CN112581404A - Rail transit video monitoring image enhancement algorithm - Google Patents
Rail transit video monitoring image enhancement algorithm Download PDFInfo
- Publication number
- CN112581404A CN112581404A CN202011566292.7A CN202011566292A CN112581404A CN 112581404 A CN112581404 A CN 112581404A CN 202011566292 A CN202011566292 A CN 202011566292A CN 112581404 A CN112581404 A CN 112581404A
- Authority
- CN
- China
- Prior art keywords
- image
- frequency
- low
- frequency component
- enhanced
- 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.)
- Pending
Links
- 238000012544 monitoring process Methods 0.000 title claims abstract description 16
- 230000002708 enhancing effect Effects 0.000 claims abstract description 17
- 238000000034 method Methods 0.000 claims abstract description 16
- 238000012937 correction Methods 0.000 claims abstract description 9
- 238000012545 processing Methods 0.000 claims description 10
- 230000014509 gene expression Effects 0.000 claims description 9
- 238000000605 extraction Methods 0.000 claims description 6
- 230000001131 transforming effect Effects 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 5
- 238000001228 spectrum Methods 0.000 claims description 3
- 238000000926 separation method Methods 0.000 abstract description 2
- 230000000694 effects Effects 0.000 description 8
- 238000010586 diagram Methods 0.000 description 3
- 238000005286 illumination Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000004445 quantitative analysis Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/80—Geometric correction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/50—Image enhancement or restoration using two or more images, e.g. averaging or subtraction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/90—Dynamic range modification of images or parts thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10016—Video; Image sequence
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20048—Transform domain processing
- G06T2207/20056—Discrete and fast Fourier transform, [DFT, FFT]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a track traffic video monitoring image enhancement algorithm, which comprises the following steps: s1, dividing an image into two color blocks and two edge components, and respectively corresponding to the frequency domain, wherein the color blocks correspond to the low-frequency component of the image, and the edge details correspond to the high-frequency component of the image; s2, enhancing the low-frequency component by using global self-adaptive gamma correction, and enhancing the high-frequency component by using Fourier transform; and S3, overlapping the enhancement maps corresponding to the low-frequency component and the high-frequency component to obtain an output image. The invention belongs to a brightness adjustment algorithm with high and low frequency separation, and better results can be obtained by respectively enhancing low-frequency images and high-frequency images by adopting different suitable methods.
Description
Technical Field
The invention relates to the technical field of image enhancement processing, in particular to a track traffic video monitoring image enhancement algorithm.
Background
The rail transit video monitoring system can facilitate the on-duty personnel to monitor the subway operation condition, the passenger flow and the conditions of passengers getting on and off the train in real time, facilitate the maintenance personnel to monitor the equipment condition in time, and ensure that the management personnel effectively control and command the field condition. The frame of the rail transit video system often has the low illumination condition, which directly leads to poor quality of the collected image, and greatly reduces the practical value of the video monitoring system. Therefore, exploring the image enhancement algorithm and application of the rail transit video monitoring system has great practical significance for fully exerting the use value of the video system.
For low-light images in rail transit video surveillance, the main purpose is to adjust their brightness in order to view the target. However, the existing enhancement algorithms such as BBHE and DHE have low fidelity to details, often lose original details while enhancing contrast, and have poor enhancement effect.
Disclosure of Invention
In order to improve the enhancement effect of low-illumination images, the patent provides a method for enhancing track traffic video monitoring images, and the specific scheme is as follows:
the rail transit video monitoring image enhancement algorithm comprises the following steps:
s1, dividing an image into two color blocks and two edge components, and respectively corresponding to the frequency domain, wherein the color blocks correspond to the low-frequency component of the image, and the edge details correspond to the high-frequency component of the image;
s2, enhancing the low-frequency component by using global self-adaptive gamma correction, and enhancing the high-frequency component by using Fourier transform;
and S3, overlapping the enhancement maps corresponding to the low-frequency component and the high-frequency component to obtain an output image.
Specifically, step S1 specifically includes: acquiring low-frequency components of the image by a Gaussian low-pass filter, and recording the obtained low-frequency image as Z; in the high-frequency component extraction process, a manner of retaining all pixels instead of direct subtraction is adopted.
Specifically, the high-frequency component extraction process in step S1 is:
respectively judging the pixel value a of the original image in R, G, B three channelscAnd a low-frequency component pixel value bcThe size of the image is c ∈ { R, G, B }, if the original image pixel is larger than the low-frequency image pixel, the original image pixel is directly subtracted, and the obtained difference value is stored in a high-frequency image X; if the original image pixel is smaller than the low-frequency component pixel, taking the opposite number of the difference value and storing the opposite number into a high-frequency image Y; the mathematical expressions of the images calculated in the R, G, B channels are as follows:
X＝ac-bc，ac＞bc，c∈{R，G，B}
Y＝bc-ac，ac＜bc，c∈{R，G，B}。
specifically, the step of enhancing the low-frequency component in step S2 specifically includes: the method is realized by a gamma correction power function, and the mathematical expression of the power function is as follows:
L＝Zγ
where L is the output enhanced low-frequency image, Z is the input low-frequency original, and γ is the power exponent.
Specifically, γ takes a value of 2.2.
Specifically, the enhancement of the high-frequency component in step S2 is specifically:
s21, transforming the image from a space domain to a frequency domain through Fourier transform;
s22, homomorphic filtering operation and processing are carried out on the frequency spectrum in the frequency domain;
s23, inverse transforming the image into a spatial domain to obtain an enhanced image, wherein the enhanced image of the high-frequency image X is P and the enhanced image of the high-frequency image Y is Q.
Specifically, step S3 is as follows: and (3) superposing the enhanced image and the low-frequency enhanced image corresponding to the high-frequency image X and the high-frequency image Y, wherein the pixel value stored by q is a negative value, and the superposition needs to be subjected to inverse number processing, and the mathematical expression of the method is as follows:
J＝Pc+Lc+Qc，c∈{R，G，B}
wherein J is the superimposed image, PcFor the X enhanced image, LcFor low-frequency images Z enhanced images, QcIs the image after Y enhancement.
The invention has the beneficial effects that:
(1) the invention belongs to a brightness adjustment algorithm with high and low frequency separation, and better results can be obtained by respectively enhancing low-frequency images and high-frequency images by adopting different suitable methods.
(2) The low frequency components of an image are the color block portions of the image, i.e., most of the pixels of the image, which determine the brightness of the image to some extent. The low-frequency components are enhanced by adopting a gamma correction method, wherein the gamma correction refers to a mode of editing a gamma curve of an image to perform nonlinear brightness adjustment on the image, and essentially, a dark color pixel and a light color pixel in the image are separated by a gray level transformation equation, and the difference between the two parts is enlarged, so that the contrast of the image is enhanced.
(3) The common algorithm for enhancing high-frequency images is guide map filtering, which requires separately processing steps for R, G, B three channels, and the enhancing effect is not significant. The texture characteristics of the image after the guide map filtering can be maintained, but the detail highlighting cannot be realized. According to the method, the Fourier transform mode is adopted, so that the steps are simple, and the details can be highlighted.
Drawings
Fig. 1 is a flowchart of an image enhancement algorithm for rail transit video surveillance provided by the present invention.
Fig. 2 shows the original image of the experimental treatment.
Fig. 3 is a diagram illustrating the effect of the BBHE algorithm process in the prior art.
FIG. 4 is a diagram of the effect of DHE algorithm processing in the prior art.
Fig. 5 is a diagram illustrating the effect of the algorithm process of the present application.
Detailed Description
Referring to fig. 1, the invention provides an image enhancement algorithm for rail transit video surveillance, which comprises the following steps:
s1, dividing an image into two color blocks and two edge components, and respectively corresponding to the frequency domain, wherein the color blocks correspond to the low-frequency component of the image, and the edge details correspond to the high-frequency component of the image; specifically, the low-frequency component of the image is obtained by a Gaussian low-pass filter, and the obtained low-frequency image is recorded as Z; in the high-frequency component extraction process, a manner of retaining all pixels instead of direct subtraction is adopted. The original image is a live video slice collected by a track traffic video monitoring system in a certain city randomly extracted in video monitoring, and the picture size is 702 × 525, as shown in fig. 2.
The high-frequency component extraction process comprises the following steps:
respectively judging the pixel value a of the original image in R, G, B three channelscAnd a low-frequency component pixel value bcThe size of the image is c ∈ { R, G, B }, if the original image pixel is larger than the low-frequency image pixel, the original image pixel is directly subtracted, and the obtained difference value is stored in a high-frequency image X; if the original image pixel is smaller than the low-frequency component pixel, taking the opposite number of the difference value and storing the opposite number into a high-frequency image Y; the mathematical expressions of the images calculated in the R, G, B channels are as follows:
X＝ac-bc，ac＞bc，c∈{R，G，B}
Y＝bc-ac，ac＜bc，c∈{R，G，B}。
s2, enhancing the low-frequency component by using global self-adaptive gamma correction, and enhancing the high-frequency component by using Fourier transform;
the step of enhancing the low frequency component specifically comprises:
the method is realized by a gamma correction power function, and the mathematical expression of the power function is as follows:
L＝Zγ
where L is the output enhanced low-frequency image, Z is the input low-frequency original, and γ is the power exponent.
Specifically, γ takes a value of 2.2.
The enhancement of the high-frequency component is specifically:
s21, transforming the image from a space domain to a frequency domain through Fourier transform;
s22, homomorphic filtering operation and processing are carried out on the frequency spectrum in the frequency domain;
s23, inverse transforming the image into a spatial domain to obtain an enhanced image, wherein the enhanced image of the high-frequency image X is P and the enhanced image of the high-frequency image Y is Q.
And S3, overlapping the enhancement maps corresponding to the low-frequency component and the high-frequency component to obtain an output image. The method comprises the following specific steps:
and (3) superposing the enhanced image and the low-frequency enhanced image corresponding to the high-frequency image X and the high-frequency image Y, wherein the pixel value stored by q is a negative value, and the superposition needs to be subjected to inverse number processing, and the mathematical expression of the method is as follows:
J＝Pc+Lc+Qc，c∈{R，G，B}
wherein J is the superimposed image, PcFor the X enhanced image, LcFor low-frequency images Z enhanced images, QcIs the image after Y enhancement.
Based on the above algorithm, the effect graph shown in fig. 5 can be obtained, and as can be seen by comparing fig. 3 and fig. 4: the information entropy and the average brightness of the BBHE algorithm are large, but partial details are blurred; the contrast of the processing result of the DHE algorithm is obviously enhanced, the definition is high, but an obvious gray overflow effect is generated at the site indication bar; the algorithm result of the method not only keeps higher brightness and definition, but also solves the problems of detail blurring, gray overflow effect and the like. The following table shows the superiority of the algorithm by means of quantitative analysis:
| Method | entropy of information | Contrast ratio | Average brightness | Definition of |
| BBHE | 7.94 | 377 | 138 | 9.88 |
| DHE | 7.14 | 628 | 85 | 11.73 |
| Algorithm of the patent | 7.91 | 644 | 145 | 12.52 |
As can be seen from the table above, the entropy of the image information obtained by the algorithm of the patent is equivalent to that of the BBHE algorithm, but better than that of the DHE algorithm; in contrast, the algorithm of the patent is obviously higher than a BBHE algorithm and slightly higher than a DHE algorithm; in the aspect of average brightness, the algorithm of the patent is slightly higher than a BBHE algorithm and is obviously higher than a DHE algorithm; in the aspect of definition, the algorithm of the patent is also due to BBHE algorithm and DHE algorithm; in general, the algorithm of the present patent is superior.
The above description is only for the preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art should be considered to be within the technical scope of the present invention, and the technical solutions and the inventive concepts thereof according to the present invention should be equivalent or changed within the scope of the present invention.
Claims (7)
1. The rail transit video monitoring image enhancement algorithm is characterized by comprising the following steps of:
s1, dividing an image into two color blocks and two edge components, and respectively corresponding to the frequency domain, wherein the color blocks correspond to the low-frequency component of the image, and the edge details correspond to the high-frequency component of the image;
s2, enhancing the low-frequency component by using global self-adaptive gamma correction, and enhancing the high-frequency component by using Fourier transform;
and S3, overlapping the enhancement maps corresponding to the low-frequency component and the high-frequency component to obtain an output image.
2. The track traffic video monitoring image enhancement algorithm according to claim 1, wherein the step S1 specifically comprises: acquiring low-frequency components of the image by a Gaussian low-pass filter, and recording the obtained low-frequency image as Z; in the high-frequency component extraction process, a manner of retaining all pixels instead of direct subtraction is adopted.
3. The track traffic video surveillance image enhancement algorithm of claim 2, wherein the extraction process of the high frequency component in step S1 is as follows:
respectively judging the pixel value a of the original image in R, G, B three channelscAnd a low-frequency component pixel value bcThe size of the image is c ∈ { R, G, B }, if the original image pixel is larger than the low-frequency image pixel, the original image pixel is directly subtracted, and the obtained difference value is stored in a high-frequency image X; if the original image pixel is smaller than the low-frequency component pixel, taking the opposite number of the difference value and storing the opposite number into a high-frequency image Y; the mathematical expressions of the images calculated in the R, G, B channels are as follows:
X＝ac-bc，ac＞bc，c∈{R，G，B}
Y＝bc-ac，ac＜bc，c∈{R，G，B}。
4. the track traffic video monitoring image enhancement algorithm according to claim 1 or 3, wherein the step of enhancing the low frequency component in step S2 is specifically: the method is realized by a gamma correction power function, and the mathematical expression of the power function is as follows:
L＝Zγ
where L is the output enhanced low-frequency image, Z is the input low-frequency original, and γ is the power exponent.
5. The track traffic video surveillance image enhancement algorithm of claim 3, wherein γ takes the value of 2.2.
6. The track traffic video monitoring image enhancement algorithm according to claim 3, wherein the enhancement of the high frequency component in step S2 specifically comprises:
s21, transforming the image from a space domain to a frequency domain through Fourier transform;
s22, homomorphic filtering operation and processing are carried out on the frequency spectrum in the frequency domain;
s23, inverse transforming the image into a spatial domain to obtain an enhanced image, wherein the enhanced image of the high-frequency image X is P and the enhanced image of the high-frequency image Y is Q.
7. The track traffic video monitoring image enhancement algorithm according to claim 6, wherein the step S3 is as follows: and (3) superposing the enhanced image and the low-frequency enhanced image corresponding to the high-frequency image X and the high-frequency image Y, wherein the pixel value stored by q is a negative value, and the superposition needs to be subjected to inverse number processing, and the mathematical expression of the method is as follows:
J＝Pc+Lc+Qc，c∈{R，G，B}
wherein J is the superimposed image, PcFor the X enhanced image, LcFor low-frequency images Z enhanced images, QcIs the image after Y enhancement.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011566292.7A CN112581404A (en) | 2020-12-25 | 2020-12-25 | Rail transit video monitoring image enhancement algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011566292.7A CN112581404A (en) | 2020-12-25 | 2020-12-25 | Rail transit video monitoring image enhancement algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112581404A true CN112581404A (en) | 2021-03-30 |
Family
ID=75140663
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011566292.7A Pending CN112581404A (en) | 2020-12-25 | 2020-12-25 | Rail transit video monitoring image enhancement algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112581404A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120201455A1 (en) * | 2008-06-04 | 2012-08-09 | Vijay Kumar Kodavalla | Method and apparatus for dynamic and adaptive enhancement of colors in digital video images using value bright-gain |
| CN105931201A (en) * | 2016-04-20 | 2016-09-07 | 北京航空航天大学 | Image subjective visual effect enhancing method based on wavelet transformation |
| CN109919861A (en) * | 2019-01-29 | 2019-06-21 | 浙江数链科技有限公司 | Infrared image enhancing method, device, computer equipment and storage medium |
| CN111383299A (en) * | 2018-12-28 | 2020-07-07 | Tcl集团股份有限公司 | Image processing method and device and computer readable storage medium |
-
2020
- 2020-12-25 CN CN202011566292.7A patent/CN112581404A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120201455A1 (en) * | 2008-06-04 | 2012-08-09 | Vijay Kumar Kodavalla | Method and apparatus for dynamic and adaptive enhancement of colors in digital video images using value bright-gain |
| CN105931201A (en) * | 2016-04-20 | 2016-09-07 | 北京航空航天大学 | Image subjective visual effect enhancing method based on wavelet transformation |
| CN111383299A (en) * | 2018-12-28 | 2020-07-07 | Tcl集团股份有限公司 | Image processing method and device and computer readable storage medium |
| CN109919861A (en) * | 2019-01-29 | 2019-06-21 | 浙江数链科技有限公司 | Infrared image enhancing method, device, computer equipment and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 杨紫烟: ""轨道交通闭路电视监控***视频图像增强算法研究及应用"", 《中国优秀博硕士学位论文全文数据库(硕士) 工程科技II辑》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| KR102234092B1 (en) | Method for inverse tone mapping of an image | |
| CN102611828B (en) | Real-time enhanced processing system for foggy continuous video image | |
| CN104240194B (en) | A kind of enhancement algorithm for low-illumination image based on parabolic function | |
| CN107292830B (en) | Low-illumination image enhancement and evaluation method | |
| CN111598791B (en) | Image defogging method based on improved dynamic atmospheric scattering coefficient function | |
| CN106875358A (en) | Image enchancing method and image intensifier device based on Bayer format | |
| CN105550999A (en) | Video image enhancement processing method based on background reuse | |
| CN105427255A (en) | GRHP based unmanned plane infrared image detail enhancement method | |
| CN100367770C (en) | Method for removing isolated noise point in video | |
| CN111598814A (en) | Single image defogging method based on extreme scattering channel | |
| CN109544470A (en) | A kind of convolutional neural networks single image to the fog method of boundary constraint | |
| Lal et al. | Automatic method for contrast enhancement of natural color images | |
| CN112581404A (en) | Rail transit video monitoring image enhancement algorithm | |
| CN111161189A (en) | Single image re-enhancement method based on detail compensation network | |
| CN111429375A (en) | Night monitoring video quality improving method assisted by daytime image reference | |
| CN107871311A (en) | A kind of image enhaucament and fusion method applied to cmos image sensor | |
| Tang et al. | Sky-preserved image dehazing and enhancement for outdoor scenes | |
| CN112019774B (en) | High-quality display method of infrared high-bit-width digital image | |
| Lyu et al. | A novel visual perception enhancement algorithm for high-speed railway in the low light condition | |
| CN111028184B (en) | Image enhancement method and system | |
| CN114222033A (en) | Adaptive Euler video amplification method based on empirical mode decomposition | |
| CN109886901B (en) | Night image enhancement method based on multi-channel decomposition | |
| Kaur et al. | Image enhancement of underwater digital images by utilizing L* A* B* color space on gradient and CLAHE based smoothing | |
| Dhingra et al. | Fusion of fuzzy enhanced overexposed and underexposed images | |
| CN108259873B (en) | Gradient domain video contrast enhancement method |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210330 |