CN111754418B - Image enhancement method based on Hess matrix - Google Patents
Image enhancement method based on Hess matrix Download PDFInfo
- Publication number
- CN111754418B CN111754418B CN202010430976.8A CN202010430976A CN111754418B CN 111754418 B CN111754418 B CN 111754418B CN 202010430976 A CN202010430976 A CN 202010430976A CN 111754418 B CN111754418 B CN 111754418B
- Authority
- CN
- China
- Prior art keywords
- image
- fractional order
- hess
- matrix
- 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
Links
- 239000011159 matrix material Substances 0.000 title claims abstract description 37
- 238000000034 method Methods 0.000 title claims abstract description 20
- 230000001186 cumulative effect Effects 0.000 claims abstract description 13
- 238000005315 distribution function Methods 0.000 claims abstract description 13
- 238000013507 mapping Methods 0.000 claims abstract description 13
- 238000012545 processing Methods 0.000 claims abstract description 10
- 230000009466 transformation Effects 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 3
- 230000006872 improvement Effects 0.000 abstract description 2
- 238000012795 verification Methods 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 11
- 230000000007 visual effect Effects 0.000 description 5
- 238000004364 calculation method Methods 0.000 description 3
- 238000002474 experimental method Methods 0.000 description 3
- 241000282412 Homo Species 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 description 1
- 238000007792 addition Methods 0.000 description 1
- 238000012271 agricultural production Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000000969 carrier Substances 0.000 description 1
- 230000001419 dependent effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
-
- 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
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
The invention discloses an image enhancement method based on a Hess matrix, and relates to the technical fields of digital image processing, computer vision and the like. The method comprises the following specific steps: 1) Processing the input image by using a multiscale fractional order Hess matrix to generate a characteristic weighted image; 2) Suppressing the characteristic value of the strong characteristic information pixel in the characteristic weighted image to obtain a background weighted image; 3) Obtaining a final mapping function by using a histogram of the background weighted image and a cumulative distribution function formed by normalizing the histogram; 4) And obtaining the enhanced image by using the mapping function. The method uses matlab as a platform for verification, and combines the related knowledge of the Hess matrix, thereby finishing the improvement of the image contrast to achieve the enhancement purpose, having practical significance and obtaining better enhancement effect.
Description
Technical Field
The invention relates to an image enhancement method based on a Hess matrix, and belongs to the technical fields of digital image processing, machine vision and the like.
Background
Images are information carriers commonly used in human society. Studies have shown that human acquired visual image information is approximately 80% of the human accepted information. It follows that visual information is of importance to humans, and images are the primary way humans acquire visual information. From the 60 s of the 20 th century, with the continuous improvement and popularization of computer technology, digital image processing rapidly developed at home and abroad, and is widely used in the fields of scientific research, industrial and agricultural production, biomedical engineering, aerospace, military, industry, robot industry and the like, and plays an increasingly important role in human life.
In daily life we can record the instant of life with photos. Today, with the development of mobile devices and the internet, people increasingly like to take pictures and share with others. However, sometimes, due to the influence of equipment, environment and the like, the acquired photo may have low contrast, poor brightness and large noise, so that the visual effect is poor and the visual requirement of people cannot be met. The image enhancement method can solve the problems of color, contrast and image definition. To this end, professionals can interactively trim pictures, such as histogram equalization, sharpening, contrast adjustment, and color mapping, through tools provided by professional image processing software, while also fine-tuning the various objects and details in the picture. The quality of the results is largely dependent on the skill and aesthetic of the software user. Still other software can be automatically enhanced with one-touch enhancement, but this may not be largely responsible for changing its contrast and color due to the difficulty in adjusting the balance of the various factors.
Disclosure of Invention
In view of the above, it is an object of the present invention to solve some of the problems of the prior art, or at least to alleviate them. Therefore, the image enhancement method based on the Hess matrix is an effective method applicable to image enhancement and generated based on digital image processing. The method solves the problem of image enhancement, and saves economic cost and time cost.
The technical scheme adopted by the invention for realizing the purpose is as follows: an image enhancement method based on a Hess matrix comprises the following steps:
1) Rapidly generating a Hess matrix of an input image under different scales by using an integral graph, expanding the multi-scale Hess matrix to a fractional order to obtain a multi-scale fractional order Hess matrix, and processing the input image by using the multi-scale fractional order Hess matrix to generate a characteristic weighted image;
3) Suppressing the characteristic value of the strong characteristic information pixel in the characteristic weighted image to obtain a background weighted image;
4) Obtaining a mapping function by using a histogram of the background weighted image and a cumulative distribution function formed by normalizing the histogram;
5) And obtaining the enhanced image by using the mapping function.
Further, the step 4) specifically includes obtaining the histogram H thereof by background weighted image f And normalize the histogram to obtain P f (n) obtaining a cumulative distribution function by using the normalized histogram, and then obtaining a mapping function by combining the cumulative distribution function with the gray level transformation function.
The invention has the advantages and beneficial effects as follows:
the invention realizes real-time automatic image enhancement by utilizing the technologies in the fields of digital image processing, machine vision and the like. The invention can obtain the enhanced image with improved contrast by inputting a low-quality image without other operations. The invention has the following advantages: 1) The matlab platform is utilized for development and test, so that the economic cost is low; 2) The operation is simple and convenient, the image with enhanced contrast can be obtained by inputting a pair of low-contrast pictures, and the effect is obvious; 3) The processing time is short, and a result can be obtained by inputting a pair of images.
The multiscale fractional order Hess matrix can detect and quantify the characteristic information of each pixel in the input image; the intensity of the characteristic pixels (i.e. pixels with high pixel values) can be suppressed and the contrast of the background pixels is excited by the characteristic suppression, and the histogram formed by the background weighted image mainly comprises the background pixels with low contrast, so that the contrast enhancement mainly aims at the background pixels, and the characteristic pixels are less or remain; the integral image is introduced to improve the calculation speed and reduce the calculation complexity; the algorithm can do global contrast enhancement and local contrast enhancement and can make the enhancement result look more natural.
Drawings
FIG. 1 is a system flow diagram of the present invention;
fig. 2 is an image to be enhanced and after enhancement (for gray scale image) in order from left to right;
fig. 3 is a sequence of images to be enhanced and enhanced (for color maps) from left to right.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and specifically described below with reference to the drawings in the embodiments of the present invention. The described embodiments are only a few embodiments of the present invention.
The system flow chart is shown in fig. 1, and the image enhancement method based on the Hess matrix comprises the following steps:
the first step: the method comprises the steps of rapidly generating a Hess matrix of an input image under different scales by using an integral graph, expanding the Hess matrix to fractional order, and processing the input image by using a multi-scale fractional order Hess matrix to generate a characteristic weighted image;
and a second step of: suppressing the characteristic value of the strong characteristic information pixel in the characteristic weighted image to obtain a background weighted image;
and a third step of: obtaining a mapping function by using a histogram of the background weighted image and a cumulative distribution function formed by normalizing the histogram;
fourth step: and obtaining the enhanced image by using the mapping function.
Further, the fractional order in step 1) is Grunwald-Letnikov (G-L) fractional order, and corresponding parameters are set for the G-L fractional order and the Hess matrix, and the Hess matrix of the image is defined as follows:
wherein Lxx (x, y, sigma) is the second partial derivative of a Gaussian functionConvolutions with the image I at coordinate points (x, y), g (σ) represents a gaussian function. L (L) xy (x, y, sigma) and L yy (x, y, σ) are the same.
G-L fractional order based on partial derivative of fractional order Gaussian functionThe definition is as follows:
where α is a fractional order. g (x, y, sigma) is a gaussian function,the method is based on the partial derivative of a fractional Gaussian function, namely G-L fractional order, wherein sigma is Gaussian standard deviation, different sigma can generate images with different scales, and the larger the sigma value is, the larger the scale of the generated images is.
Combining fractional order Hess matrixes under different scales in a weighted summation mode:
wherein MSFrH (x, y) represents a multiscale fractional order Hess matrix FrH (x, y, σ) j ) Representing a fractional order Hess matrix, w j Denoted as sigma at the j FrH th parameter j FrH of (a). j represents the j-th scale, m represents the total number of scales, sigma j As parameters, representing Gaussian standard deviation, different sigma can generate images with different scales
Further, in the step 1), the integral graph I ∑(x,y) Defined as the sum of gray values of all pixel points in a rectangular area enclosed by the upper left corner of the image I and the point (x, y), namely
I (I, j) represents the pixel point gray value of the image I at the coordinate point (I, j).
Through the integral graph, the sum of gray values in the rectangular area of the image can be calculated by only 3 additions, so that the integral graph can calculate an approximate Hess matrix under the condition of very low calculation time complexity.
Further, in step 2), the threshold τ=1×10 is set when suppressing the feature value of the strong feature information pixel in the feature weighted image -6 If the value in the feature weighted image is smaller than the threshold, the original value is unchanged, and if the value is equal to or larger than the threshold, the value in the feature weighted image is set to 0.
Further, the histogram H of the Background Weighted Image (BWI) in step 3) f The definition is as follows:
H f ={h f (n)|0≤n≤2 B -1} (5)
wherein ,
wherein ,
the input image size is M×N, the pixel value at the coordinates (x, y) is f (x, y) =n, n∈ {0,1, L,2 B -1}, where B is the number of image bits, in this embodiment the number of image bits is 8, so b=8, BWI represents the feature weighted image generated in step 2, BWI (x, y) is the value of (x, y) in BWI. And to H f Normalized H f The definition is as follows:
P f (n) represents normalized H f Histogram equalization.
The corresponding cumulative distribution function is:
P f (i) Represents H f Normalized P f P when n=i in (n) f A value of (n).
The general definition of the gray scale transformation function is:
T(n)=(2 B -1)c(n)+0.5 (10)
wherein c (n) is a cumulative distribution function, T (n) is an enhanced image, and the final mapping function is obtained by taking the formula (9) into the formula (10) as follows:
T(n)=(2 B -1)C f (n)+0.5 (11)
C f (n) represents a cumulative distribution function. T (n) is the enhanced image.
In order to verify the effect of the present invention, the following experiments were performed:
verification experiments were performed on a computer with a matlab2016a configured as an E5-2603 processor (1.7 GHz), 32GB memory.
The experimental method comprises the following steps:
in the experimental process, the image enhancement method based on the Hess matrix provided by the invention is used for enhancing the selected low-contrast image (aiming at the gray level image and the color image respectively) to obtain the final enhanced image with improved contrast. The experimental results are shown in fig. 2 and 3.
The above examples should be understood as illustrative only and not limiting the scope of the invention. Various changes and modifications to the present invention may be made by one skilled in the art after reading the teachings herein, and such equivalent changes and modifications are intended to fall within the scope of the invention as defined in the appended claims.
Claims (3)
1. An image enhancement method based on a Hess matrix is characterized by comprising the following steps:
1) Rapidly generating a Hess matrix of an input image under different scales by using an integral graph, expanding the multi-scale Hess matrix to a fractional order to obtain a multi-scale fractional order Hess matrix, and processing the input image by using the multi-scale fractional order Hess matrix to generate a characteristic weighted image;
the fractional order is G-L, corresponding parameters are set for the G-L fractional order and the Hess matrix, and the fractional Hess matrix under different scales is combined in a weighted summation mode to obtain a multi-scale fractional Hess matrix;
the Hess matrix is:
wherein Lxx (x, y, sigma) is the second partial derivative of a Gaussian functionConvolutions with image I at points (x, y), g (σ) representingGaussian function, L xy (x, y, sigma) and L yy (x, y, σ) is the same;
the fractional order Hess matrix is determined by G-L fractional order based on fractional order Gaussian partial derivativeG-L of (C) is defined as:
wherein alpha is a fractional order, sigma is a Gaussian standard deviation, g (x, y, sigma) is a Gaussian function with standard deviation sigma,is based on the partial derivative of a fractional order gaussian function;
combining fractional order Hess matrixes under different scales in a weighted summation mode:
wherein MSFrH (x, y) represents a multiscale fractional order Hess matrix FrH (x, y, σ) j ) Representing a fractional order Hess matrix, w j Denoted as sigma at the j FrH th parameter j The weight of FrH of (2), j represents the j-th scale, m represents the total number of scales, sigma j As parameters, the standard deviation of gauss is represented;
2) Suppressing the characteristic value of the strong characteristic information pixel in the characteristic weighted image to obtain a background weighted image;
3) The mapping function is obtained by using a histogram of the background weighted image and a cumulative distribution function formed by normalizing the histogram, and the specific steps are as follows: obtaining a histogram of the background weighted image through the background weighted image, and normalizing the histogram to obtain P f (n) obtaining a cumulative distribution function by using the normalized histogram, and then combining the cumulative distribution function with the gray level transformation function to obtain a final mapping function;
histogram H of the background weighted image f The definition is as follows:
wherein ,
wherein ,
the input image size is M×N, the pixel value at the coordinates (x, y) is f (x, y) =n, N ε {0,1, …,2 B -1}, wherein B represents the number of image bits, BWI (x, y) being the value of (x, y) coordinates in the feature weighted image;
the cumulative distribution function is:
P f (i) Represents H f Normalized P f P when n=i in (n) f A value of (n);
the mapping function is as follows:
T(n)=(2 B -1)C f (n)+0.5
t (n) is the enhanced image, C f (n) represents a cumulative distribution function;
4) And obtaining the enhanced image by using the mapping function.
2. The image enhancement method based on a Hess matrix according to claim 1, characterized in that: the integral graph is the sum of gray values of all pixel points in a rectangular area surrounded by the upper left corner of the image I and (x, y).
3. A Hess matrix based image according to claim 1The enhancement method is characterized in that: in the step 3), the characteristic value of the strong characteristic information pixel in the characteristic weighted image is suppressed by adopting a threshold value setting method, and the threshold value tau=1×10 -6 If the value in the feature weighted image is smaller than the threshold, the original value is unchanged, and if the value is equal to or larger than the threshold, the value in the feature weighted image is set to 0.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010430976.8A CN111754418B (en) | 2020-05-20 | 2020-05-20 | Image enhancement method based on Hess matrix |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010430976.8A CN111754418B (en) | 2020-05-20 | 2020-05-20 | Image enhancement method based on Hess matrix |
Publications (2)
Publication Number | Publication Date |
---|---|
CN111754418A CN111754418A (en) | 2020-10-09 |
CN111754418B true CN111754418B (en) | 2023-09-26 |
Family
ID=72673727
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010430976.8A Active CN111754418B (en) | 2020-05-20 | 2020-05-20 | Image enhancement method based on Hess matrix |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111754418B (en) |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109447936A (en) * | 2018-12-21 | 2019-03-08 | 江苏师范大学 | A kind of infrared and visible light image fusion method |
CN110910323A (en) * | 2019-11-19 | 2020-03-24 | 常州工学院 | Underwater image enhancement method based on self-adaptive fractional order multi-scale entropy fusion |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP3289430B1 (en) * | 2015-04-27 | 2019-10-23 | Snap-Aid Patents Ltd. | Estimating and using relative head pose and camera field-of-view |
-
2020
- 2020-05-20 CN CN202010430976.8A patent/CN111754418B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109447936A (en) * | 2018-12-21 | 2019-03-08 | 江苏师范大学 | A kind of infrared and visible light image fusion method |
CN110910323A (en) * | 2019-11-19 | 2020-03-24 | 常州工学院 | Underwater image enhancement method based on self-adaptive fractional order multi-scale entropy fusion |
Non-Patent Citations (2)
Title |
---|
Image denoising and enhancement based on adaptive fractional calculus of small probability strategy;Li Bo;Neurocomputing;全文 * |
基于图像增强与多特征的森林火灾烟雾检测算法;朱磊;《消防科学与技术》;全文 * |
Also Published As
Publication number | Publication date |
---|---|
CN111754418A (en) | 2020-10-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN104156921B (en) | Self-adaptive low-illuminance or non-uniform-brightness image enhancement method | |
WO2022199583A1 (en) | Image processing method and apparatus, computer device, and storage medium | |
WO2020124873A1 (en) | Image processing method | |
CN109919859B (en) | Outdoor scene image defogging enhancement method, computing device and storage medium thereof | |
CN104680500A (en) | Image intensification algorithm based on histogram equalization | |
CN110298792B (en) | Low-illumination image enhancement and denoising method, system and computer equipment | |
CN111105371A (en) | Low-contrast infrared image enhancement method | |
CN108305232A (en) | A kind of single frames high dynamic range images generation method | |
Srinivas et al. | Remote sensing image segmentation using OTSU algorithm | |
Kaur et al. | Image enhancement techniques: A selected review | |
Wei et al. | An image fusion dehazing algorithm based on dark channel prior and retinex | |
Kim | Image enhancement using patch-based principal energy analysis | |
CN113592750A (en) | Infrared enhancement method based on gradient histogram | |
Mi et al. | A generalized enhancement framework for hazy images with complex illumination | |
CN111754418B (en) | Image enhancement method based on Hess matrix | |
Yang et al. | Improved retinex image enhancement algorithm based on bilateral filtering | |
Cheng et al. | Image quality analysis of a novel histogram equalization method for image contrast enhancement | |
CN112308793A (en) | Novel method for enhancing contrast and detail of non-uniform illumination image | |
CN114429426B (en) | Low-illumination image quality improvement method based on Retinex model | |
CN110992287A (en) | Method for clarifying non-uniform illumination video | |
CN116452447A (en) | Low-illumination high-definition image processing method | |
CN112991361B (en) | Image segmentation method based on local graph structure similarity | |
CN112766338B (en) | Method, system and computer readable storage medium for calculating distance image | |
CN112581400B (en) | Gaussian standard deviation and contrast ratio based tuned image enhancement method | |
Fu et al. | A new image enhancement algorithm for low illumination 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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |