CN109785251B - Image quality improving method and device and automatic driving system - Google Patents

Image quality improving method and device and automatic driving system Download PDF

Info

Publication number
CN109785251B
CN109785251B CN201811589078.6A CN201811589078A CN109785251B CN 109785251 B CN109785251 B CN 109785251B CN 201811589078 A CN201811589078 A CN 201811589078A CN 109785251 B CN109785251 B CN 109785251B
Authority
CN
China
Prior art keywords
contrast
roi
target image
brightness
value
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.)
Active
Application number
CN201811589078.6A
Other languages
Chinese (zh)
Other versions
CN109785251A (en
Inventor
郝源
姜安
崔峰
朱海涛
刘永才
孙钊
苏文秀
肖志鹏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Beijing Smarter Eye Technology Co Ltd
Original Assignee
Beijing Smarter Eye Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Beijing Smarter Eye Technology Co Ltd filed Critical Beijing Smarter Eye Technology Co Ltd
Priority to CN201811589078.6A priority Critical patent/CN109785251B/en
Publication of CN109785251A publication Critical patent/CN109785251A/en
Application granted granted Critical
Publication of CN109785251B publication Critical patent/CN109785251B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Landscapes

  • Image Analysis (AREA)

Abstract

The invention provides an image quality improving method and device and an automatic driving system, which are applied to an imaging system. The image quality improving method comprises the following steps: dividing the ROI of the target image into a preset number of regions, and performing contrast detection probability analysis on each region to obtain an analysis result; and comparing the analysis result with a preset contrast threshold interval to obtain a first comparison result, judging the first comparison result, and performing tone adjustment on the target image when the first comparison result exceeds the preset contrast threshold interval so as to improve the quality of the target image. The invention detects the contrast of the image through CDP, and improves the image quality by utilizing tone mapping, thereby solving the problem that the contrast is influenced by external sunlight glare and other factors.

Description

Image quality improving method and device and automatic driving system
Technical Field
The invention relates to the field of image imaging, in particular to an image quality improving method and device and an automatic driving system.
Background
In recent years, automobile driving assistance systems have been increasingly demanded in the market, and many of them have been standardized for automobiles from the beginning of ABS anti-lock systems, to systems for automatically switching lights, to automatic wiper systems for sensing rain, and to backing systems with backing images and ultrasonic radar assistance. Especially, the camera is widely applied to the field of automobile auxiliary driving and the field of automatic driving at present, and becomes the extension of human eyes. The front images of the automobile can be used for identifying automobile signal lamps, pedestrians, obstacles, lane lines and the like, and are an indispensable part in the fields of automobile auxiliary systems and automatic driving in the future. As the binocular ADAS camera has both monocular identification and recording functions and accurate depth ranging function, the binocular ADAS camera is widely applied to high-end automobiles and has wider market prospect compared with the monocular camera widely applied at present.
With the application of the binocular camera to the fields of automobile assistance and automatic driving, the requirement on the image effect of the binocular camera is higher and higher, but if the binocular camera is in foggy weather, heavy rain, dust in the surrounding environment of a windshield is large, and the glare effect of external sunlight influences the image effect, so that the detection, the depth measurement and the three-dimensional matching of a target object are influenced. As shown in fig. 1, the entire image is grayed out due to the windshield dust and the sunlight glare effect, but due to the high sensitivity of human eyes to the contrast, it is possible to find the object in front, but it may be difficult for the obstacle recognition algorithm. Fig. 2 shows the sudden appearance of an object after no glare. If the algorithm does not detect the obstacle in time, the practicability of the camera for detecting the obstacle is greatly influenced.
In view of this, the present invention is proposed.
Disclosure of Invention
The invention provides an image quality improving method and device and an automatic driving system, which are used for solving the problem that the image effect in the prior art cannot meet the requirement of the automatic driving system on obstacle detection and identification.
In order to achieve the above object, according to an aspect of the present invention, an image quality improving method is provided, and the following technical solutions are adopted:
an image quality improving method is applied to an imaging system and comprises the following steps: dividing the ROI of the target image into a preset number of regions, and performing contrast detection probability analysis on each region to obtain an analysis result; and comparing the analysis result with a preset contrast threshold interval to obtain a first comparison result, judging the first comparison result, and performing tone adjustment on the target image when the first comparison result exceeds the preset contrast threshold interval so as to improve the quality of the target image.
According to another aspect of the present invention, an image quality improving apparatus is provided, and the following technical solutions are adopted:
an image quality improvement device comprising: the analysis module is used for dividing the ROI of the target image into a preset number of regions and carrying out contrast detection probability analysis on each region to obtain an analysis result; and the adjusting module is used for comparing the analysis result with a preset contrast threshold interval to obtain a first comparison result, judging the first comparison result, and adjusting the tone of the target image when the first comparison result exceeds the preset contrast threshold interval so as to improve the quality of the target image.
According to another aspect of the invention, an automatic driving system is provided, and the following technical scheme is adopted:
the automatic driving system comprises the lifting device.
The invention improves the image quality of the binocular ADAS camera through CDP and tone mapping, and obtains good effect. Meanwhile, a certain objective basis is found for the image contrast and is consistent with the feeling of people. The camera has certain relieving and improving effects on the condition of contrast reduction caused by glare and other environments, and also provides powerful support for the camera to be applied to obstacle recognition, lane line recognition, pedestrian recognition and the like.
Drawings
In order to more clearly illustrate the embodiments of the present application or technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments described in the present invention, and other drawings can be obtained by those skilled in the art according to the drawings.
FIG. 1 is a schematic view of a windshield dust and glare influencing scene according to the background of the invention;
FIG. 2 is a schematic view of the same scene as FIG. 1 but showing the glare disappearance as described in the background of the invention;
FIG. 3 is a flowchart of an image quality improvement method according to an embodiment of the invention;
FIG. 4 is a schematic diagram illustrating CDP statistics performed on a target image ROI according to an embodiment of the present invention;
FIG. 5 is a schematic diagram illustrating a comparison between before and after processing of a target image according to an embodiment of the present invention;
fig. 6 is a schematic structural diagram of an image quality improving apparatus according to an embodiment of the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Fig. 3 shows a flowchart of an image quality improving method according to an embodiment of the invention.
Referring to fig. 3, an image quality improving method includes:
s101: an image quality improving method is applied to an imaging system and comprises the following steps: dividing the ROI of the target image into a preset number of regions, and performing contrast detection probability analysis on each region to obtain an analysis result;
s103: and comparing the analysis result with a preset contrast threshold interval to obtain a first comparison result, judging the first comparison result, and performing tone adjustment on the target image when the first comparison result exceeds the preset contrast threshold interval so as to improve the quality of the target image.
Specifically, in step S101, an ROI of the image is selected, the ROI is divided into N × M regions, and CDP, i.e., contrast detection probability, is performed on each region, as shown in fig. 4. In step S103, the analysis result is compared with a preset contrast threshold interval to obtain a first comparison result. I.e., the CDP of each region is compared to a preset contrast threshold interval, as in equation 14.
Figure BDA0001919811210000031
Wherein the content of the first and second substances,
Figure BDA0001919811210000032
is the maximum threshold value for the CDP,
Figure BDA0001919811210000033
is the minimum threshold for CDP.
It can therefore be seen that the above comparison result, i.e. the first comparison result, has two cases, being within the preset contrast threshold interval and exceeding the preset contrast threshold interval. And regarding the preset contrast threshold interval, the contrast requirement is considered to be met. And for the image which exceeds the preset contrast threshold interval, carrying out tone adjustment on the target image so as to improve the quality of the target image.
Furthermore, when the comparison result does not exceed the preset contrast threshold interval, the method for increasing further includes: calculating the average brightness of the region to obtain the average brightness; and comparing the average brightness with a preset brightness threshold value to obtain a second comparison result, and performing tone adjustment on the target image when the second comparison result exceeds the preset brightness threshold value, so that the quality of the target image is improved.
And further analyzing and comparing the brightness of the interval meeting the contrast threshold, and adjusting the color tone of the target image when the average brightness exceeds a preset brightness threshold, so that the quality of the target image is improved.
Suppose that the ROI area is divided into N rows and M columns of sub-ROIs, each sub-ROI area having an average brightness LSub-ROISatisfy the requirement of
Figure BDA0001919811210000042
Wherein 1 is not less than i not more than (N M), LMaxIs the average brightness maximum threshold, LMinIs the average luminance minimum threshold.
The image contrast is considered to be satisfactory and can be used for object recognition. Adjustments are required when conditions are exceeded.
Preferably, the performing a contrast detection probability analysis on each region to obtain an analysis result includes: acquiring the contrast of the target image, and defining a random variable and a confidence interval of the target image according to the contrast; and obtaining the contrast detection probability of the target image according to the random variable and the confidence interval.
Specifically, the acquiring the contrast of the target image includes: based on the following weber contrast:
KWeber＝KW＝Emax/Emin -1wherein K is more than or equal to 0Weber≤∞ (1)
Wherein KWeber、KWIs a Weber contrast value, EmaxTo the maximum brightness value, EminIs the minimum brightness value;
defining a relative luminance difference p:
Emax＝(1+p)Emin (2)
Figure BDA0001919811210000041
the brightness difference percentage p is obtainedWeber；
The definition of michelson contrast in the optical domain is:
Figure BDA0001919811210000051
wherein K is more than or equal to 0Michelson≤1 (4)
Converting the Michelson contrast into a Weber contrast to obtain a Weber contrast value KW：
Figure BDA0001919811210000052
Figure BDA0001919811210000053
Figure BDA0001919811210000054
Figure BDA0001919811210000055
Wherein KMichelson、KMIs the value of the Michelson contrast, KWIs a Weber contrast value, EmaxTo the maximum brightness value, EminIs the minimum luminance value.
Preferably, the obtaining the contrast detection probability of the target image according to the random variable and the confidence interval includes:
calculating the contrast detection probability by a probability function and the contrast
Figure BDA0001919811210000056
The calculation formula is as follows:
Figure BDA0001919811210000057
wherein KmeasIs a random variable, epsilon is a confidence interval, the probability function is Prob (-) and the contrast is Kin。
Preferably, the tone adjustment of the target image includes: calculating the overall brightness of the region to be adjusted, and performing weight matching on the ROI region of the region to be adjusted and other regions; and mapping each pixel point in the region needing to be adjusted according to the average brightness based on the matching result.
Specifically, after CDP analysis, for a target image, for which tone mapping is required, the overall image brightness is estimated, generally, a log-average method is used for brightness estimation, but here, a certain weight ratio is performed on an ROI region and other regions:
Figure BDA0001919811210000061
wherein
Figure BDA0001919811210000062
For estimated overall luminance, WeightRoi，Weightnon-RoiFor the weight value, the general ROI weight is larger than the non-ROI weight.
WeightRoi+Weightnon-Roi＝1 (11)
NROI，Nnon-ROINumber of pixels, δ, for Roi and non-Roi areasRoi，δnon-RoiIs a very small number, corresponding to the situation where the pixel is purely black.
Preferably, the mapping each pixel point in the region needing to be adjusted according to the average brightness includes:
and de-mapping the brightness value of each pixel point according to the average brightness:
Lwhiteis pure white brightness threshold, exceeds LwhiteAll values of (A) are pure white
Figure BDA0001919811210000063
Wherein L isw(x, y) is the original luminance of each pixel, α is the luminance tendency, and L (x, y) is the final per-pixel mapping result L of the luminance calculationd(x, y) is:
Figure BDA0001919811210000064
the invention improves the image quality of the binocular ADAS camera through CDP and tone mapping, and obtains good effect. Meanwhile, a certain objective basis is found for the image contrast and is consistent with the feeling of people. The camera has certain relieving and improving effects on the condition of contrast reduction caused by glare and other environments, and also provides powerful support for the camera to be applied to obstacle recognition, lane line recognition, pedestrian recognition and the like. The results of the experiment are shown in FIG. 5, where the images before and after treatment are seen to be compared.
Fig. 6 is a schematic structural diagram of an image quality improving apparatus according to an embodiment of the present invention.
Referring to fig. 6, an image quality improving apparatus includes: and the analysis module 50 is configured to divide the ROI of the target image into a preset number of regions, and perform contrast detection probability analysis on each region to obtain an analysis result. The adjusting module 52 is configured to compare the analysis result with a preset contrast threshold interval to obtain a first comparison result, determine the first comparison result, and perform color tone adjustment on the target image when the first comparison result exceeds the preset contrast threshold interval, so that the quality of the target image is improved.
The invention provides an automatic driving system which comprises the lifting device.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present invention, and all the changes or substitutions should be covered within the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims (6)

1. An image quality improvement method applied to an imaging system is characterized by comprising the following steps:
dividing the ROI of the target image into a preset number of regions, and performing contrast detection probability analysis on each region to obtain an analysis result;
comparing the analysis result with a preset contrast threshold interval to obtain a first comparison result, judging the first comparison result, and performing tone adjustment on the target image when the first comparison result exceeds the preset contrast threshold interval so as to improve the quality of the target image;
wherein, the performing contrast detection probability analysis on each region to obtain an analysis result comprises:
acquiring the contrast of the target image, and defining a random variable and a confidence interval of the target image according to the contrast;
acquiring the contrast detection probability of the target image according to the random variable and the confidence interval;
wherein the obtaining the contrast of the target image comprises:
based on the following weber contrast:
KWeber＝KW＝Emax/Emin-1 wherein 0. ltoreq. KWeber≤∞ (1)
Wherein KWeber、KWIs a Weber contrast value, EmaxTo the maximum brightness value, EminIs the minimum brightness value;
defining a relative luminance difference p:
Emax＝(1+p)Emin (2)
Figure FDA0002757881390000011
the brightness difference percentage p is obtainedWeber；
The definition of michelson contrast in the optical domain is:
Figure FDA0002757881390000012
converting the Michelson contrast into a Weber contrast to obtain a Weber contrast value KW：
Figure FDA0002757881390000013
Figure FDA0002757881390000021
Figure FDA0002757881390000022
Figure FDA0002757881390000023
Wherein KMichelson、KMIs the value of the Michelson contrast, KWIs a Weber contrast value, EmaxTo the maximum brightness value, EminIs the minimum luminance value.
2. The lifting method of claim 1, wherein when the comparison result does not exceed the preset contrast threshold interval, the lifting method further comprises:
calculating the average brightness of the region to obtain the average brightness;
and comparing the average brightness with a preset brightness threshold value to obtain a second comparison result, and performing tone adjustment on the target image when the second comparison result exceeds the preset brightness threshold value, so that the quality of the target image is improved.
3. The method of claim 1, wherein the obtaining the contrast detection probability of the target image according to the random variable and the confidence interval comprises:
calculating the contrast detection probability by a probability function and the contrast
Figure FDA0002757881390000024
The calculation formula is as follows:
Figure FDA0002757881390000025
wherein KmeasIs a random variable, epsilon is a confidence interval, outlineThe rate function is Prob (-) and the contrast is Kin。
4. A lifting method as recited in claim 2, wherein the performing a tone adjustment on the target image comprises:
calculating the overall brightness of the region to be adjusted, and performing weight matching on the ROI region of the region to be adjusted and other regions;
and mapping each pixel point in the region needing to be adjusted according to the average brightness based on the matching result.
5. The method as claimed in claim 4, wherein the method for weighting and matching the ROI area of the area to be adjusted with other areas comprises:
Figure FDA0002757881390000031
wherein
Figure FDA0002757881390000032
As a whole brightness, WeightRoi，Weightnon-RoiIs a weight value of the weight value,
WeightRoi+Weightnon-Roi＝1 (11)
NROI，Nnon-ROInumber of pixels, δ, for Roi and non-Roi areasRoi，δnon-RoiIs a very small number, corresponding to the situation where the pixel is purely black.
6. The improvement method of claim 4, wherein said mapping each pixel in the region requiring adjustment according to the average brightness comprises:
and de-mapping the brightness value of each pixel point according to the average brightness:
Lwhiteis pure white brightness threshold, exceeds LwhiteAll values of (A) are pure white
Figure FDA0002757881390000033
Wherein L isw(x, y) is the original luminance of each pixel, a is the luminance trend, and L (x, y) is the calculated luminance, where
Figure FDA0002757881390000034
For overall brightness, the final per-pixel mapping result Ld(x, y) is:
Figure FDA0002757881390000035
CN201811589078.6A 2018-12-25 2018-12-25 Image quality improving method and device and automatic driving system Active CN109785251B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811589078.6A CN109785251B (en) 2018-12-25 2018-12-25 Image quality improving method and device and automatic driving system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811589078.6A CN109785251B (en) 2018-12-25 2018-12-25 Image quality improving method and device and automatic driving system

Publications (2)

Publication Number Publication Date
CN109785251A CN109785251A (en) 2019-05-21
CN109785251B true CN109785251B (en) 2021-02-12

Family

ID=66497631

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811589078.6A Active CN109785251B (en) 2018-12-25 2018-12-25 Image quality improving method and device and automatic driving system

Country Status (1)

Country Link
CN (1) CN109785251B (en)

Families Citing this family (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112561906A (en) * 2020-12-24 2021-03-26 百果园技术(新加坡)有限公司 Image processing method, device, equipment and medium
GB2621820A (en) * 2022-08-16 2024-02-28 Advanced Risc Mach Ltd Method, apparatus and program for processing an image

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105578063B (en) * 2015-07-14 2018-04-10 宇龙计算机通信科技(深圳)有限公司 A kind of image processing method and terminal
CN106503707B (en) * 2016-10-21 2019-08-02 浙江宇视科技有限公司 Licence plate recognition method and device under a kind of infrared lighting condition

Also Published As

Publication number Publication date
CN109785251A (en) 2019-05-21

Similar Documents

Publication Publication Date Title
KR101204556B1 (en) Method for removing image noise and night-vision system using the same
TWI607901B (en) Image inpainting system area and method using the same
US10163195B2 (en) Spatio-temporal differential synthesis ofdetail images for high dynamic range imaging
JP2019096320A (en) Determination of clear or dirty captured image
US9224052B2 (en) Method for in-image periodic noise pixel inpainting
JP2000039306A (en) Device and method for vehicle system area detection
US9478016B2 (en) Image restoration method and image processing apparatus using the same
CN109785251B (en) Image quality improving method and device and automatic driving system
CN111860120A (en) Automatic shielding detection method and device for vehicle-mounted camera
US10129458B2 (en) Method and system for dynamically adjusting parameters of camera settings for image enhancement
Choi et al. Fog detection for de-fogging of road driving images
JPH09282452A (en) Monitor
KR101522757B1 (en) Method for removing noise of image
KR101146417B1 (en) Apparatus and method for tracking salient human face in robot surveillance
US20210089818A1 (en) Deposit detection device and deposit detection method
CN105447827A (en) Image noise reduction method and system thereof
KR100801989B1 (en) Recognition system for registration number plate and pre-processor and method therefor
US11568547B2 (en) Deposit detection device and deposit detection method
CN108259819B (en) Dynamic image feature enhancement method and system
CN106651939B (en) Method based on the atmospheric scattering theory building more depth of field car lights of model inspection night traffic video
KR20150011097A (en) Driver assistance system and method thereof
US11393128B2 (en) Adhered substance detection apparatus
CN115240170A (en) Road pedestrian detection and tracking method and system based on event camera
Mittal et al. IoT based image defogging system for road accident control during winters
TW201824864A (en) Dynamic image feature enhancement method and system capable of directly and instantly adjusting imaging parameters of the camera and improving parameters of the captured image

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant