowly enters our life.It can obtain road and road at night
The infrared image of the scenes such as other trees building, but infrared image is without colour, is a gray level image and shadow-free, contrast
It is low, therefore to the human eye, resolution ratio is low, and visual effect obscures.Because the color levels that human eye can be differentiated are the several of tonal gradation
Hundred times, therefore, if by vehicle mounted infrared image colorization, the visual analysis effect of image will be greatly improved, help driver to figure
Made rapidly as content and correctly understand and judge, and then improve night running safety.
The content of the invention
It is an object of the invention to provide it is a kind of can be by the system of vehicle mounted infrared image colorization.
In order to achieve the above object, driven the technical scheme is that providing a kind of vehicle mounted infrared image colorization auxiliary
System is sailed, including the infrared thermal imaging device being loaded on vehicle body, the infrared thermal imaging device and the numeral being loaded on vehicle body
Processing unit is connected, and digital processing element is transferred to after gathering infrared image in real time by infrared thermal imaging device, through digital processing
Infrared image after cell processing is shown by the display unit on vehicle body, it is characterised in that digital processing element is to receiving
Infrared image be handled as follows:
Step 1, image preprocessing is carried out to infrared image, while the noise on removing infrared image, increase is infrared
The contrast of image;
Step 2, using watershed algorithm to pretreated infrared image carry out background segment, by the background of infrared image
Different zones are divided into, corresponding color is assigned to corresponding region using the distinctive texture information in each region;
Step 3, the object for possessing higher heat are located at the prospect of infrared image, and the prospect of infrared image is handled,
Assigned specifically after the principle that dynamic threshold is combined comes out the object detection for possessing higher heat using statistics with histogram
Color with warning function;
Step 4, background and prospect are assigned into coloured colorization infrared image shown by display unit.
Preferably, in the step 2, background segment is carried out to pretreated infrared image using watershed algorithm
Concretely comprise the following steps:
Step 2.1, the unrelated marginal information using priori removal infrared image, infrared image background content is divided into
Following three regions, the above is sky areas, be below road surface region, the right and left is trees region;
Step 2.2, using watershed algorithm regional is calculated, the calculating process of water ridge algorithm is an iteration
Annotation process, concretely comprise the following steps:The gray level of each pixel in each region is sorted from low to high first, then from low
During being flooded to height realization, the domain of influence of each local minimum in h rank height is sentenced using first in first out structure
Disconnected and mark.
Preferably, in the step 3, higher thermal will be possessed with the principle that dynamic threshold is combined using statistics with histogram
What the object detection of amount came out concretely comprises the following steps:
Step 3.1, according to priori set an initial threshold, by the initial threshold assign dynamic threshold variable, will
First frame infrared image is set to present frame infrared image;
Step 3.2, with dynamic threshold variable binary conversion treatment is carried out to present frame infrared image, higher than the pixel of threshold value
The coordinate range of pedestrian is referred to as, at the same time, the statistics with histogram of gray scale is carried out to present frame infrared image, finds appearance frequency
The maximum gray value of rate, dynamic threshold variable is assigned using it as new threshold value;
Step 3.3, after next frame infrared image is set into present frame infrared image, return to step 3.2 re-executes, until
All infrared images are disposed.
Compared with prior art, the beneficial effects of the invention are as follows provide one kind under overnight sight by infrared image colour
The scheme of change, simple infrared hybrid optical system is difficult to the difficulty distinguished before compensate for, and enhances driver and obtains road at night
The ability of road information, it is the important reference of nighttime driving.The incidence of vehicle at night accident can greatly be reduced.
Brief description of the drawings
Fig. 1 is system framework figure provided by the invention;
Fig. 2 is the colorization algorithm flow chart of the present invention.
Embodiment
To become apparent the present invention, hereby with preferred embodiment, and accompanying drawing is coordinated to be described in detail below.
A kind of vehicle mounted infrared auxiliary drive system main purpose provided by the invention is to improve the security of running car, is led to
The infrared camera crossed on vehicle obtains the information of vehicle, condition of road surface and pedestrian's situation, and the system is finally realized
Colorization result figure provide effective information for driver.
As shown in figure 1, system architecture provided by the invention mainly includes:FLIRA615 on-vehicle night visions instrument, digital processing unit
DSP DM642, Vehicular screen, power supply etc..
Real-time image is carried out using the image capture module of FLIR A615 infrared thermographies and DSP DM642 compositions
Data are quickly uploaded to DSP DM642 by collection, the data collected by Transmission Control Protocol and twisted-pair feeder.DSP DM642 are TI public affairs
Equipment turns into a variety of in the chip that a new generation of department designs exclusively for field of video applications, its powerful computing capability and abundant piece
The first choice of video processing applications.
Image preprocessing is carried out to the infrared picture collected by DSP DM642 processing units after collection, utilizes figure
Noise on infrared image etc. is removed as the method for filtering, while strengthens the contrast of infrared image.After obtained pretreatment
It is as shown in Figure 2 using the colorization algorithm split based on image, the flow chart of algorithm.
The algorithm is firstly the need of segmentation and the colorization that background is carried out to background image.Partitioning algorithm is based on mathematical morphology
Watershed segmentation methods, it is cutting techniques that are a kind of very outstanding and being widely applied, and in essence, it belongs to
A kind of dividing method increased based on region, but it obtain be the border of target, and be continuous, closure but pixel is wide
Border.To eliminate over-segmentation caused by watershed algorithm, we remove unrelated marginal information using priori.By in picture
Appearance is divided into following three major types, and the above is sky, be below road surface, the right and left is tree.The calculating process in watershed is one and changed
For annotation process.In two steps, one is sequencer procedure to the algorithm, and one is the process of flooding.First to the ash of each pixel
Degree level is sorted from low to high, then during realization is flooded from low to high, to each local minimum in h ranks height
The domain of influence judged and marked using first in first out structure.The each several part of image background is assigned after image segments
Different colors.
Secondly algorithm is handled the prospect of infrared image.Due to pedestrian, the higher ash on image of animal equitemperature
Angle value is larger, uses the Target Recognition Algorithms based on statistics with histogram.Its general principle is the frequency of occurrences to 256 gray levels
Statistics.One initial threshold is set according to priori.With this two field picture of threshold process first, i.e., image is carried out at binaryzation
Reason.At the same time the statistics with histogram of gray scale is carried out, the maximum gray value of the frequency of occurrences is found, is assigned using it as new threshold value
Dynamic threshold variable.Can carries out binaryzation using this dynamic threshold to image afterwards, higher than the pixel quilt of threshold value
The coordinate range of pedestrian is denoted as, improves the arithmetic speed of algorithm.
New threshold value is handled as the threshold value of next frame new images, while continues to update new threshold in aforementioned manners
Value.Such circular treatment.When the advantages of this algorithm can save calculating dynamic threshold, first to all pixels point of global image
Processing, and in the binarization segmentation after statistics is put into, improve the real-time performance of system.
Then the coordinate range of pedestrian is found in the image after watershed segmentation, assigns colour, so whole is infrared
After the prospect and background of image are all combined into piece image by colorization, entire image is exported.Can be with night road dusk
Abundant road information is provided in the case of dark to driver.
Claims (2)
1. a kind of vehicle mounted infrared image colorization DAS (Driver Assistant System), including the infrared thermal imaging device being loaded on vehicle body, should
Infrared thermal imaging device is connected with the digital processing element being loaded on vehicle body, and infrared figure is gathered in real time by infrared thermal imaging device
Digital processing element is transferred to as after, the infrared image after digital processing element is handled is shown by the display unit on vehicle body
Show, it is characterised in that the infrared image received is handled as follows digital processing element:
Step 1, image preprocessing is carried out to infrared image, while the noise on removing infrared image, increase infrared image
Contrast;
Step 2, using watershed algorithm to pretreated infrared image carry out background segment, by the background segment of infrared image
Into different zones, corresponding color is assigned to corresponding region using the distinctive texture information in each region;Utilize watershed algorithm
Background segment is carried out to pretreated infrared image to concretely comprise the following steps:
Step 2.1, the unrelated marginal information using priori removal infrared image, infrared image background content are divided into following
Three regions, the above is sky areas, be below road surface region, the right and left is trees region;
Step 2.2, using watershed algorithm regional is calculated, the calculating process of water ridge algorithm is an iteration mark
Process, concretely comprise the following steps:The gray level of each pixel in each region is sorted from low to high first, then from low to high
During realization is flooded, each local minimum is judged using first in first out structure in the domain of influence of h rank height and
Mark
Step 3, the object for possessing high heat are located at the prospect of infrared image, and the prospect of infrared image is handled, using straight
Side's figure statistics assigns after the principle that dynamic threshold is combined comes out the object detection for possessing higher heat specifically has police
The color being shown as;
Step 4, background and prospect are assigned into coloured colorization infrared image shown by display unit.
2. a kind of vehicle mounted infrared image colorization DAS (Driver Assistant System) as claimed in claim 1, it is characterised in that in the step
In rapid 3, the specific step of the object detection of high heat out will be possessed with the principle that dynamic threshold is combined using statistics with histogram
Suddenly it is:
Step 3.1, according to priori set an initial threshold, by the initial threshold assign dynamic threshold variable, by first
Frame infrared image is set to present frame infrared image;
Step 3.2, with dynamic threshold variable binary conversion treatment is carried out to present frame infrared image, remembered higher than the pixel of threshold value
Make the coordinate range of pedestrian, at the same time, the statistics with histogram of gray scale is carried out to present frame infrared image, finds the frequency of occurrences most
Big gray value, dynamic threshold variable is assigned using it as new threshold value;
Step 3.3, after next frame infrared image is set into present frame infrared image, return to step 3.2 re-executes, until by institute
There is infrared image to be disposed.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510141721.9A CN104751138B (en) | 2015-03-27 | 2015-03-27 | A kind of vehicle mounted infrared image colorization DAS (Driver Assistant System) |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510141721.9A CN104751138B (en) | 2015-03-27 | 2015-03-27 | A kind of vehicle mounted infrared image colorization DAS (Driver Assistant System) |
Publications (2)
Publication Number | Publication Date |
---|---|
CN104751138A CN104751138A (en) | 2015-07-01 |
CN104751138B true CN104751138B (en) | 2018-02-23 |
Family
ID=53590795
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201510141721.9A Expired - Fee Related CN104751138B (en) | 2015-03-27 | 2015-03-27 | A kind of vehicle mounted infrared image colorization DAS (Driver Assistant System) |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN104751138B (en) |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105787906A (en) * | 2016-03-25 | 2016-07-20 | 北京环境特性研究所 | Method and system for rejecting bright noises from infrared image |
CN108566523A (en) * | 2018-05-16 | 2018-09-21 | 深圳市科纳实业有限公司 | A kind of vehicle-mounted obstacle identification optical projection system |
CN112233023A (en) * | 2020-09-27 | 2021-01-15 | 轩辕智驾科技(深圳)有限公司 | Vehicle-mounted infrared camera and dimming method and device thereof |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101286239A (en) * | 2008-04-22 | 2008-10-15 | 北京航空航天大学 | Aerial shooting traffic video frequency vehicle rapid checking method |
CN201325373Y (en) * | 2008-12-16 | 2009-10-14 | 武汉高德红外股份有限公司 | Driven infravision automotive auxiliary driving system based on image intelligent treatment |
US7672481B2 (en) * | 2004-11-30 | 2010-03-02 | Honda Motor Co., Ltd. | Image information processing system, image information processing method, image information processing program, and automobile |
CN102840853A (en) * | 2012-07-25 | 2012-12-26 | 中国航空工业集团公司洛阳电光设备研究所 | Obstacle detection and alarm method for vehicle-mounted night vision system |
CN103310006A (en) * | 2013-06-28 | 2013-09-18 | 电子科技大学 | ROI extraction method in auxiliary vehicle driving system |
-
2015
- 2015-03-27 CN CN201510141721.9A patent/CN104751138B/en not_active Expired - Fee Related
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7672481B2 (en) * | 2004-11-30 | 2010-03-02 | Honda Motor Co., Ltd. | Image information processing system, image information processing method, image information processing program, and automobile |
CN101286239A (en) * | 2008-04-22 | 2008-10-15 | 北京航空航天大学 | Aerial shooting traffic video frequency vehicle rapid checking method |
CN201325373Y (en) * | 2008-12-16 | 2009-10-14 | 武汉高德红外股份有限公司 | Driven infravision automotive auxiliary driving system based on image intelligent treatment |
CN102840853A (en) * | 2012-07-25 | 2012-12-26 | 中国航空工业集团公司洛阳电光设备研究所 | Obstacle detection and alarm method for vehicle-mounted night vision system |
CN103310006A (en) * | 2013-06-28 | 2013-09-18 | 电子科技大学 | ROI extraction method in auxiliary vehicle driving system |
Non-Patent Citations (4)
Title |
---|
《车载红外图像的行人检测与跟踪技术》;苏晓倩 等;《激光与红外》;20120831;第42卷(第8期);949-953 * |
Color Estimation for Thermal Infrared Imagery Based on Kernel PCA and Sparse Representation;SUN Shao-yuan;《Journal of Donghua University》;20121231;第29卷(第6期);475-479 * |
基于纹理特征库的微光图像色彩纹理传递;许真 等;《红外技术》;20110131;第33卷(第1期);49-55 * |
车载红外视频图像快速彩色化技术研究;范德营 等;《华东交通大学学报》;20130430;第30卷(第2期);摘要,1.1-1.4节,2.1-2.3节 * |
Also Published As
Publication number | Publication date |
---|---|
CN104751138A (en) | 2015-07-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Huang et al. | An efficient visibility enhancement algorithm for road scenes captured by intelligent transportation systems | |
CN101739827B (en) | Vehicle detecting and tracking method and device | |
TWI607901B (en) | Image inpainting system area and method using the same | |
CN107808392B (en) | Automatic tracking and positioning method and system for security check vehicle in open scene | |
CN104657735A (en) | Lane line detection method and system, as well as lane departure early warning method and system | |
CN104036466B (en) | A kind of video defogging method and system | |
CN107862290A (en) | Method for detecting lane lines and system | |
CN102938057B (en) | A kind of method for eliminating vehicle shadow and device | |
CN103871079A (en) | Vehicle tracking method based on machine learning and optical flow | |
CN101872546A (en) | Video-based method for rapidly detecting transit vehicles | |
CN109887276B (en) | Night traffic jam detection method based on fusion of foreground extraction and deep learning | |
CN104751138B (en) | A kind of vehicle mounted infrared image colorization DAS (Driver Assistant System) | |
CN103020930A (en) | Nighttime monitoring video enhancing method | |
Mandal et al. | Real-time automotive night-vision system for drivers to inhibit headlight glare of the oncoming vehicles and enhance road visibility | |
CN103400351B (en) | Low light based on KINECT depth map shines image enchancing method and system | |
Cheng et al. | Visibility enhancement of single hazy images using hybrid dark channel prior | |
CN107122732B (en) | High-robustness rapid license plate positioning method in monitoring scene | |
CN104992160B (en) | A kind of heavy truck night front vehicles detection method | |
Zhao et al. | Research of fire smoke detection algorithm based on video | |
CN111598814A (en) | Single image defogging method based on extreme scattering channel | |
CN109800693B (en) | Night vehicle detection method based on color channel mixing characteristics | |
CN116757949A (en) | Atmosphere-ocean scattering environment degradation image restoration method and system | |
Shi et al. | Image haze removal using dark channel prior and minimizing energy function | |
CN114565597B (en) | Night road pedestrian detection method based on YOLO v3-tiny-DB and transfer learning | |
CN106651939B (en) | Method based on the atmospheric scattering theory building more depth of field car lights of model inspection night traffic video |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
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: 20180223 Termination date: 20210327 |