CN117218026B - Infrared image enhancement method and device - Google Patents
Infrared image enhancement method and device Download PDFInfo
- Publication number
- CN117218026B CN117218026B CN202311202900.XA CN202311202900A CN117218026B CN 117218026 B CN117218026 B CN 117218026B CN 202311202900 A CN202311202900 A CN 202311202900A CN 117218026 B CN117218026 B CN 117218026B
- Authority
- CN
- China
- Prior art keywords
- image
- low
- pass filter
- infrared
- infrared 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
- 238000000034 method Methods 0.000 title claims abstract description 43
- 238000005286 illumination Methods 0.000 claims abstract description 86
- 238000001914 filtration Methods 0.000 claims abstract description 75
- 238000012545 processing Methods 0.000 claims abstract description 72
- 238000012937 correction Methods 0.000 claims abstract description 14
- 238000006243 chemical reaction Methods 0.000 claims description 10
- 230000002708 enhancing effect Effects 0.000 claims description 9
- 230000002776 aggregation Effects 0.000 claims description 6
- 238000004220 aggregation Methods 0.000 claims description 6
- 230000009466 transformation Effects 0.000 claims description 5
- 238000007499 fusion processing Methods 0.000 claims description 4
- 238000012886 linear function Methods 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 238000004321 preservation Methods 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 18
- 230000000694 effects Effects 0.000 description 6
- 230000003044 adaptive effect Effects 0.000 description 5
- 230000009286 beneficial effect Effects 0.000 description 5
- 238000003331 infrared imaging Methods 0.000 description 5
- 230000001186 cumulative effect Effects 0.000 description 4
- 238000005315 distribution function Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 238000001514 detection method Methods 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 238000011160 research Methods 0.000 description 3
- 230000002146 bilateral effect Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000012935 Averaging Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000014759 maintenance of location Effects 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000003062 neural network model Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
Landscapes
- Image Processing (AREA)
Abstract
The invention discloses an infrared image enhancement method and device, wherein the method comprises the following steps: acquiring a low-pass filter scale parameter set and a weighting coefficient set; processing the original infrared image by using a preset sequential guided filtering image enhancement model based on the low-pass filter scale parameter set to obtain a primary enhanced infrared image set; and processing the primary enhanced infrared image set according to the weighting coefficient set to obtain a final infrared enhanced image. The invention enhances the contrast of the image by carrying out low-pass filtering and self-adaptive gamma correction processing based on a histogram on the illumination image; guiding and filtering the reflected image to enhance image details, edges and texture information; and finally, fusing the enhanced images of all scales to obtain a final result. Therefore, the technical scheme provided by the invention effectively realizes the noise filtering, contrast enhancement and edge and detail preservation of the infrared image.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to an infrared image enhancement method and device.
Background
The infrared imaging device has the characteristic of all-weather operation, and is widely applied to the scenes such as monitoring, detection, early warning and the like. In various infrared imaging devices, the airborne long-focus infrared camera has long focal length and wide viewing angle, can capture target scene images in a long distance and a large range, and has very important application value. However, due to the limitations of the detection distance and the infrared imaging detection mechanism, the airborne long-focus infrared camera is more easily affected by the environmental factors of the transmission path, the signal attenuation is strong, and the received infrared thermal radiation signal is far weaker than that of the near-distance infrared imaging equipment, so that the generated infrared image has more noise, low contrast, insufficient definition and weak and small targets are easily covered. The infrared image enhancement processing technology is an important means for improving the quality of infrared images and enhancing the usability of the images. However, since the airborne long-focus infrared camera is not used in a large scale in society, most of the documents currently mainly research the image enhancement processing technology of the near-distance infrared imaging equipment, and the research on the image enhancement processing technology of the airborne long-focus infrared camera is little.
At present, common image enhancement algorithms comprise algorithms based on histogram equalization, and the algorithms can effectively enhance the image contrast, but can cause a great deal of loss of detail information of an airborne long-focus infrared camera due to simple cutting-off and merging of image gray levels. In addition, there is image enhancement algorithm based on bilateral filtering, the remarkable characteristic of this kind of algorithm is to decompose the image signal into basic layer and detail layer, carry on smooth filtering treatment to the basic layer with different algorithms separately, carry on enhancement treatment to the detail layer, fuse the processing result finally, this kind of processing tactics try to keep image detail and marginal information while filtering the image noise, enhancing the contrast. Recently, image enhancement processing algorithms based on guided filtering have received a lot of attention, and most researches show that the effect of guided filtering is better than that of bilateral filtering. In guided filtering, the guided image plays an important role, which provides critical information about the image, guiding which pixels of the filter should be filtered out, which pixels should be preserved, and which pixels should be enhanced. However, all the existing image processing algorithms based on the guided filtering use the original image as a guided image, the original image is an image to be processed, noise is included, detail features are not obvious, guiding can be performed wrongly according to the guided filtering principle, the image noise is unavoidably amplified, and meanwhile, the enhancement of the detail information of the image is insufficient.
Disclosure of Invention
The invention aims at solving the technical problems that the on-board long-focus infrared camera has the characteristics of large image dynamic range, low contrast, weak target signal and strong noise, and provides the infrared image enhancement method and the device which are efficient and practical, thereby being beneficial to enhancing key detail information, reducing noise and improving image quality in the infrared image enhancement processing.
In order to solve the above technical problems, a first aspect of an embodiment of the present invention discloses an infrared image enhancement method, which includes:
s1, acquiring a low-pass filter scale parameter set and a weighting coefficient set; the low-pass filter scale parameter set comprises S groups of low-pass filter scale parameters; the low-pass filter scale parameters in the low-pass filter scale parameter set are sequentially reduced; the weighting coefficient set comprises S weighting coefficients; the weighting coefficient in the weighting coefficient set corresponds to the low-pass filter scale parameter in the low-pass filter scale parameter set;
s2, processing an original infrared image by using a preset sequential guided filtering image enhancement model based on the low-pass filter scale parameter set to obtain a primary enhanced infrared image set; the preset sequential guide filtering image enhancement model comprises an illumination image generation unit, a reflection image generation unit and a guide filtering unit; the original infrared image characterizes the enhanced infrared image;
and S3, processing the primary enhanced infrared image set according to the weighting coefficient set to obtain a final infrared enhanced image.
In a first aspect of the embodiment of the present invention, the processing, based on the low-pass filter scale parameter set, the original infrared image with a preset sequential guided filtered image enhancement model to obtain a primary enhanced infrared image set includes:
s21, sequentially selecting low-pass filter scale parameters from the low-pass filter scale parameter set, and processing an original infrared image by using a preset sequential guided filter image enhancement model to obtain S primary enhanced infrared images; the S represents the number of low-pass filter scale parameters contained in the low-pass filter scale parameter set;
s22, carrying out aggregation processing on the S primary enhanced infrared images to obtain a primary enhanced infrared image set, wherein the sequence of the primary enhanced infrared images in the primary enhanced infrared image set corresponds to the low-pass filter scale parameters in the low-pass filter scale parameter set.
As an optional implementation manner, in the first aspect of the embodiment of the present invention, the sequentially selecting low-pass filter scale parameters from the low-pass filter scale parameter set, and processing the original infrared image by using a preset sequential guided filtered image enhancement model to obtain S primary enhanced infrared images includes:
s211, constructing a pointer variable p; the initial value of the pointer variable p is 1;
s212, selecting a p-th group of low-pass filter scale parameters from the low-pass filter scale parameter set;
s213, processing the original infrared image by using the illumination image generation unit based on the p-th group low-pass filter scale parameters to obtain a p-th illumination image; the illumination image generation unit comprises a low-pass filtering process and an adaptive gamma correction process, and is used for decomposing an illumination image from an original image;
s214, processing the p-th illumination image and the original infrared image by using the reflection image generation unit to obtain a p-th reflection image; the reflected image generating unit comprises logarithmic conversion processing and exponential conversion processing and is used for resolving an illumination image from an original image and the illumination image;
s215, conducting guide filtering processing on the p-th reflected image and the guide image by utilizing the guide filtering unit to obtain a p-th image after guide filtering; when the pointer variable p is 1, the guide image is an original infrared image, and when the pointer variable p is greater than 1, the guide image is a p-1 th primary enhanced infrared image; the guiding filtering unit is used for performing linear transformation on the input reflected image by using the guiding image so as to achieve the purpose of optimizing the reflected image;
s216, carrying out fusion processing on the p-th illumination image and the p-th image after the guide filtering to obtain a p-th primary enhanced infrared image;
s217, updating the pointer variable p to p+1, and judging whether the updated pointer variable p is greater than S or not to obtain a first judgment result;
if the first judgment result is yes, executing a step S22;
if the first judgment result is no, step S212 is executed.
In a first aspect of the embodiment of the present invention, the processing, by the illumination image generating unit, the original infrared image based on the p-th set of low-pass filter scale parameters to obtain a p-th illumination image includes:
s2131, based on the p-th group of low-pass filter scale parameters, performing convolution processing on an original infrared image by using a low-pass filter model to obtain a p-th illumination image estimation result;
the low-pass filtering model is expressed as:
wherein L is p Is the firstp illumination image estimation results Γ p For the low pass filter function at the p-th set of low pass filter scale parameters,the method is a convolution operator, and I is an original infrared image;
the low pass filter function is expressed as:
wherein, Γ p (i, j) is a low-pass filter function under the p-th group of low-pass filter scale parameters, (i, j) is an image pixel point coordinate, Z is a normalization constant, c p The p-th group of low-pass filter scale parameters;
s2132, performing self-adaptive gamma correction processing based on a histogram on the p-th illumination image estimation result to obtain a p-th illumination image.
In a first aspect of the embodiment of the present invention, the performing the histogram-based adaptive gamma correction on the estimation result of the p-th illumination image to obtain the p-th illumination image includes:
s21321, calculating a probability density function of each gray level in the p-th illumination image estimation result:
where p (m) is a probability density function of gray level m, n m For the number of pixels with the gray level of M in the p-th illumination image estimation result, M is the gray level of the p-th illumination image estimation result, and for an 8-bit image, m=2 8 N is the total number of pixels of the p-th illumination image estimation result;
s21322 calculating cumulative distribution function for each gray level
Wherein p (M) is a probability density function of the gray level M, F (M) is a cumulative distribution function of the gray level M, and M is the gray level of the p-th illumination image estimation result;
s21323, correcting the gray value of each pixel point by utilizing the self-adaptive gamma function to obtain a p-th illumination image;
the adaptive gamma function is expressed as:
wherein T (m) is L s And the pixel point with the middle gray level of m is subjected to self-adaptive gamma correction to obtain a gray value.
In a first aspect of the embodiment of the present invention, the processing, by using the reflection image generating unit, the p-th illumination image and the original infrared image to obtain a p-th reflection image includes:
s2141, carrying out logarithmic conversion treatment on the p-th illumination image to obtain a p-th illumination image in a logarithmic domain;
s2142, carrying out logarithmic conversion treatment on the original infrared image to obtain a logarithmic domain original infrared image;
s2143, subtracting the p-th illumination image of the logarithmic domain from the original infrared image of the logarithmic domain to obtain a logarithmic domain image difference value;
s2144, performing exponential transformation on the logarithmic domain image difference value to obtain a p-th reflection image.
As an optional implementation manner, in the first aspect of the embodiment of the present invention, after being processed by the guiding filtering unit, the p-th image after guiding filtering is expressed as a linear function of the p-th reflection image, specifically:
wherein w is k B is a weighting coefficient k To compensate for the coefficient omega k A sliding window with a center point coordinate of k and a radius of r is shown;
solving the parameter w by minimizing the cost function k And b k ：
In the formula, lambda is a regularization parameter,for regularization parameter correction factors, measure the homogeneity of the image region in the filter sliding window, μ k 、/>Respectively guide images G p Mean and variance of pixels in sliding window, +.>For the original infrared image R p Gray scale average of pixels in sliding window, |omega k The I represents the total number of pixel points in the sliding window;
for first order gradient of image->And second order gradient->Specifically:
wherein, alpha is the proportion of the first-order gradient and the second-order gradient of the image, the value of the alpha depends on the processed infrared image, and when the homogeneous area in the image is dominant, the alpha takes a larger value (more than 0.5); when heterogeneous areas in the image are dominant, α takes a small value (less than 0.5);
the image first order gradient is a gradient energy (energy of gradient) operator:
in the formula g i ，g j The gradient in the direction of coordinates i and j:
g i ＝I(i+1,j)-I(i,j)
g j ＝I(i,j+1)-I(i,j)
the image second order gradient is a laplacian of gaussian (Laplacian of Gaussian) operator:
in the formula, τ is an auxiliary parameter, and LoG operator at pixel point i (= (x, y)) is specifically:
where σ is the standard deviation of the sliding window.
In a first aspect of the embodiment of the present invention, the fusing the p-th illumination image and the p-th image after the guided filtering to obtain a p-th primary enhanced infrared image includes:
and multiplying the p-th illumination image and the p-th image after the guide filtering to obtain a p-th primary enhanced infrared image.
In a first aspect of the embodiment of the present invention, the processing the primary enhanced infrared image set according to the weighting coefficient set to obtain a final infrared enhanced image includes:
and carrying out weighted summation on all primary enhanced infrared images in the primary enhanced infrared image set to obtain a final infrared enhanced image.
A second aspect of an embodiment of the present invention discloses an infrared image enhancement apparatus, the apparatus including:
an acquisition module; the method comprises the steps of acquiring a low-pass filter scale parameter set and a weighting coefficient set; the low-pass filter scale parameter set comprises S groups of low-pass filter scale parameters; the low-pass filter scale parameters in the low-pass filter scale parameter set are sequentially reduced; the weighting coefficient set comprises S weighting coefficients; the weighting coefficient in the weighting coefficient set corresponds to the low-pass filter scale parameter in the low-pass filter scale parameter set;
a primary enhancement module; the method comprises the steps of processing an original infrared image by using a preset sequential guided filtering image enhancement model based on the low-pass filter scale parameter set to obtain a primary enhanced infrared image set; the preset sequential guide filtering image enhancement model comprises an illumination image generation unit, a reflection image generation unit and a guide filtering unit;
a weighting enhancing module; and the primary enhanced infrared image set is processed according to the weighting coefficient set to obtain a final infrared enhanced image.
Compared with the prior art, the embodiment of the invention has the following beneficial effects:
(1) The illumination image obtained after the low-pass filtering of the original infrared image reserves the low-frequency signal component of the image, key detail information in the original image is reserved in the reflection image, the gray level distribution of a histogram of the illumination image is better and more uniform, the illumination image is processed by utilizing the self-adaptive gamma correction based on the histogram, the image noise is not amplified, the key detail information is not lost, and a better image contrast enhancement effect can be obtained.
(2) According to the invention, the reflection image is guided and filtered, and because the reflection image contains key detail information of an original image, such as edges, textures and the like, the guided and filtered enhances the details of the image and simultaneously filters high-frequency noise signals in the reflection image; the method aims at solving the problems that in the prior art, original images are used as guide images, noise is easily amplified, and detail information is not enhanced sufficiently, and by taking a sequential processing thought as a guide image of guide filtering of the next scale, the filtering effect and detail enhancement effect of a guide filter on the image noise are further enhanced.
(3) Aiming at the characteristics of large dynamic range and multiple image scenes of an airborne long-focus infrared image, the invention introduces a plurality of image gradients as regularization parameters of the guided filtering by referring to the idea of neural network model average, so that the guided filtering can automatically adapt to various types of areas in the image, and compared with the prior art, the invention has stronger robustness.
Drawings
FIG. 1 is a schematic flow chart of an infrared image enhancement method according to an embodiment of the present invention;
FIG. 2 is a schematic flow chart of an infrared image enhancement method based on a preset sequential guided filtered image enhancement model according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of an infrared image enhancement device according to an embodiment of the present invention.
Detailed Description
In order to make the present invention better understood by those skilled in the art, the following description will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings, and it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The terms first, second and the like in the description and in the claims and in the above-described figures are used for distinguishing between different objects and not necessarily for describing a sequential or chronological order. Furthermore, the terms "comprise" and "have," as well as any variations thereof, are intended to cover a non-exclusive inclusion. For example, a process, method, apparatus, article, or device that comprises a list of steps or elements is not limited to the list of steps or elements but may, in the alternative, include other steps or elements not expressly listed or inherent to such process, method, article, or device.
Reference herein to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the invention. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. Those of skill in the art will explicitly and implicitly appreciate that the embodiments described herein may be combined with other embodiments.
The invention discloses an infrared image enhancement method and device, which are beneficial to enhancing key detail information of an infrared image, and meanwhile, image noise cannot be amplified, so that a better image contrast enhancement effect can be obtained. The following will describe in detail.
Example 1
Please refer to fig. 1. Fig. 1 is a schematic flow chart of an infrared image enhancement method according to an embodiment of the present invention. The infrared image enhancement method described in fig. 1 is applied to an infrared image enhancement system, such as a local server or a cloud server for infrared image enhancement, which is not limited in the embodiment of the present invention. As shown in fig. 1, the infrared image enhancement method may include the operations of:
s1, acquiring a low-pass filter scale parameter set and a weighting coefficient set;
in the embodiment of the invention, the low-pass filter scale parameter set comprises S groups of low-pass filter scale parameters, and the low-pass filter scale parameters in the low-pass filter scale parameter set are sequentially reduced so as to reduce the smoothness degree of an illumination image and prevent image texture loss and image distortion caused by excessive smoothness; the weighting coefficient set comprises S weighting coefficients, and the weighting coefficients are sequentially increased; the weighting coefficients in the weighting coefficient set correspond to the low-pass filter scale parameters in the low-pass filter scale parameter set.
It should be noted that, the value of S is a positive integer not less than 5, that is, the low-pass filter scale parameter set includes not less than 5 groups of low-pass filter scale parameters, and the user may increase or decrease according to the effect after the actual processing.
S2, processing an original infrared image by using a preset sequential guided filter image enhancement model based on a low-pass filter scale parameter set to obtain a primary enhanced infrared image set;
in the embodiment of the invention, the preset sequential guide filtering image enhancement model comprises an illumination image generation unit, a reflection image generation unit and a guide filtering unit; the original infrared image characterizes the enhanced infrared image.
And S3, processing the primary enhanced infrared image set according to the weighting coefficient set to obtain a final infrared enhanced image.
In the embodiment of the invention, the weighting coefficient of the primary enhanced infrared image of each scale is a constant 1 minus the normalized low-pass filter scale parameter, so as to increase the proportion of the primary enhanced image obtained by subsequent processing in the final generation of the infrared enhanced image.
In an alternative embodiment, as shown in fig. 2, fig. 2 is a schematic flow chart of an infrared image enhancement method based on a preset sequential guided filtered image enhancement model according to an embodiment of the present invention; the processing of the original infrared image by using a preset sequential guided filtered image enhancement model based on the low-pass filter scale parameter set to obtain a primary enhanced infrared image set includes:
s21, sequentially selecting low-pass filter scale parameters from the low-pass filter scale parameter set, and processing an original infrared image by using a preset sequential guided filter image enhancement model to obtain S primary enhanced infrared images; the S represents the number of low-pass filter scale parameters contained in the low-pass filter scale parameter set;
s22, carrying out aggregation processing on the S primary enhanced infrared images to obtain a primary enhanced infrared image set, wherein the sequence of the primary enhanced infrared images in the primary enhanced infrared image set corresponds to the low-pass filter scale parameters in the low-pass filter scale parameter set.
In another optional embodiment, the sequentially selecting low-pass filter scale parameters from the low-pass filter scale parameter set, and processing the original infrared image by using a preset sequential guided filtered image enhancement model to obtain S primary enhanced infrared images includes:
s211, constructing a pointer variable p; the initial value of the pointer variable p is 1;
s212, selecting a p-th group of low-pass filter scale parameters from the low-pass filter scale parameter set;
s213, processing the original infrared image by using the illumination image generation unit based on the p-th group low-pass filter scale parameters to obtain a p-th illumination image;
s214, processing the p-th illumination image and the original infrared image by using the reflection image generation unit to obtain a p-th reflection image; the reflected image generating unit comprises logarithmic conversion processing and exponential conversion processing and is used for resolving an illumination image from an original image and the illumination image;
s215, conducting guide filtering processing on the p-th reflected image and the guide image by utilizing the guide filtering unit to obtain a p-th image after guide filtering; when the pointer variable p is 1, the guide image is an original infrared image, and when the pointer variable p is greater than 1, the guide image is a p-1 th primary enhanced infrared image; the guiding filtering unit is used for performing linear transformation on the input reflected image by using the guiding image so as to achieve the purpose of optimizing the reflected image;
s216, carrying out fusion processing on the p-th illumination image and the p-th image after the guide filtering to obtain a p-th primary enhanced infrared image;
s217, updating the pointer variable p to p+1, and judging whether the updated pointer variable p is greater than S or not to obtain a first judgment result;
if the first judgment result is yes, executing a step S22; that is, when the pointer variable p is greater than S, it is indicated that all of the low-pass filter scale parameters in the set of low-pass filter scale parameters have been traversed.
If the first judgment result is negative, executing step S212; that is, when the pointer variable p is not greater than S, it is indicated that the low-pass filter scale parameter is still unused in the low-pass filter scale parameter set, and the loop execution needs to be continued.
In yet another optional embodiment, the processing the original infrared image with the illumination image generating unit based on the p-th set of low-pass filter scale parameters to obtain a p-th illumination image includes:
s2131, based on the p-th group of low-pass filter scale parameters, performing convolution processing on an original infrared image by using a low-pass filter model to obtain a p-th illumination image estimation result;
the low-pass filtering model is expressed as:
wherein L is p For the p-th illumination image estimation result, Γ p For the low pass filter function at the p-th set of low pass filter scale parameters,the method is a convolution operator, and I is an original infrared image;
the low pass filter function is expressed as:
wherein, Γ p (i, j) is a low-pass filter function under the p-th group of low-pass filter scale parameters, (i, j) is an image pixel point coordinate, Z is a normalization constant, c p The p-th group of low-pass filter scale parameters;
s2132, performing self-adaptive gamma correction processing based on a histogram on the p-th illumination image estimation result to obtain a p-th illumination image.
Therefore, the infrared image enhancement method described by the embodiment of the invention is beneficial to enhancing the image contrast by performing low-pass filtering and self-adaptive gamma correction processing based on a histogram on the original image.
In yet another optional embodiment, the performing the histogram-based adaptive gamma correction on the p-th illumination image estimation result to obtain a p-th illumination image includes:
s21321, calculating a probability density function of each gray level in the p-th illumination image estimation result:
where p (m) is a probability density function of gray level m, n m For the number of pixels with the gray level of M in the p-th illumination image estimation result, M is the gray level of the p-th illumination image estimation result, and for an 8-bit image, m=2 8 N is the total number of pixels of the p-th illumination image estimation result;
s21322 calculating cumulative distribution function for each gray level
Wherein p (M) is a probability density function of the gray level M, F (M) is a cumulative distribution function of the gray level M, and M is the gray level of the p-th illumination image estimation result;
s21323, correcting the gray value of each pixel point by utilizing the self-adaptive gamma function to obtain a p-th illumination image;
the adaptive gamma function is expressed as:
wherein T (m) is L s And the pixel point with the middle gray level of m is subjected to self-adaptive gamma correction to obtain a gray value.
In yet another optional embodiment, the processing, by the reflection image generating unit, the p-th illumination image and the original infrared image to obtain a p-th reflection image includes:
s2141, carrying out logarithmic conversion treatment on the p-th illumination image to obtain a p-th illumination image in a logarithmic domain;
s2142, carrying out logarithmic conversion treatment on the original infrared image to obtain a logarithmic domain original infrared image;
s2143, subtracting the p-th illumination image of the logarithmic domain from the original infrared image of the logarithmic domain to obtain a logarithmic domain image difference value;
s2144, performing exponential transformation on the logarithmic domain image difference value to obtain a p-th reflection image.
In yet another alternative embodiment, after the processing of the guided filtering unit, the p-th image after the guided filtering is represented as a linear function of the p-th reflected image, specifically:
wherein w is k B is a weighting coefficient k To compensate for the coefficient omega k A sliding window with a center point coordinate of k and a radius of r is shown;
solving the parameter w by minimizing the cost function k And b k ：
In the formula, lambda is a regularization parameter,for regularization parameter correction factors, measure the homogeneity of the image region in the filter sliding window, μ k 、/>Respectively guide images G p Mean and variance of pixels in sliding window, +.>For the original infrared image R s Gray scale average of pixels in sliding window, |omega k The I represents the total number of pixel points in the sliding window;
for first order gradient of image->And second order gradient->Specifically:
wherein alpha is the proportion of the first-order gradient and the second-order gradient of the image;
the image first order gradient is a gradient energy (energy of gradient) operator:
in the formula g i ，g j The gradient in the direction of coordinates i and j:
g i ＝I(i+1,j)-I(i,j)
g j ＝I(i,j+1)-I(I,k)
the image second order gradient is a laplacian of gaussian (Laplacian of Gaussian) operator:
in the formula, τ is an auxiliary parameter, and LoG operator at pixel point i (= (x, y)) is specifically:
where σ is the standard deviation of the sliding window.
It should be noted that, the sequential guided filtering disclosed in this embodiment is characterized by referring to the idea of model averaging in a neural network, aiming at the problem that an airborne long-focus infrared image with a high dynamic range generally has rich area scenes, and a single image area information measurement index is difficult to completely adapt, various types of image area information measurement indexes are introduced as regularization parameters so as to enhance the edge and detail retention performance of the filter; meanwhile, the sequential processing thought is used, the enhanced image of the above scale is used as the guiding image of the guiding filtering of the next scale, and noise filtering, contrast enhancement, edge and detail preservation are better realized.
Therefore, the implementation of the infrared image enhancement method described by the embodiment of the invention is beneficial to enhancing the image detail, edge and texture information by guiding and filtering the reflected image.
In yet another optional embodiment, the fusing the p-th illumination image and the p-th image after the guided filtering to obtain a p-th primary enhanced infrared image includes:
and multiplying the p-th illumination image and the p-th image after the guide filtering to obtain a p-th primary enhanced infrared image.
In yet another alternative embodiment, the processing the primary enhanced infrared image set according to the weighting coefficient set to obtain a final infrared enhanced image includes:
and carrying out weighted summation on all primary enhanced infrared images in the primary enhanced infrared image set to obtain a final infrared enhanced image, wherein the method comprises the following steps of:
wherein O represents the final infrared enhanced image, O' p Representing a p-th primary enhanced infrared image in the set of primary enhanced infrared images, S representing a number of low pass filter scale parameters included in the set of low pass filter scale parameters, L' p Representing the p-th illumination image, R' p Representing the image after pilot filtering of the p-th reflected image.
Example two
Referring to fig. 3, fig. 3 is a schematic structural diagram of an infrared image enhancement device according to an embodiment of the invention. The device described in fig. 3 can be applied to an infrared image enhancement system, such as a local server or a cloud server for infrared image enhancement, and the embodiment of the invention is not limited. As shown in fig. 3, the apparatus may include:
an acquisition module 201; the method comprises the steps of acquiring a low-pass filter scale parameter set and a weighting coefficient set; the low-pass filter scale parameter set comprises S groups of low-pass filter scale parameters; the low-pass filter scale parameters in the low-pass filter scale parameter set are sequentially reduced; the weighting coefficient set comprises S weighting coefficients; the weighting coefficient in the weighting coefficient set corresponds to the low-pass filter scale parameter in the low-pass filter scale parameter set;
a primary boost module 202; the method comprises the steps of processing an original infrared image by using a preset sequential guided filtering image enhancement model based on the low-pass filter scale parameter set to obtain a primary enhanced infrared image set; the preset sequential guide filtering image enhancement model comprises an illumination image generation unit, a reflection image generation unit and a guide filtering unit;
a weight enhancement module 203; and the primary enhanced infrared image set is processed according to the weighting coefficient set to obtain a final infrared enhanced image.
It should be noted that, the infrared image enhancement device disclosed in the second embodiment is a product embodiment corresponding to the infrared image enhancement method disclosed in the first embodiment, and specific processing steps and methods are the same, which are not described in detail in the second embodiment.
The apparatus embodiments described above are merely illustrative, in which the modules illustrated as separate components may or may not be physically separate, and the components shown as modules may or may not be physical, i.e., may be located in one place, or may be distributed over multiple network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment. Those of ordinary skill in the art will understand and implement the present invention without undue burden.
From the above detailed description of the embodiments, it will be apparent to those skilled in the art that the embodiments may be implemented by means of software plus necessary general hardware platforms, or of course by means of hardware. Based on such understanding, the foregoing technical solutions may be embodied essentially or in part in the form of a software product that may be stored in a computer-readable storage medium including Read-Only Memory (ROM), random-access Memory (Random Access Memory, RAM), programmable Read-Only Memory (Programmable Read-Only Memory, PROM), erasable programmable Read-Only Memory (Erasable Programmable Read Only Memory, EPROM), one-time programmable Read-Only Memory (OTPROM), electrically erasable programmable Read-Only Memory (EEPROM), compact disc Read-Only Memory (Compact Disc Read-Only Memory, CD-ROM) or other optical disc Memory, magnetic disc Memory, tape Memory, or any other medium that can be used for computer-readable carrying or storing data.
Finally, it should be noted that: the embodiment of the invention discloses an infrared image enhancement method and device, which are disclosed as preferred embodiments of the invention, and are only used for illustrating the technical scheme of the invention, but not limiting the technical scheme; although the invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art will understand that; the technical scheme recorded in the various embodiments can be modified or part of technical features in the technical scheme can be replaced equivalently; such modifications and substitutions do not depart from the spirit and scope of the corresponding technical solutions.
Claims (7)
1. A method of infrared image enhancement, the method comprising:
s1, acquiring a low-pass filter scale parameter set and a weighting coefficient set; the low-pass filter scale parameter set comprises S groups of low-pass filter scale parameters; the low-pass filter scale parameters in the low-pass filter scale parameter set are sequentially reduced; the weighting coefficient set comprises S weighting coefficients; the weighting coefficient in the weighting coefficient set corresponds to the low-pass filter scale parameter in the low-pass filter scale parameter set;
s2, processing an original infrared image by using a preset sequential guided filtering image enhancement model based on the low-pass filter scale parameter set to obtain a primary enhanced infrared image set; the preset sequential guide filtering image enhancement model comprises an illumination image generation unit, a reflection image generation unit and a guide filtering unit; the original infrared image characterizes the enhanced infrared image; specific:
s21, sequentially selecting low-pass filter scale parameters from the low-pass filter scale parameter set, processing an original infrared image by using a preset sequential guided filter image enhancement model to obtain S primary enhanced infrared images, wherein the S primary enhanced infrared images comprise:
s211, constructing a pointer variable p; the initial value of the pointer variable p is 1;
s212, selecting a p-th group of low-pass filter scale parameters from the low-pass filter scale parameter set;
s213, processing the original infrared image by using the illumination image generation unit based on the p-th group low-pass filter scale parameters to obtain a p-th illumination image;
s214, processing the p-th illumination image and the original infrared image by using the reflection image generation unit to obtain a p-th reflection image;
s215, conducting guide filtering processing on the p-th reflected image and the guide image by utilizing the guide filtering unit to obtain a p-th image after guide filtering; when the pointer variable p is 1, the guide image is an original infrared image, and when the pointer variable p is greater than 1, the guide image is a p-1 th primary enhanced infrared image;
s216, carrying out fusion processing on the p-th illumination image and the p-th image after the guide filtering to obtain a p-th primary enhanced infrared image;
s217, updating the pointer variable p to p+1, and judging whether the updated pointer variable p is greater than S or not to obtain a first judgment result;
if the first judgment result is yes, executing a step S22;
if the first judgment result is negative, executing step S212;
s22, carrying out aggregation processing on the S primary enhanced infrared images to obtain a primary enhanced infrared image aggregation;
and S3, processing the primary enhanced infrared image set according to the weighting coefficient set to obtain a final infrared enhanced image.
2. The method according to claim 1, wherein the processing the original infrared image by the illumination image generating unit based on the p-th set of low-pass filter scale parameters to obtain a p-th illumination image includes:
s2131, based on the p-th group of low-pass filter scale parameters, performing convolution processing on an original infrared image by using a low-pass filter model to obtain a p-th illumination image estimation result;
the low-pass filtering model is expressed as:
wherein L is p For the p-th illumination image estimation result, Γ p For the low pass filter function at the p-th set of low pass filter scale parameters,the method is a convolution operator, and I is an original infrared image;
the low pass filter function is expressed as:
wherein, Γ p (i, j) is a low-pass filter function under the p-th group of low-pass filter scale parameters, (i, j) is an image pixel point coordinate, Z is a normalization constant, c p The p-th group of low-pass filter scale parameters;
s2132, performing self-adaptive gamma correction processing based on a histogram on the p-th illumination image estimation result to obtain a p-th illumination image.
3. The method for enhancing an infrared image according to claim 1, wherein said processing the p-th illumination image and the original infrared image by the reflected image generating unit to obtain the p-th reflected image includes:
s2141, carrying out logarithmic conversion treatment on the p-th illumination image to obtain a p-th illumination image in a logarithmic domain;
s2142, carrying out logarithmic conversion treatment on the original infrared image to obtain a logarithmic domain original infrared image;
s2143, subtracting the p-th illumination image of the logarithmic domain from the original infrared image of the logarithmic domain to obtain a logarithmic domain image difference value;
s2144, performing exponential transformation on the logarithmic domain image difference value to obtain a p-th reflection image.
4. The method according to claim 1, wherein after processing by the guide filtering unit, the p-th image after guide filtering is represented as a linear function of the p-th reflected image, specifically:
wherein w is k B is a weighting coefficient k To compensate for the coefficient omega k A sliding window with a center point coordinate of k and a radius of r is shown;
solving the parameter w by minimizing the cost function k And b k ：
In the formula, lambda is a regularization parameter,for regularization parameter correction factors, measure the homogeneity of the image region in the filter sliding window, μ k 、/>Respectively guide images G p Mean and variance of pixels in sliding window, +.>For the original infrared image R p Gray scale average of pixels in sliding window, |omega k The I represents the total number of pixel points in the sliding window;
for first order gradient of image->And second order gradient->Specifically:
wherein alpha is the proportion of the first-order gradient and the second-order gradient of the image;
the image first order gradient is a gradient energy (energy of gradient) operator:
in the formula g i ，g j The gradient in the direction of coordinates i and j:
g i ＝I(i+1,j)-I(i,j)
g j ＝I(i,j+1)-I(i,j)
the image second order gradient is a laplacian of gaussian (Laplacian of Gaussian) operator:
in the formula, τ is an auxiliary parameter, and LoG operator at pixel point i (= (x, y)) is specifically:
where σ is the standard deviation of the sliding window.
5. The method of claim 1, wherein the fusing the p-th illumination image and the p-th image after the guided filtering to obtain a p-th primary enhanced infrared image comprises:
and multiplying the p-th illumination image and the p-th image after the guide filtering to obtain a p-th primary enhanced infrared image.
6. The method of claim 1, wherein processing the primary enhanced infrared image set according to the set of weighting coefficients to obtain a final infrared enhanced image comprises:
and carrying out weighted summation on all primary enhanced infrared images in the primary enhanced infrared image set to obtain a final infrared enhanced image.
7. An infrared image enhancement device, the device comprising:
an acquisition module; the method comprises the steps of acquiring a low-pass filter scale parameter set and a weighting coefficient set; the low-pass filter scale parameter set comprises S groups of low-pass filter scale parameters; the low-pass filter scale parameters in the low-pass filter scale parameter set are sequentially reduced; the weighting coefficient set comprises S weighting coefficients; the weighting coefficient in the weighting coefficient set corresponds to the low-pass filter scale parameter in the low-pass filter scale parameter set;
a primary enhancement module; the method comprises the steps of processing an original infrared image by using a preset sequential guided filtering image enhancement model based on the low-pass filter scale parameter set to obtain a primary enhanced infrared image set; the preset sequential guide filtering image enhancement model comprises an illumination image generation unit, a reflection image generation unit and a guide filtering unit; specific:
s21, sequentially selecting low-pass filter scale parameters from the low-pass filter scale parameter set, processing an original infrared image by using a preset sequential guided filter image enhancement model to obtain S primary enhanced infrared images, wherein the S primary enhanced infrared images comprise:
s211, constructing a pointer variable p; the initial value of the pointer variable p is 1;
s212, selecting a p-th group of low-pass filter scale parameters from the low-pass filter scale parameter set;
s213, processing the original infrared image by using the illumination image generation unit based on the p-th group low-pass filter scale parameters to obtain a p-th illumination image;
s214, processing the p-th illumination image and the original infrared image by using the reflection image generation unit to obtain a p-th reflection image;
s215, conducting guide filtering processing on the p-th reflected image and the guide image by utilizing the guide filtering unit to obtain a p-th image after guide filtering; when the pointer variable p is 1, the guide image is an original infrared image, and when the pointer variable p is greater than 1, the guide image is a p-1 th primary enhanced infrared image;
s216, carrying out fusion processing on the p-th illumination image and the p-th image after the guide filtering to obtain a p-th primary enhanced infrared image;
s217, updating the pointer variable p to p+1, and judging whether the updated pointer variable p is greater than S or not to obtain a first judgment result;
if the first judgment result is yes, executing a step S22;
if the first judgment result is negative, executing step S212;
s22, carrying out aggregation processing on the S primary enhanced infrared images to obtain a primary enhanced infrared image aggregation;
a weighting enhancing module; and the primary enhanced infrared image set is processed according to the weighting coefficient set to obtain a final infrared enhanced image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311202900.XA CN117218026B (en) | 2023-09-18 | 2023-09-18 | Infrared image enhancement method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311202900.XA CN117218026B (en) | 2023-09-18 | 2023-09-18 | Infrared image enhancement method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117218026A CN117218026A (en) | 2023-12-12 |
| CN117218026B true CN117218026B (en) | 2024-03-22 |
Family
ID=89038539
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311202900.XA Active CN117218026B (en) | 2023-09-18 | 2023-09-18 | Infrared image enhancement method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117218026B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117409083B (en) * | 2023-12-14 | 2024-03-22 | 珠海市金锐电力科技有限公司 | Cable terminal identification method and device based on infrared image and improved YOLOV5 |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110728645A (en) * | 2019-10-23 | 2020-01-24 | 华中科技大学鄂州工业技术研究院 | Image detail enhancement method and device based on guide filter regularization parameter and electronic equipment |
| CN111292257A (en) * | 2020-01-15 | 2020-06-16 | 重庆邮电大学 | Retinex-based image enhancement method in dark vision environment |
| CN113096053A (en) * | 2021-03-17 | 2021-07-09 | 西安电子科技大学 | High dynamic infrared image detail enhancement method based on multi-scale guided filtering |
| CN114092353A (en) * | 2021-11-19 | 2022-02-25 | 长春理工大学 | Infrared image enhancement method based on weighted guided filtering |
| CN116071259A (en) * | 2023-02-08 | 2023-05-05 | 重庆邮电大学 | Infrared image enhancement method based on secondary guide filtering |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US7590303B2 (en) * | 2005-09-29 | 2009-09-15 | Samsung Electronics Co., Ltd. | Image enhancement method using local illumination correction |
| US20100142790A1 (en) * | 2008-12-04 | 2010-06-10 | New Medical Co., Ltd. | Image processing method capable of enhancing contrast and reducing noise of digital image and image processing device using same |
-
2023
- 2023-09-18 CN CN202311202900.XA patent/CN117218026B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110728645A (en) * | 2019-10-23 | 2020-01-24 | 华中科技大学鄂州工业技术研究院 | Image detail enhancement method and device based on guide filter regularization parameter and electronic equipment |
| CN111292257A (en) * | 2020-01-15 | 2020-06-16 | 重庆邮电大学 | Retinex-based image enhancement method in dark vision environment |
| CN113096053A (en) * | 2021-03-17 | 2021-07-09 | 西安电子科技大学 | High dynamic infrared image detail enhancement method based on multi-scale guided filtering |
| CN114092353A (en) * | 2021-11-19 | 2022-02-25 | 长春理工大学 | Infrared image enhancement method based on weighted guided filtering |
| CN116071259A (en) * | 2023-02-08 | 2023-05-05 | 重庆邮电大学 | Infrared image enhancement method based on secondary guide filtering |
Non-Patent Citations (2)
| Title |
|---|
| 基于照射_反射模型和有界运算的多谱段图像增强;毕国玲;续志军;赵建;孙强;;物理学报(第10期);第100701-1-100701-9页 * |
| 消除光晕和细节增强的多尺度Retinex红外图像增强;温海滨;毕笃彦;马时平;何林远;;红外技术;第38卷(第02期);第149-155页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117218026A (en) | 2023-12-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Sun et al. | Fast single image haze removal via local atmospheric light veil estimation | |
| CN117218026B (en) | Infrared image enhancement method and device | |
| Zhou et al. | Multi-scale retinex-based adaptive gray-scale transformation method for underwater image enhancement | |
| CN114119436A (en) | Infrared image and visible light image fusion method and device, electronic equipment and medium | |
| Yao et al. | The Retinex-based image dehazing using a particle swarm optimization method | |
| Gao et al. | Single image dehazing using local linear fusion | |
| Sandoub et al. | A low‐light image enhancement method based on bright channel prior and maximum colour channel | |
| CN111311562B (en) | Ambiguity detection method and device for virtual focus image | |
| Li et al. | Image enhancement algorithm based on depth difference and illumination adjustment | |
| Wu et al. | A novel scheme for infrared image enhancement by using weighted least squares filter and fuzzy plateau histogram equalization | |
| CN112508803A (en) | Denoising method and device for three-dimensional point cloud data and storage medium | |
| CN108614998A (en) | A kind of single pixel infrared target detection method | |
| CN109712134B (en) | Iris image quality evaluation method and device and electronic equipment | |
| CN111652821B (en) | Low-light video image noise reduction processing method, device and equipment based on gradient information | |
| CN113962900A (en) | Method, device, equipment and medium for detecting infrared dim target under complex background | |
| CN117649694A (en) | Face detection method, system and device based on image enhancement | |
| CN111311610A (en) | Image segmentation method and terminal equipment | |
| Rama Lakshmi et al. | A Review on Image Denoising Algorithms for Various Applications | |
| CN116862809A (en) | Image enhancement method under low exposure condition | |
| US8311358B2 (en) | Method and system for image extraction and identification | |
| CN116543022A (en) | Gas infrared image processing method, gas leak detector and storage medium | |
| Dong et al. | Content-adaptive noise estimation for color images with cross-channel noise modeling | |
| Zhou et al. | An improved algorithm using weighted guided coefficient and union self‐adaptive image enhancement for single image haze removal | |
| CN115713620A (en) | Infrared small target detection method and device, computing equipment and storage medium | |
| CN115170420A (en) | Image contrast processing method and system |
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 |