CN108307121B - Local image mapping method and vehicle camera - Google Patents
Local image mapping method and vehicle camera Download PDFInfo
- Publication number
- CN108307121B CN108307121B CN201711475229.0A CN201711475229A CN108307121B CN 108307121 B CN108307121 B CN 108307121B CN 201711475229 A CN201711475229 A CN 201711475229A CN 108307121 B CN108307121 B CN 108307121B
- Authority
- CN
- China
- Prior art keywords
- image
- mapping
- pixel value
- local
- reference image
- 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
Links
- 238000013507 mapping Methods 0.000 title claims abstract description 62
- 238000000034 method Methods 0.000 title claims abstract description 37
- 238000001914 filtration Methods 0.000 claims abstract description 7
- 238000012545 processing Methods 0.000 claims description 18
- 238000012986 modification Methods 0.000 claims 1
- 230000004048 modification Effects 0.000 claims 1
- 102100034112 Alkyldihydroxyacetonephosphate synthase, peroxisomal Human genes 0.000 abstract description 3
- 101000799143 Homo sapiens Alkyldihydroxyacetonephosphate synthase, peroxisomal Proteins 0.000 abstract description 3
- 238000000848 angular dependent Auger electron spectroscopy Methods 0.000 abstract description 3
- 230000000694 effects Effects 0.000 abstract description 2
- 230000000007 visual effect Effects 0.000 abstract description 2
- 230000011218 segmentation Effects 0.000 description 4
- 238000012937 correction Methods 0.000 description 2
- 238000007781 pre-processing Methods 0.000 description 2
- 230000003750 conditioning effect Effects 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000002203 pretreatment Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/70—Circuitry for compensating brightness variation in the scene
- H04N23/741—Circuitry for compensating brightness variation in the scene by increasing the dynamic range of the image compared to the dynamic range of the electronic image sensors
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- Image Processing (AREA)
Abstract
The invention relates to a local image mapping method, which is characterized by comprising the following steps: acquiring a source image, filtering the image to obtain a fuzzy reference image and segmenting a local image; constructing a compensation parameter to dynamically compensate the brightness of the local image pixel; inputting the relation between the pixel value and the corresponding pixel value of the fuzzy reference image, and mapping the current pixel by constructing a mapping function to obtain an output parameter; adjusting the input pixel value by using the output parameter to complete local mapping; wherein the mapping function is a polynomial of order. Meanwhile, the vehicle camera adopting the mapping method is also disclosed. The image adjusted by the method has better effect on visual identification (such as an ADAS system), can be used for special scenes, and can be adjusted by different coefficients aiming at different scenes.
Description
Technical Field
The present invention relates to image processing, and more particularly, to a local image mapping method and a vehicle camera.
Background
The local image mapping technique is an image processing technique for adjusting the contrast of an area, and a processed image has better performance in vision or image identification, but the current local image mapping algorithm is not particularly used for processing extreme brightness or darkness. Typically, the contrast is lost after the adjustment, or the dynamic range is lost. The slave causes an image problem that exposure control is not added.
The method can not meet the requirement of a scene with high requirements on exposure control and detail reduction in the field of image processing of vehicle-mounted cameras, particularly an ADAS system based on image processing.
Disclosure of Invention
In order to solve the above technical problems, the present invention provides a local image mapping method.
A local image mapping method comprises the following steps:
s1, acquiring a source image, and filtering the image to obtain a fuzzy reference image;
s2, segmenting the original image according to the color difference of the fuzzy reference image and the size of the local kernel to obtain a local image;
s3, constructing a compensation parameter to dynamically compensate the brightness of the local image pixel to obtain an input pixel value;
s4, mapping the current pixel by constructing a mapping function according to the relation between the current input pixel value and the corresponding pixel value of the fuzzy reference image to obtain an output parameter;
s5, adjusting the input pixel value by using the output parameter to complete local mapping;
wherein the mapping function is a polynomial of order.
Further, the relationship between the current pixel value after the dynamic compensation and the corresponding pixel value of the blurred reference image is as follows:
wherein xi is the compensated current pixel value, xr is the corresponding pixel value on the fuzzy reference image, and r is the proportional variable substituted into the mapping function.
Further, the mapping function is:
a, B, C and D are functions which are always involved, and certain adjustment is carried out according to different application scenes, so that the lightness mapping inflection point and the darkness mapping inflection point are modified.
Further, the specific adjustment method of step S5 is as follows:
wherein, R is the brightness value width of the current image, and y is the pixel value output after mapping.
Further, the luminance dynamic compensation method in step S3 is as follows:
s31, calculating the corresponding average brightness value of the fuzzy reference image in the current area;
s32, determining a dynamic compensation amount according to the average brightness value;
and S33, dynamically compensating each pixel value of the current area by using the dynamic compensation amount to obtain an input pixel value.
Wherein the brightness average is inversely related to the dynamic compensation amount.
Further, the dynamic compensation amount is a constant or a function.
Further, in step S33, the following luminance correction is included before generating the input pixel value:
wherein, R is the brightness value width of the current image, and L is the brightness value.
Further, in step S1, a step of resizing the source image is further included before the filtering process is performed on the image.
Further, in step S1, the step of adjusting the size of the blurred reference image after obtaining the blurred reference image is performed, where the size of the blurred reference image is equal to the size of the source image.
Based on the local image mapping method, the invention also provides a vehicle camera which comprises a processing unit, wherein the processing unit has the processing function of the local image mapping method.
The local image mapping method of the invention solves the problem of poor image exposure control of the image, ensures that the details of the image do not lose image details under extreme conditions, has better effect on visual identification (such as an ADAS system) and can be used in special scenes, and can be adjusted by different coefficients aiming at different scenes.
Drawings
FIG. 1 is a flowchart of the pretreatment step in example 1 of the present invention.
Fig. 2 is a flowchart of local image mapping in embodiment 1 of the present invention.
Fig. 3 is a flowchart of luminance dynamic compensation in embodiment 2 of the present invention.
Detailed Description
The following detailed description of the preferred embodiments of the present invention, taken in conjunction with the accompanying drawings, will make the advantages and features of the invention easier to understand for those skilled in the art and will therefore make the scope of the invention more clearly defined.
Example 1:
the embodiment provides a local image mapping method which can be used in a plurality of different image acquisition fields and is particularly suitable for image processing by using a camera for an automobile. And carrying out local adjustment on the acquired image signals. The method comprises the steps of copying a source image into two parts after the source image is obtained, wherein one part is used for preprocessing to generate a reference image, and the other part is used for local adjustment, wherein image adjustment parameters for local adjustment are derived from the reference image.
As shown in fig. 1, the pretreatment steps include the following:
s1, firstly, obtaining a source image through an ISP unit, wherein the image format of the source image can include, but is not limited to YUV, La b, RGB and the like, it can be understood that according to different source image formats, a proper filtering mode can be selected to filter the image,
preferably, in order to improve the processing efficiency and accuracy, the source image may be resized before the blurring process is performed so as to be more suitable for the blurring process.
S2, in addition, the source image needs to be segmented to form a plurality of local areas for subsequent processing, in this embodiment, the segmentation may be performed on the source image according to the blur reference image color difference and the local kernel size to obtain a local image. The segmentation method can be performed by a convolutional neural network and the like, and each region has a relatively similar brightness value and the like.
Through the preprocessing, a fuzzy reference image and local segmentation information can be obtained, and the fuzzy reference image can be used for the brightness average value of the to-be-processed area in the subsequent processing process. The local segmentation information is used for determining the subsequent local image needing to be processed. It should be noted that, when the size of the image is changed before the filtering operation is performed, the size of the obtained blurred reference image is different from that of the source image, and therefore, the size of the blurred reference image needs to be adjusted again before the blurred reference image is output, so that the blurred reference image is equal to the size of the source image. The size in this embodiment may refer to the number of image pixels, the pixel density, and the like. Therefore, each pixel point of the source image corresponds to one pixel point of the fuzzy reference image.
After the pre-treatment is completed, the local conditioning step is started, as shown in fig. 2:
s3, because the brightness of the obtained source image is uncertain, in order to better retain details, a compensation parameter needs to be constructed to dynamically compensate the brightness of the local image pixels, and proper adjustment is carried out according to the original brightness of the local image pixels, so that a larger adjustment space is provided for distinguishing details, and finally an input pixel value is obtained. It is understood that the pixel value refers to all pixel values in the local image, and for convenience of the following description, each pixel value is represented by xiAnd (4) showing.
S4, preferably determining a relationship between a pixel value currently processed in the local image and a pixel value corresponding to the blurred reference image, where in this embodiment, the relationship between the two may be related by the following formula:
wherein xi is the compensated current pixel value, xr is the corresponding pixel value on the fuzzy reference image, and r is the proportional variable substituted into the mapping function.
After the relationship between the current pixel and the current pixel is determined, mapping is carried out on the current pixel by constructing a mapping function, wherein the mapping function is a polynomial of a order of several, specifically a polynomial of a 3-order:
a, B, C and D are functions which are always involved, and certain adjustment is carried out according to different application scenes, so that the lightness mapping inflection point and the darkness mapping inflection point are modified. A plurality of function constant parameters can be set through a plurality of tests, so that the mapping function has an optimal function constant parameter correspondence under various brightness conditions. Thus forming a parameter library, and calling as required during mapping.
It should be noted that, in the mapping process, the parameters of the mapping function itself are adjusted so that the mapping function is adjusted
The output value is a value between 0 and 1, and the value output in this step is an output parameter. The output parameters will participate in the final adjustment of the pixel values of the source image.
S5, adjusting the input pixel value by using the output parameter to complete local mapping; the specific adjustment method adopts the following formula:
wherein, R is the brightness value width of the current image, and y is the pixel value output after mapping. And performing the above-mentioned operation mapping on all pixel values in the local image to be adjusted, thereby completing the adjustment of the whole image.
It should be noted that the value of R is based on the bit width of the current image, and when the color bit width is 8 bits, R takes 256. By analogy, a value when the color bit width is 12 bits or even higher can be obtained, and details will not be described in this embodiment.
Based on the local image mapping method, the invention also provides a vehicle camera which comprises a processing unit, wherein the processing unit has the processing function of the local image mapping method.
Example 2:
the present embodiment is further optimized based on embodiment 1, and as shown in fig. 3, in the present embodiment, the luminance dynamic compensation method in step S3 is as follows:
s31, calculating the corresponding average brightness value of the fuzzy reference image in the current area;
s32, determining a dynamic compensation amount according to the average brightness value, wherein the dynamic compensation amount can be a constant or a function for adjusting the brightness setting to be corrected under different scenesAnd (4) determining. Such as night, day or fog, snow, etc. The brightness average value is inversely related to the dynamic compensation amount. When the luminance is large, compensation may not be necessary, and when the luminance is low, appropriate compensation is performed. For convenience of description, use
And (4) showing.
S33, dynamically compensating each pixel value of the current region by using the dynamic compensation amount, which can be represented by the following formula:
s34, in order to prevent data overflow, the following luminance correction is also included before generating the input pixel value:
wherein, R is the brightness value width of the current image, and L is the brightness value.
Optimally, according to image information in different formats, the compensated brightness value is combined into the pixel value to obtain an input pixel value for subsequent mapping operation.
The embodiments of the present invention have been described in detail with reference to the drawings, but the present invention is not limited to the above embodiments, and various changes can be made within the knowledge of those skilled in the art without departing from the gist of the present invention.
Claims (8)
1. A local image mapping method is characterized by comprising the following steps:
s1, acquiring a source image, and filtering the image to obtain a fuzzy reference image;
s2, segmenting the original image according to the color difference of the fuzzy reference image and the size of the local kernel to obtain a local image;
s31, calculating the corresponding average brightness value of the fuzzy reference image in the current area;
s32, determining a dynamic compensation amount according to the average brightness value, wherein the brightness average value is inversely related to the dynamic compensation amount;
s33, dynamically compensating each pixel value of the current area by using the dynamic compensation amount to obtain an input pixel value;
s4, mapping the input pixel value by constructing a mapping function according to the proportional relation between the input pixel value and the corresponding pixel value of the fuzzy reference image to obtain an output parameter, wherein the mapping function is a polynomial of a order;
s5, adjusting the input pixel value by using the output parameter, completing local mapping by using the following algorithm,
wherein,
for the input pixel value, f (R) is the mapping function, R is the brightness value width of the current image, and y is the pixel value output after mapping.
2. The local image mapping method according to claim 1, wherein the relationship between the input pixel value and the corresponding pixel value of the blurred reference image is as follows:
wherein xi is an input pixel value, xr is a corresponding pixel value on the blurred reference image, and r is a proportional variable substituted into the mapping function.
3. The local image mapping method according to claim 1 or 2, wherein the mapping function is:
a, B, C and D are functions which are always involved, and certain adjustment is carried out according to different application scenes, so that the lightness mapping inflection point and the darkness mapping inflection point are modified.
4. The local image mapping method according to claim 1, wherein the dynamic compensation amount is a constant or a function.
5. The local image mapping method as claimed in claim 1, wherein the step S33 further comprises the following luminance modification before generating the input pixel value:
wherein, R is the brightness value width of the current image, and L is the brightness value.
6. The local image mapping method as claimed in claim 1, wherein the step S1 further includes a step of resizing the source image before the filtering process is performed on the image.
7. The local image mapping method as claimed in claim 1, wherein the step S1, after obtaining the blurred reference image, further comprises adjusting the size of the blurred reference image, wherein the blurred reference image has the same size as the source image.
8. A vehicle camera is characterized by comprising a processing unit, wherein the processing unit has the processing function of the local image mapping method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711475229.0A CN108307121B (en) | 2017-12-29 | 2017-12-29 | Local image mapping method and vehicle camera |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711475229.0A CN108307121B (en) | 2017-12-29 | 2017-12-29 | Local image mapping method and vehicle camera |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108307121A CN108307121A (en) | 2018-07-20 |
| CN108307121B true CN108307121B (en) | 2020-11-13 |
Family
ID=62867786
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711475229.0A Active CN108307121B (en) | 2017-12-29 | 2017-12-29 | Local image mapping method and vehicle camera |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108307121B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115001689A (en) * | 2022-05-24 | 2022-09-02 | 中国科学技术大学 | High-embedding-rate reversible data hiding method and system capable of resisting internal attack |
| CN117935414B (en) * | 2024-01-23 | 2024-10-29 | 湖北华中电力科技开发有限责任公司 | Traffic strategy analysis system based on image content big data identification |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101853497A (en) * | 2010-02-25 | 2010-10-06 | 杭州海康威视软件有限公司 | Image enhancement method and device |
| CN102158670A (en) * | 2010-02-11 | 2011-08-17 | 富士通微电子(上海)有限公司 | Digital video image contrast adaptive-stretching method and system |
| CN103024300A (en) * | 2012-12-25 | 2013-04-03 | 华为技术有限公司 | Device and method for high dynamic range image display |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8103120B2 (en) * | 2008-09-22 | 2012-01-24 | Solomon Systech Limited | Method and apparatus of local contrast enhancement |
-
2017
- 2017-12-29 CN CN201711475229.0A patent/CN108307121B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102158670A (en) * | 2010-02-11 | 2011-08-17 | 富士通微电子(上海)有限公司 | Digital video image contrast adaptive-stretching method and system |
| CN101853497A (en) * | 2010-02-25 | 2010-10-06 | 杭州海康威视软件有限公司 | Image enhancement method and device |
| CN103024300A (en) * | 2012-12-25 | 2013-04-03 | 华为技术有限公司 | Device and method for high dynamic range image display |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108307121A (en) | 2018-07-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111986129B (en) | HDR image generation method, equipment and storage medium based on multi-shot image fusion | |
| CN104717432B (en) | Handle method, image processing equipment and the digital camera of one group of input picture | |
| EP3156968B1 (en) | Haze removal device and image generation method | |
| CN112734650A (en) | Virtual multi-exposure fusion based uneven illumination image enhancement method | |
| US20140092221A1 (en) | Image processing apparatus and method, and related computer program | |
| US9307162B2 (en) | Local enhancement apparatus and method to generate high dynamic range images by blending brightness-preserved and brightness-adjusted blocks | |
| CN108133462B (en) | A Single Image Restoration Method Based on Gradient Field Segmentation | |
| WO2015141050A1 (en) | Multi-area white balance control device, multi-area white balance control method, multi-area white balance control program, computer on which multi-area white balance control program has been recorded, multi-area white balance image processing device, multi-area white balance image processing method, multi-area white balance image processing program, computer on which multi-area white balance image processing program has been recorded, and imaging device provided with multi-area white balance image processing device | |
| EP4101161B1 (en) | Multi-frame depth-based multi-camera relighting of images | |
| CN106131526A (en) | A kind of white balancing treatment method based on rgb space and device | |
| CN105245785A (en) | Brightness balance adjustment method of vehicle panoramic camera | |
| US20190052860A1 (en) | Multi-Image Color-refinement with Application to Disparity Estimation | |
| CN113592739B (en) | Lens shading correction method, device and storage medium | |
| CN107169973A (en) | The background removal and synthetic method and device of a kind of image | |
| CN108307121B (en) | Local image mapping method and vehicle camera | |
| CN119624842B (en) | A method, device and storage medium for image brightness correction | |
| CN117808699A (en) | Real world single image defogging method based on sky region segmentation and image fusion | |
| CN112991163B (en) | Panoramic image acquisition method, device and equipment | |
| CN110086997B (en) | Face image exposure brightness compensation method and device | |
| CN116957948A (en) | Image processing method, electronic product and storage medium | |
| CN111028184B (en) | Image enhancement method and system | |
| CN111968188A (en) | Low-illumination image enhancement processing method, system, device and storage medium | |
| CN110070508B (en) | Unsharp enhancement method based on threshold value and linear mapping | |
| Jin et al. | Image dehazing using non-local haze-lines and multi-exposure fusion | |
| JP5952574B2 (en) | Image processing apparatus and control method thereof |
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 |