CN110443807A - A kind of even carrying out image threshold segmentation method of uneven illumination based on luminance proportion - Google Patents
A kind of even carrying out image threshold segmentation method of uneven illumination based on luminance proportion Download PDFInfo
- Publication number
- CN110443807A CN110443807A CN201910568740.8A CN201910568740A CN110443807A CN 110443807 A CN110443807 A CN 110443807A CN 201910568740 A CN201910568740 A CN 201910568740A CN 110443807 A CN110443807 A CN 110443807A
- Authority
- CN
- China
- Prior art keywords
- image
- pixel
- luminance
- value
- threshold
- 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
- 238000000034 method Methods 0.000 title claims abstract description 39
- 238000005286 illumination Methods 0.000 title claims abstract description 25
- 230000011218 segmentation Effects 0.000 title claims abstract description 23
- 238000005315 distribution function Methods 0.000 claims abstract description 7
- 238000009795 derivation Methods 0.000 claims abstract description 4
- 239000011159 matrix material Substances 0.000 claims description 21
- 238000001914 filtration Methods 0.000 claims description 8
- 238000010586 diagram Methods 0.000 claims description 4
- 230000009977 dual effect Effects 0.000 abstract 1
- 230000000694 effects Effects 0.000 description 6
- 230000006870 function Effects 0.000 description 6
- 230000007547 defect Effects 0.000 description 2
- 230000002708 enhancing effect Effects 0.000 description 2
- 238000003709 image segmentation Methods 0.000 description 2
- RZVHIXYEVGDQDX-UHFFFAOYSA-N 9,10-anthraquinone Chemical compound C1=CC=C2C(=O)C3=CC=CC=C3C(=O)C2=C1 RZVHIXYEVGDQDX-UHFFFAOYSA-N 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000013467 fragmentation Methods 0.000 description 1
- 238000006062 fragmentation reaction Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 230000003313 weakening effect Effects 0.000 description 1
Classifications
-
- G06T5/94—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- 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
- G06T2207/10004—Still image; Photographic image
-
- 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
- G06T2207/20021—Dividing image into blocks, subimages or windows
Abstract
The present invention provides a kind of even carrying out image threshold segmentation methods of uneven illumination based on luminance proportion, carry out luminance compensation to the different luminance areas in image first, the luminance background of whole image is made to reach unanimity.Then the data such as the corresponding pixel number of each gray level in gained image are fitted, obtain a gauss of distribution function, by carrying out second order derivation to the function, obtain two threshold values.Using dual threshold, to luminance proportion, treated that image is split processing, if the gray value of pixel is between high-low threshold value in image, determines the pixel for background dot；Otherwise, it is determined that the pixel is target point.Pixel all in image is traversed, the bianry image that segmentation finishes is finally obtained.
Description
Technical field
The present invention relates to technical field of image segmentation, and in particular to a kind of even image threshold of uneven illumination based on luminance proportion
It is worth dividing method.
Background technique
Carrying out image threshold segmentation, the also known as binaryzation of image are widely used in Car license recognition, OCR identification, fingerprint recognition, red
The fields such as outer moving target.Bianry image has the characteristics that memory space is small, processing speed is fast, can be convenient carry out Boolean logic
Operation.The basic thought of image binaryzation is that one gray threshold of setting makes image show in white and black effect.Binaryzation
Method is all the method based on threshold value mostly, that is, finds a suitable threshold value and the pixel grey scale of original image is divided into greater than threshold value
Be less than threshold value two classes, to obtain the result of binaryzation.Currently, binarization method can be divided into Global thresholding, local threshold
Value method two major classes.Simple for the Global thresholding principle of representative with Da-Jin algorithm (OTSU), processing speed is fast, but is easily lost mesh
Mark details and even very sensitive to uneven illumination；Although local thresholding method can preferably handle the even image of uneven illumination,
But it will lead to the problems such as partial region is broken there are false target or target object in binarization result.Not for a width illumination
For uniform image, the optimal threshold in high luminance area and low-light level area differs greatly, and is difficult at this time using Global thresholding
All divide successfully, and be easy to cause local segmentation to be distorted using local thresholding method.Even uneven illumination is frequent in image procossing
The problem of encountering, mainly image background ambient lighting was bad and body surface is reflective the reason of causing the problem etc..
Currently, mainly having histogram equalization method for the even algorithm for image enhancement of uneven illumination, based on illumination-reflection
Homomorphic filtering method, Retinex Enhancement Method etc..The flat equalization of histogram, substantially carries out Nonlinear extension, weight to image
New distribution image pixel value, keeps the quantity of pixel value in certain tonal range roughly equal.In this way, the peak among original histogram
Top part contrast is enhanced, and the lowest point part contrast of two sides reduces, and the histogram for exporting image is one more flat
The visual effect of rude classification can be generated if the fragmentation value of output data is lesser by being segmented histogram；Homomorphic filtering is one
The specific process of picture superposition and compression brightness of image range that kind carries out in a frequency domain.Homomorphic filter can be reduced
Low frequency and increase high frequency, to can be reduced illumination variation and sharpen edge details.The homomorphic filtering technology of image foundation be
Illumination catoptric imaging principle in image acquisition procedures.It belongs to frequency domain processing, and effect is adjusted to image grayscale range,
Uneven problem is illuminated on image by eliminating.But homomorphic filtering method can not cut down multiplying property and convolution noise；It is general
Retinex algorithm to light image estimate when, can all assume initial light image be it is slowly varying, i.e., light image is smooth
's.But practical really not so, brightness differs greatly the edge in region, and image irradiation variation is simultaneously unsmooth.So in this feelings
Under condition, Retinex, which enhances enhancing image of the algorithm in the big region of luminance difference, can generate halation.
For the defect for avoiding above-mentioned algorithm, the invention proposes a kind of even image thresholds of the uneven illumination based on luminance proportion
Dividing method first carries out piecemeal to the background luminance of image and dynamically adjusts, locates adjusted image background brightness substantially
It is split in identical level, then using dual-threshold voltage, to solve problems of the prior art.
Summary of the invention
The technical problem to be solved in the present invention is that using Global thresholding to the even image of uneven illumination at present for above-mentioned
Or the common Binarization methods such as local thresholding method are difficult to the technical issues of obtaining satisfactory segmentation effect, provide a kind of base
Above-mentioned technological deficiency is solved in the even carrying out image threshold segmentation method of the uneven illumination of luminance proportion.
A kind of even carrying out image threshold segmentation method of uneven illumination based on luminance proportion, comprising:
Step 1, input source picture seek the average gray I of source images；
Step 2 divides the image into M × N block according to a certain size, finds out every piece of average brightness, obtains the bright of sub-block
Spend matrix D；Wherein, the average brightness of a sub-block represents an element of luminance matrix D；
Step 3, the average gray I that source images are subtracted with the value of each element of matrix D, obtain the luminance difference square of sub-block
Battle array E；
Matrix E is extended to and source images Luminance Distribution matrix R of a size by step 4 with bicubic interpolation method；
The pixel brightness value of source images is subtracted corresponding numerical value in matrix R, then imposes smothing filtering by step 5, can be obtained
Image after to luminance proportion；
In step 6, obtaining step 5 equalize after image grey level histogram, measure image in dimensional parameters [s, t] simultaneously
Average gray value greay_mean is calculated, counting number T, s, t that each gray level occurs indicates the size of image；
Step 7 goes out intensity histogram diagram data obtained in step 6, including pixel grayscale and each gray level
Existing number T is recorded and is fitted to Gaussian distribution curve；
Step 8, the expectation μ and variances sigma that One-Dimensional Normal distribution probability density function is obtained according to Gaussian distribution curve, two
The threshold value of value image chooses condition | X- μ |≤σ, and to obtain two threshold Xs1、X2, wherein X1> X2, after luminance proportion
The pixel gray value G of image is compared with resulting two threshold values, if X2< G < X1, then G is set to 255, otherwise sets G
It is 0, traverses each pixel in image, final bianry image can be obtained.
Further, the specific method of step 2 includes: to divide the image into M × N block according to a certain size, finds out every piece
Average value, the luminance matrix D for obtaining sub-block, which refers to, is divided into M × N block for a secondary picture, and gray level is (0 ..., L), then average
Brightness are as follows:Wherein p (i, j) is the pixel brightness value that coordinate is in image；With the son of size M × N
Block carries out piecemeal to the image, then the luminance mean value of each sub-block are as follows:
Further, in the step 7, input be image after compensation grey level histogram, and by pixel grayscale
And its corresponding frequency of occurrence T is fitted to Gaussian distribution curve.
Further, the segmentation threshold of bianry image is chosen in the step 8 | X- μ |≤σ, specific steps include:
S81, the One-Dimensional Normal distribution probability density function that fitting obtains is denoted as f (x):
S82, derivation is carried out to f (x) function, obtains derivative f ' (x),
In gauss of distribution function, abscissa is pixel grey scale, and ordinate is pixel number, when | f ' (x) | take maximum
When value, the quantity and gray-value variation of pixel are the most obvious.
S83, by the property of gauss of distribution function it is found that as f " when (x)=0, f ' (x) obtains maximum and minimum, this
When
X1=μ-σ, X2=μ+σ；
Enable X1、X2For segmentation threshold, by the pixel gray value G of image after luminance proportion and resulting two threshold values into
Row compares, if X2< G < X1, then G is set to 255, G is otherwise set to 0, traversed each pixel in image, obtain final two
It is worth image.
Compared with prior art, the invention has the advantages that: the present invention proposes that a kind of uneven illumination based on luminance proportion is even
Carrying out image threshold segmentation method, this method, by carrying out luminance compensation to image block, keep image background bright in terms of image enhancement
Degree is generally in identical level；In terms of threshold value selection, by carrying out one-dimensional gaussian profile fitting to grey level histogram, obtain
Two optimal segmenting thresholds, segmentation effect become apparent from, and the various indexs such as recall rate, precision rate, error rate are improved.
Detailed description of the invention
Present invention will be further explained below with reference to the attached drawings and examples, in attached drawing:
Fig. 1 is a kind of even carrying out image threshold segmentation method flow diagram of uneven illumination based on luminance proportion of the invention.
Specific embodiment
For a clearer understanding of the technical characteristics, objects and effects of the present invention, now control attached drawing is described in detail
A specific embodiment of the invention.
A kind of even carrying out image threshold segmentation method of uneven illumination based on luminance proportion, as shown in Figure 1, comprising:
Step 1, input source picture seek the average gray I of source images；
Step 2 divides the image into M × N block according to a certain size, finds out every piece of average value, obtains the brightness square of sub-block
Battle array D；
Step 3, the average gray I that source images are subtracted with the value of each element of matrix D, obtain the luminance difference square of sub-block
Battle array E；
Matrix E is extended to and source images Luminance Distribution matrix R of a size by step 4 with bicubic interpolation method；
The pixel brightness value of source images is subtracted corresponding numerical value in matrix R, then imposes smothing filtering by step 5, can be obtained
Image after to luminance proportion；
In step 6, obtaining step 5 equalize after image grey level histogram, measure image in dimensional parameters [s, t] simultaneously
Average gray value greay_mean is calculated, counting number T, s, t that each gray level occurs indicates the size of image；
Step 7 goes out intensity histogram diagram data obtained in step 6, including pixel grayscale and each gray level
Existing number T is recorded and is fitted to Gaussian distribution curve；
The threshold value of step 8, the expectation μ and variances sigma for obtaining One-Dimensional Normal distribution probability density function, bianry image is chosen
Condition can be set as | X- μ |≤σ, available two threshold Xs1、X2, wherein X1> X2.By the pixel of image after luminance proportion
Gray value G is compared with resulting two threshold values, if X2< G < X1, then G is set to 255, G is otherwise set to 0, traverse image
In each pixel, final bianry image can be obtained.
The step 2, in 3,4, background and target should be evenly distributed on image 4 parts up and down, if illumination
It is uniformly, then the average value of this 4 part gray value should be not much different；, whereas if difference is larger, then made by processing
Its similar mean values shows the effect of similar uniform illumination.The specific method is as follows: dividing the image into M × N according to a certain size
Block finds out every piece of average value, and the luminance matrix D for obtaining sub-block, which refers to, is divided into M × N block for a secondary picture, and gray level is
(0 ..., L), then average brightness are as follows:Wherein p (i, j) is the pixel intensity that coordinate is in image
Value.Piecemeal is carried out to the image with the sub-block of size M × N, then the luminance mean value of each sub-block are as follows:Then the difference of sub-block luminance mean value and full figure luminance mean value is Δlum=Lumav_bm-Lumav, right
For ΔlumThe brightness of positive sub-block carries out Weakening treatment, to ΔlumThe brightness for the sub-block being negative carries out enhancing processing.To avoid office
The defect of portion's threshold method does not directly add or subtracts the same adjustment numerical value to each sub-block, but according to piecemeal
Format interpolation is carried out to the matrix of sub-block, be allowed to expand to entire original image size by using bicubic interpolation method.So
Δ after subtracting extension with the value of the pixel of original image afterwardslumMatrix is just able to achieve the luminance proportion adjustment of entire image.
The segmentation threshold of bianry image is chosen in the step 8 | X- μ |≤σ, the specific steps are as follows:
S81, the One-Dimensional Normal distribution probability density function that fitting obtains is denoted as f (x), is shown below:
S82, derivation is carried out to above formula function, obtains derivative f ' (x),
Image binaryzation method will usually find a threshold value when handling grayscale image, so that the difference between profile and background
Different maximum.And in gauss of distribution function, abscissa is pixel grey scale, and ordinate is pixel number, when | f ' (x) | take maximum
When value, the quantity and gray-value variation of pixel are the most obvious.
S83, by the property of gauss of distribution function it is found that as f " when (x)=0, f ' (x) obtains maximum and minimum, this
When X1=μ-σ, X2=μ+σ.
Enable X1、X2For segmentation threshold, by the pixel gray value G of image after luminance proportion and resulting two threshold values into
Row compares, if X2< G < X1, then G is set to 255, G is otherwise set to 0, traversed each pixel in image, obtain final two
It is worth image.
In summary: the present invention, by carrying out luminance compensation to image block, keeps image background bright in terms of image enhancement
Degree is generally in identical level；In terms of threshold value selection, by carrying out one-dimensional gaussian profile fitting to grey level histogram, obtain
Two optimal segmenting thresholds, the image segmentation even for uneven illumination becomes apparent from, in recall rate, precision rate, error rate etc.
It is better than homomorphic filtering and Retinex algorithm for image enhancement in index.Present invention can apply to aerospaces, UAV Flight Control
Deng the military field for needing to carry out feature extraction under complex background environment；Civilian aspect, mobile robot, vision SLAM,
Also there is more wide application prospect in the fields such as dam body monitoring.
The embodiment of the present invention is described with above attached drawing, but the invention is not limited to above-mentioned specific
Embodiment, the above mentioned embodiment is only schematical, rather than restrictive, those skilled in the art
Under the inspiration of the present invention, without breaking away from the scope protected by the purposes and claims of the present invention, it can also make very much
Form, all of these belong to the protection of the present invention.
Claims (4)
1. a kind of even carrying out image threshold segmentation method of uneven illumination based on luminance proportion characterized by comprising
Step 1, input source picture seek the average gray I of source images；
Step 2 divides the image into M × N block according to a certain size, finds out every piece of average brightness, obtains the brightness square of sub-block
Battle array D；Wherein, the average brightness of a sub-block represents an element of luminance matrix D；
Step 3, the average gray I that source images are subtracted with the value of each element of matrix D, obtain the luminance difference matrix E of sub-block；
Matrix E is extended to and source images Luminance Distribution matrix R of a size by step 4 with bicubic interpolation method；
The pixel brightness value of source images is subtracted corresponding numerical value in matrix R, then imposes smothing filtering by step 5, can be obtained bright
Image after degree equalization；
The grey level histogram of image after equalizing in step 6, obtaining step 5 measures the dimensional parameters [s, t] in image and calculates
Average gray value greay_mean, counting number T, s, t that each gray level occurs indicates the size of image；
Step 7, by intensity histogram diagram data obtained in step 6, including pixel grayscale and each gray level occur
Number T is recorded and is fitted to Gaussian distribution curve；
Step 8, the expectation μ and variances sigma that One-Dimensional Normal distribution probability density function is obtained according to Gaussian distribution curve, binary map
The threshold value of picture chooses condition | X- μ |≤σ, and to obtain two threshold Xs1、X2, wherein X1> X2, by image after luminance proportion
Pixel gray value G be compared with resulting two threshold values, if X2< G < X1, then G is set to 255, G is otherwise set to 0,
Each pixel in image is traversed, final bianry image can be obtained.
2. a kind of even carrying out image threshold segmentation method of uneven illumination based on luminance proportion according to claim 1, feature
It is, the specific method of step 2 includes: to divide the image into M × N block according to a certain size, finds out every piece of average value, obtains son
The luminance matrix D of block, which refers to, is divided into M × N block for a secondary picture, and gray level is (0 ..., L), then average brightness are as follows:Wherein p (i, j) is the pixel brightness value that coordinate is in image；With the sub-block of size M × N to this
Image carries out piecemeal, then the luminance mean value of each sub-block are as follows:
3. a kind of even carrying out image threshold segmentation method of uneven illumination based on luminance proportion according to claim 1, feature
Be, in the step 7, input be compensation after image grey level histogram, and by pixel grayscale and its it is corresponding go out
Occurrence number T is fitted to Gaussian distribution curve.
4. a kind of even carrying out image threshold segmentation method of uneven illumination based on luminance proportion according to claim 1, feature
It is, the segmentation threshold of bianry image is chosen in the step 8 | X- μ |≤σ, specific steps include:
S81, the One-Dimensional Normal distribution probability density function that fitting obtains is denoted as f (x):
S82, derivation is carried out to f (x) function, obtains derivative f ' (x),
In gauss of distribution function, abscissa is pixel grey scale, and ordinate is pixel number, when | f ' (x) | when being maximized,
The quantity and gray-value variation of pixel are the most obvious；
S83, by the property of gauss of distribution function, it is found that as f, " when (x)=0, f ' (x) obtains maximum and minimum, at this time
X1=μ-σ, X2=μ+σ；
Enable X1、X2For segmentation threshold, the pixel gray value G of image after luminance proportion and resulting two threshold values are compared
Compared with if X2< G < X1, then G is set to 255, G is otherwise set to 0, traversed each pixel in image, obtain final binary map
Picture.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910568740.8A CN110443807A (en) | 2019-06-27 | 2019-06-27 | A kind of even carrying out image threshold segmentation method of uneven illumination based on luminance proportion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910568740.8A CN110443807A (en) | 2019-06-27 | 2019-06-27 | A kind of even carrying out image threshold segmentation method of uneven illumination based on luminance proportion |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110443807A true CN110443807A (en) | 2019-11-12 |
Family
ID=68428356
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910568740.8A Pending CN110443807A (en) | 2019-06-27 | 2019-06-27 | A kind of even carrying out image threshold segmentation method of uneven illumination based on luminance proportion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110443807A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111539967A (en) * | 2020-04-24 | 2020-08-14 | 电子科技大学 | Method and system for identifying and processing interference fringe region in terahertz imaging of focal plane |
| CN111986113A (en) * | 2020-08-20 | 2020-11-24 | 浙江理工大学 | Optical image shadow eliminating method and system |
| CN112215237A (en) * | 2020-11-06 | 2021-01-12 | 泰康保险集团股份有限公司 | Image processing method and device, electronic equipment and computer readable storage medium |
| CN113160098A (en) * | 2021-04-16 | 2021-07-23 | 浙江大学 | Processing method of dense particle image under condition of uneven illumination |
| CN113781292A (en) * | 2021-08-23 | 2021-12-10 | 北京达佳互联信息技术有限公司 | Image processing method and device, electronic device and storage medium |
| CN113870233A (en) * | 2021-09-30 | 2021-12-31 | 常州市宏发纵横新材料科技股份有限公司 | Binding yarn detection method, computer equipment and storage medium |
| CN114241426A (en) * | 2022-02-23 | 2022-03-25 | 北京奥康达体育产业股份有限公司 | Outdoor intelligent gymnasium management system |
| CN114418890A (en) * | 2022-01-20 | 2022-04-29 | 中国人民解放军国防科技大学 | Uneven-illumination text image processing method |
| CN114549670A (en) * | 2022-02-23 | 2022-05-27 | 京东方数字科技有限公司 | Image processing method and image processing system |
| CN114742784A (en) * | 2022-03-31 | 2022-07-12 | 精诚工坊电子集成技术（北京）有限公司 | Skin image red blood silk marking method, evaluation method and system |
| CN114897923A (en) * | 2022-05-25 | 2022-08-12 | 中国海洋大学 | Natural gas hydrate CT image threshold segmentation method, system, equipment and medium |
| CN116630425A (en) * | 2023-07-21 | 2023-08-22 | 长春市天之城科技有限公司 | Intelligent food detection system based on X rays |
| CN116993636A (en) * | 2023-07-10 | 2023-11-03 | 中国地质大学（武汉） | Image enhancement method and device for underground low-illumination deep stratum empty area |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101404086A (en) * | 2008-04-30 | 2009-04-08 | 浙江大学 | Target tracking method and device based on video |
| US7570809B1 (en) * | 2004-07-03 | 2009-08-04 | Hrl Laboratories, Llc | Method for automatic color balancing in digital images |
| US20140341467A1 (en) * | 2013-05-14 | 2014-11-20 | Hyundai Mobis Co., Ltd. | Method for brightness equalization of various images |
| CN107705314A (en) * | 2017-11-01 | 2018-02-16 | 齐鲁工业大学 | A kind of more subject image dividing methods based on intensity profile |
| CN109544581A (en) * | 2018-11-15 | 2019-03-29 | 湖南自兴智慧医疗科技有限公司 | A kind of automatic division processing method of early pregnancy uterus image |
-
2019
- 2019-06-27 CN CN201910568740.8A patent/CN110443807A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7570809B1 (en) * | 2004-07-03 | 2009-08-04 | Hrl Laboratories, Llc | Method for automatic color balancing in digital images |
| CN101404086A (en) * | 2008-04-30 | 2009-04-08 | 浙江大学 | Target tracking method and device based on video |
| US20140341467A1 (en) * | 2013-05-14 | 2014-11-20 | Hyundai Mobis Co., Ltd. | Method for brightness equalization of various images |
| CN107705314A (en) * | 2017-11-01 | 2018-02-16 | 齐鲁工业大学 | A kind of more subject image dividing methods based on intensity profile |
| CN109544581A (en) * | 2018-11-15 | 2019-03-29 | 湖南自兴智慧医疗科技有限公司 | A kind of automatic division processing method of early pregnancy uterus image |
Non-Patent Citations (3)
| Title |
|---|
| 张政寿: "《二维尺寸链理论及计算》", 30 April 1988, 国防工业出版社 * |
| 彭兴邦 等: "一种基于亮度均衡的图像阈值分割技术", 《计算机技术与发展》 * |
| 李泽宇 等: "拟合正态分布曲线的肺野图像分割与三维重建", 《计算机工程与设计》 * |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111539967B (en) * | 2020-04-24 | 2023-03-28 | 电子科技大学 | Method and system for identifying and processing interference fringe region in terahertz imaging of focal plane |
| CN111539967A (en) * | 2020-04-24 | 2020-08-14 | 电子科技大学 | Method and system for identifying and processing interference fringe region in terahertz imaging of focal plane |
| CN111986113A (en) * | 2020-08-20 | 2020-11-24 | 浙江理工大学 | Optical image shadow eliminating method and system |
| CN111986113B (en) * | 2020-08-20 | 2024-03-22 | 浙江理工大学 | Optical image shadow elimination method and system |
| CN112215237A (en) * | 2020-11-06 | 2021-01-12 | 泰康保险集团股份有限公司 | Image processing method and device, electronic equipment and computer readable storage medium |
| CN112215237B (en) * | 2020-11-06 | 2023-10-13 | 泰康保险集团股份有限公司 | Image processing method and device, electronic equipment and computer readable storage medium |
| CN113160098A (en) * | 2021-04-16 | 2021-07-23 | 浙江大学 | Processing method of dense particle image under condition of uneven illumination |
| CN113781292A (en) * | 2021-08-23 | 2021-12-10 | 北京达佳互联信息技术有限公司 | Image processing method and device, electronic device and storage medium |
| CN113870233A (en) * | 2021-09-30 | 2021-12-31 | 常州市宏发纵横新材料科技股份有限公司 | Binding yarn detection method, computer equipment and storage medium |
| CN114418890A (en) * | 2022-01-20 | 2022-04-29 | 中国人民解放军国防科技大学 | Uneven-illumination text image processing method |
| CN114418890B (en) * | 2022-01-20 | 2024-01-23 | 中国人民解放军国防科技大学 | Method for processing text image with uneven illumination |
| CN114549670A (en) * | 2022-02-23 | 2022-05-27 | 京东方数字科技有限公司 | Image processing method and image processing system |
| CN114549670B (en) * | 2022-02-23 | 2023-04-07 | 京东方数字科技有限公司 | Image processing method and image processing system |
| CN114241426A (en) * | 2022-02-23 | 2022-03-25 | 北京奥康达体育产业股份有限公司 | Outdoor intelligent gymnasium management system |
| CN114742784A (en) * | 2022-03-31 | 2022-07-12 | 精诚工坊电子集成技术（北京）有限公司 | Skin image red blood silk marking method, evaluation method and system |
| CN114897923A (en) * | 2022-05-25 | 2022-08-12 | 中国海洋大学 | Natural gas hydrate CT image threshold segmentation method, system, equipment and medium |
| CN116993636A (en) * | 2023-07-10 | 2023-11-03 | 中国地质大学（武汉） | Image enhancement method and device for underground low-illumination deep stratum empty area |
| CN116993636B (en) * | 2023-07-10 | 2024-02-13 | 中国地质大学（武汉） | Image enhancement method and device for underground low-illumination deep stratum empty area |
| CN116630425A (en) * | 2023-07-21 | 2023-08-22 | 长春市天之城科技有限公司 | Intelligent food detection system based on X rays |
| CN116630425B (en) * | 2023-07-21 | 2023-09-22 | 长春市天之城科技有限公司 | Intelligent food detection system based on X rays |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110443807A (en) | A kind of even carrying out image threshold segmentation method of uneven illumination based on luminance proportion | |
| Shanmugavadivu et al. | Particle swarm optimized multi-objective histogram equalization for image enhancement | |
| CN108765336B (en) | Image defogging method based on dark and bright primary color prior and adaptive parameter optimization | |
| CN102789635B (en) | Image enhancement method and image enhancement device | |
| CN109087254B (en) | Unmanned aerial vehicle aerial image haze sky and white area self-adaptive processing method | |
| CN111145105B (en) | Image rapid defogging method and device, terminal and storage medium | |
| CN109272461A (en) | Infrared image enhancing method based on median filtering and color histogram | |
| CN104463804A (en) | Image enhancement method based on intuitional fuzzy set | |
| CN109190632A (en) | A kind of binarization method of ancient books file and picture | |
| CN109377450A (en) | A kind of edge-protected denoising method | |
| CN110782407A (en) | Single image defogging method based on sky region probability segmentation | |
| CN112381826B (en) | Binarization method of edge defect image | |
| CN112528868B (en) | Illegal line pressing judgment method based on improved Canny edge detection algorithm | |
| CN101447027A (en) | Binaryzation method of magnetic code character area and application thereof | |
| CN111027564A (en) | Low-illumination imaging license plate recognition method and device based on deep learning integration | |
| CN113344810A (en) | Image enhancement method based on dynamic data distribution | |
| CN116823686B (en) | Night infrared and visible light image fusion method based on image enhancement | |
| Zhang et al. | Research on the License Plate Recognition based on MATLAB | |
| CN110930358B (en) | Solar panel image processing method based on self-adaptive algorithm | |
| CN108205678A (en) | A kind of nameplate Text region processing method containing speck interference | |
| CN110633705A (en) | Low-illumination imaging license plate recognition method and device | |
| CN106384103A (en) | Vehicle face recognition method and device | |
| CN116452447A (en) | Low-illumination high-definition image processing method | |
| CN110288536A (en) | A kind of borehole image processing method based on improvement bilateral filtering | |
| CN114359083B (en) | High-dynamic thermal infrared image self-adaptive preprocessing method for interference environment |
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: 20191112 |