CN113870135A - NSST domain infrared image enhancement method based on longicorn stigma optimization algorithm - Google Patents
NSST domain infrared image enhancement method based on longicorn stigma optimization algorithm Download PDFInfo
- Publication number
- CN113870135A CN113870135A CN202111147083.3A CN202111147083A CN113870135A CN 113870135 A CN113870135 A CN 113870135A CN 202111147083 A CN202111147083 A CN 202111147083A CN 113870135 A CN113870135 A CN 113870135A
- Authority
- CN
- China
- Prior art keywords
- image
- algorithm
- frequency component
- low
- 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.)
- Pending
Links
- 238000004422 calculation algorithm Methods 0.000 title claims abstract description 120
- 238000000034 method Methods 0.000 title claims abstract description 67
- 241001481710 Cerambycidae Species 0.000 title claims abstract description 46
- 238000005457 optimization Methods 0.000 title claims abstract description 39
- 230000009466 transformation Effects 0.000 claims abstract description 46
- 238000012545 processing Methods 0.000 claims abstract description 42
- 230000002708 enhancing effect Effects 0.000 claims abstract description 15
- 230000011218 segmentation Effects 0.000 claims abstract description 12
- 238000000354 decomposition reaction Methods 0.000 claims description 18
- 230000008569 process Effects 0.000 claims description 18
- 238000004364 calculation method Methods 0.000 claims description 13
- 230000000694 effects Effects 0.000 claims description 12
- 238000003064 k means clustering Methods 0.000 claims description 12
- 238000005315 distribution function Methods 0.000 claims description 9
- 230000004807 localization Effects 0.000 claims description 9
- 238000010845 search algorithm Methods 0.000 claims description 8
- 230000000007 visual effect Effects 0.000 claims description 8
- 238000006243 chemical reaction Methods 0.000 claims description 6
- 238000013507 mapping Methods 0.000 claims description 6
- 238000003709 image segmentation Methods 0.000 claims description 5
- 230000015572 biosynthetic process Effects 0.000 claims description 4
- 238000003672 processing method Methods 0.000 claims description 4
- 238000003786 synthesis reaction Methods 0.000 claims description 4
- 230000009471 action Effects 0.000 claims description 3
- 230000004927 fusion Effects 0.000 claims description 3
- 238000009827 uniform distribution Methods 0.000 claims description 3
- 230000000717 retained effect Effects 0.000 claims description 2
- 239000000126 substance Substances 0.000 claims description 2
- 238000001914 filtration Methods 0.000 abstract description 13
- 230000007547 defect Effects 0.000 description 6
- 238000001514 detection method Methods 0.000 description 6
- 238000004458 analytical method Methods 0.000 description 5
- 239000002131 composite material Substances 0.000 description 5
- 230000002159 abnormal effect Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000011156 evaluation Methods 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 235000013929 Psidium pyriferum Nutrition 0.000 description 3
- 244000236580 Psidium pyriferum Species 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 3
- 235000013305 food Nutrition 0.000 description 3
- 239000011159 matrix material Substances 0.000 description 3
- 230000005855 radiation Effects 0.000 description 3
- 238000009826 distribution Methods 0.000 description 2
- 238000003708 edge detection Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000002474 experimental method Methods 0.000 description 2
- 230000001965 increasing effect Effects 0.000 description 2
- 238000003331 infrared imaging Methods 0.000 description 2
- 239000002184 metal Substances 0.000 description 2
- 238000004800 variational method Methods 0.000 description 2
- 238000012795 verification Methods 0.000 description 2
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 description 1
- 241000282461 Canis lupus Species 0.000 description 1
- 241000283153 Cetacea Species 0.000 description 1
- 230000003321 amplification Effects 0.000 description 1
- 230000002238 attenuated effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002146 bilateral effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000009529 body temperature measurement Methods 0.000 description 1
- 230000000739 chaotic effect Effects 0.000 description 1
- 238000010835 comparative analysis Methods 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 230000004069 differentiation Effects 0.000 description 1
- 230000002349 favourable effect Effects 0.000 description 1
- 230000002431 foraging effect Effects 0.000 description 1
- 238000009499 grossing Methods 0.000 description 1
- 210000003128 head Anatomy 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000012212 insulator Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000003199 nucleic acid amplification method Methods 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 150000003839 salts Chemical class 0.000 description 1
- 238000010008 shearing Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
- G06F18/23213—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions with fixed number of clusters, e.g. K-means clustering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/004—Artificial life, i.e. computing arrangements simulating life
- G06N3/006—Artificial life, i.e. computing arrangements simulating life based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/40—Image enhancement or restoration using histogram techniques
-
- 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
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/194—Segmentation; Edge detection involving foreground-background segmentation
-
- 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
- G06T2207/10048—Infrared image
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Computational Linguistics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Probability & Statistics with Applications (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Image Processing (AREA)
Abstract
The invention discloses an NSST domain infrared image enhancement method based on a longicorn stigma optimization algorithm, which combines an improved longicorn stigma search optimization algorithm, an SANS filtering algorithm, a fuzzy enhancement algorithm and the like for the first time; firstly, decomposing an infrared image into a high-frequency part and a low-frequency part by using NSST (non-subsampled contourlet transform); carrying out threshold segmentation on low-frequency components containing a large amount of target equipment information by using an improved longicorn algorithm to decompose the low-frequency components into a background area and a target area, and then respectively carrying out enhancement processing; for high-frequency components containing noise and image detail information, proper parameters are selected to carry out SANS filtering denoising, and then fuzzy enhancement is carried out. Then enhancing the processed high-frequency sub-band image; finally, NSST inverse transformation is carried out on the processed low-frequency component and the processed high-frequency component to obtain a final enhanced image; the method can remove noise of the infrared image, improve the edge and detail information of the power equipment part in the infrared image, and improve the overall contrast of the infrared image gray scale image of the power equipment.
Description
Technical Field
The invention relates to a NSST domain infrared image enhancement method based on a longicorn silk optimization algorithm.
Background
The normal operation of the power equipment ensures the stable operation of the whole power system, and the abnormal operation or the fault of the power equipment is accompanied with the temperature rise. The infrared detection technology can display infrared rays invisible to human eyes in a temperature spectrum mode. And carrying out real-time, non-contact and nondestructive detection on the power equipment under the condition that the power equipment is not stopped when the power system is operated. The infrared detection is widely applied to the temperature detection of the power equipment, and aims to overcome the defect of equipment failure in advance and further eliminate the further damage to the whole power system. The infrared temperature measurement technology is applied to the detection of the running state of the power equipment, and has very important application value. At the present stage, advanced foreign infrared detection equipment is actively introduced into a plurality of power companies, a plurality of power equipment is patrolled and examined, and important experimental data are collected and accumulated through a large amount of practices.
Due to the limitation of the infrared focal plane production process, weak signals are difficult to distinguish in the photoelectric conversion process, so that the presented infrared image is blurred. The power equipment is of various types, the structure is complex, the metal connecting parts between the electrical equipment, insulators, temperature abnormal parts such as joints are compared with a transformer and a tower pole and are difficult to distinguish, the limitation of the infrared imaging hardware production process at the current stage enables the infrared image identification degree of the power equipment not to be high, in addition, the infrared radiation of an object received by an infrared sensor is subjected to heat transfer, the influence of external factors such as atmospheric attenuation and heat radiation causes the infrared image to have the problems of low resolution, fuzzy details, various noises and the like, and the analysis and fault location of the abnormal reasons of the power equipment are difficult to carry out. Due to the defects of the current infrared sensor manufacturing process, the micro signals are difficult to distinguish in the imaging process, the infrared ray is attenuated in the transmission process, and the influence of atmosphere and environment heat radiation causes the infrared image to contain noise, blur and low contrast, so that the quality and visual effect of the infrared image of the electrical equipment are further influenced.
Partial image enhancement and edge detection are problems in digital image processing for infrared imaging of power devices. The image enhancement processing needs to denoise the image, improve the contrast, and improve the texture and edge details of the target object.
The traditional infrared image enhancement processing method mainly comprises the following steps:
1. spatial domain enhancement algorithm
(1) Linear enhancement algorithm
The reason why the infrared image contrast is low is that the gray scale is concentrated in a region with a narrow dynamic range, and a linear enhancement method is used to expand the gray scale range.
The boundaries of the linear transformation should first be determined from the gray histogram. The extent of the concentrated gray scale region can now be determined by finding the maximum gray level of the gray histogram. The left and right boundaries of the linear enhancement can be determined by the maximum gray value, and the left boundary is set as aLThe right boundary is aR。
After the left and right boundaries are determined, the low frequency foreground portion can be enhanced in advance, and if z (x, y) represents the enhanced image, the calculation formula is:
the method is favorable for maintaining the gray part distribution rule of the original electric power infrared image, and the gray value boundary range [ a ] is obtained through calculationL,aR]And the existing pixel points are subjected to linear enhancement, so that the visual effect of the power equipment area is improved, the effect that each part does not influence each other and independent observation is realized.
(2) Histogram equalization
The histogram equalization algorithm is based on the statistical principle, and expands the range of an image gray domain by counting the distribution function of the image pixel gray value so as to achieve the purposes of enhancing the image contrast, increasing the information entropy and ensuring the visual effect to be clearer. The image gray value is a discrete variable, and the probability density with the gray value x is:
n is the total amount of pixels; n islThe number of pixels with the gray value of l; the gray scale distribution function of the image is:
the histogram equalization algorithm multiplies the gray distribution function by (L-1) as a new pixel value, and the conversion relation is as follows:
the histogram equalization method is used for processing the low-frequency and low-temperature components, the background brightness can be effectively reduced, the target pixel value of the main part of the non-electric equipment is highlighted, the gray difference between the part and the background is increased, and therefore the identification degree of the original image is improved.
2. Denoising algorithm
In an image, a factor that prevents people from receiving information is called noise. The image denoising is classified into spatial filtering, frequency domain filtering, partial differential equation denoising and variational methods. The spatial filtering method is to calculate the original image matrix and calculate and select proper pixel values to replace the original noise gray value. Common space domain denoising algorithms include domain filtering, low-pass filtering and median filtering; the frequency domain filtering refers to transforming the original image from a space domain to a frequency domain, processing a transform coefficient of the frequency domain, and finally performing inverse transformation to achieve the purpose of denoising. Common frequency domain processing includes fourier transform, wavelet transform, and the like; the partial differential equation denoising refers to establishing a partial differential equation for the original noise-containing image and solving a nonlinear partial differential equation in the partial differential equation; the variable distribution denoising refers to determining an image energy function, and enabling an image to reach a smooth state through minimization processing on the energy function.
3. Traditional frequency domain enhancement algorithm
The frequency domain enhancement is one of the most widely used methods for image enhancement at present, the frequency domain processing is developed from Fourier transform, the Fourier transform represents general signals of an image by sinusoidal signals with different frequencies, but the Fourier transform only contains frequency information and does not contain time information, and the complete information of one image is difficult to embody. The wavelet transformation improves the basic waveform of Fourier transformation, so that the transformation result contains frequency and time parameters, but the properties of all directions of the wavelet transformation are the same, and the anisotropy of the image cannot be ensured.
However, these conventional infrared image enhancement methods have various disadvantages:
the conventional linear enhancement has the following disadvantages: the interference signal is mixed into the live-action information to form noise under the influence of the infrared focal plane manufacturing process and external factors. The linear enhancement processes the whole pixel value of the image, and the real scene information and the noise information are not distinguished in the processing process. The linear algorithm processing can enhance the contrast between an object and the background, and corresponding noise information is correspondingly amplified, so that the quality and the visual effect of the infrared image of the power equipment are reduced, and more interference is caused to the abnormal analysis and fault judgment of the power equipment.
The traditional denoising algorithm has the following defects: comparing various traditional denoising algorithms can be obtained. The image denoising calculation amount based on the airspace is small, the image details can be well kept, but the complex filtering cannot be completely filtered, and the method is only suitable for the image with simple noise information; the frequency domain denoising is not limited by noise characteristics, but the calculation amount in the transformation process is large, and the operation time is too long. Partial differentiation and variational methods can preserve image edge information but are deficient in their ability to distinguish between edge information and noise.
The traditional frequency domain denoising algorithm has the following defects: the early Fourier transform does not contain time information, and the integrity of the image cannot be presented; the wavelet transform ignores the image anisotropy on the basis of Fourier transform; the shear wave transformation considers the directionality of the image, can realize the optimal sparse representation of the image, has good effects of edge detection and image enhancement in the infrared image of the power equipment, but the downsampling operation of scale transformation and direction transformation is easy to cause the pseudo Gibbs phenomenon of the image.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provide an NSST domain infrared image enhancement method based on a longicorn whisker optimization algorithm, which can remove noise of an infrared image, improve the edge and detail information of a power equipment part and improve the overall contrast of an infrared image gray scale map of the power equipment.
The technical scheme for realizing the purpose is as follows: an NSST domain infrared image enhancement method based on a longicorn whisker optimization algorithm comprises the following steps:
s1, performing multi-scale and multi-directional transformation on the original power equipment infrared image by adopting NSST transformation, and decomposing the original power equipment infrared image into a high-frequency component and a low-frequency component;
s2, performing threshold segmentation on the low-frequency component obtained in the step S1 by adopting a longicorn whisker optimization algorithm, and segmenting the low-frequency component into a background region and a foreground region of the power equipment main body;
s3, performing linear enhancement on the foreground region obtained by segmentation in the step S2, performing histogram equalization processing on the background region, and performing synthesis processing on the foreground region and the background region to obtain an enhanced low-frequency component image;
s4, processing the high-frequency component obtained in the step S1 by adopting an SANS algorithm, eliminating vibration and noise of the high-frequency component, and enhancing the contrast and definition of the image of the high-frequency component by using a Pal-King algorithm to obtain an enhanced high-frequency component image;
and S5, performing NSST inverse transformation on the enhanced high-frequency component image and the enhanced low-frequency component image to obtain an enhanced infrared image of the power equipment.
In the above NSST domain infrared image enhancement method based on the longicorn whisker optimization algorithm, in step S1, the multi-scale and multi-directional transformation includes two processes of multi-scale decomposition and directional localization, wherein:
in the multi-scale decomposition process, an original power equipment infrared image is decomposed into a high-frequency image and a low-frequency image through a non-downsampling pyramid filter bank, NSP decomposition is carried out on the basis of the low-frequency image obtained through decomposition, and k high-frequency sub-bands and 1 low-frequency sub-band image are obtained after k times of NSP decomposition;
the method comprises the following steps of a directional localization process, wherein NSST utilizes improved shear wave transformation to perform localization processing on high-frequency sub-band images and low-frequency sub-band images obtained by multi-scale decomposition, firstly, a standard shear wave filter is mapped to Cartesian coordinates from a pseudo-polar coordinate system, and inverse Fourier transformation is performed; and finally, finishing final processing by utilizing two-dimensional convolution, and decomposing the infrared image of the original power equipment into a high-frequency component and a low-frequency component.
In step S2, the n & ltth & gt region infrared image enhancement method based on the longicorn silk optimization algorithm is an image segmentation processing method formed by combining a self-adaptive longicorn silk search algorithm and a K-means clustering algorithm, and includes the following steps:
s21, solving a global optimal solution of the image data point set of the low-frequency component by adopting a self-adaptive longicorn whisker search algorithm;
s22, taking the global optimal solution obtained in the step S21 as an initial clustering center of a K-means clustering algorithm;
s23, randomly generating K clustering centers, assigning elements in the image data point set to the K classes, and calculating the distance f from each image data point in the image data point set to the clustering centers, wherein the calculation formula is as follows:
Xiis the ith point in the image data point set X, and YjAnd for the jth clustering center, updating the clustering center through the mean value, carrying out algorithm convergence, obtaining the minimum fitness value, and obtaining the optimal initial clustering center of the K-means clustering algorithm.
In the NSST domain infrared image enhancement method based on the longicorn whisker optimization algorithm, in step S3, the process of linearly enhancing the foreground region of the low-frequency component is as follows:
determining a boundary of linear enhancement according to the gray histogram, and determining the range of a concentrated gray area by searching the maximum gray level of the gray histogram; determining left and right boundaries of linear enhancement by maximum gray value, and setting the left boundary as aLThe right boundary is aR；
After the left and right boundaries are determined, the foreground region of the low-frequency component is linearly enhanced, the enhanced image is represented by z (x, y), and the calculation formula is as follows:
by calculating the resulting gray value boundary range [ a ]L,aR]The existing pixel points are subjected to linear enhancement, and the visual effect of the foreground area of the infrared image of the power equipment is improved;
the histogram equalization processing for the background region of the low-frequency component is performed as follows:
the probability density of occurrence of a gray value of l is:
n is the total amount of pixels; n islThe number of pixels with the gray value of l;
the gray scale distribution function of the image is:
the histogram equalization algorithm multiplies the gray distribution function by (L-1) as a new pixel value, and the conversion relation is as follows:
the low-frequency component subjected to histogram equalization processing reduces the background brightness, highlights the target pixel value of the main part of the non-electric equipment, increases the gray difference between the part and the background, and improves the identification degree of the original image.
In step S4, the SANS algorithm fuses the input blurred image and the restored image by using the local variance of the input blurred image, and a fusion equation and scale parameters calculated according to the local variance are as follows:
wherein the content of the first and second substances,
is the final restored image; alpha (x, y) is a fused weight coefficient, beta is a coefficient which ensures the uniform distribution of alpha between 0 and 1, the smaller the beta value is, the more remarkable the effect of derringing is, and v (x, y) is the local variance of the input image g (x, y);
after the high-frequency component is processed by the SANS algorithm, the edge details of the high-frequency component are retained while vibration and noise are removed;
the Pal-King algorithm firstly maps the high-frequency component processed by the SANS algorithm from the original space domain to the fuzzy domain, and the classical Pal-King algorithm fuzzy membership function muijComprises the following steps:
wherein, tijRefers to the gray value of the image (i, j); l denotes the gray level; kdIs a fuzzy parameter at the denominator; keIs an exponential fuzzy parameter; keAnd KdDetermining a fuzzy membership mapping rule by the combined action;
modifying a fuzzy membership function of a classic Pal-King algorithm:
wherein, tijDenotes the gray value, t, of the image (i, j)minRefers to the minimum gray value in the image of the high frequency component;
tmaxrefers to the maximum gray value in the image of the high frequency component;
and further enhancing the fuzzy membership degree after mapping, and performing flat S-type processing by adopting S-type transformation, wherein the S-type transformation formula is as follows:
μijthe fuzzy membership function after S-type transformation is enhanced by the steps and transformed to [0, L-1 according to the inverse function of the fuzzy membership]Within the range, obtaining an enhanced gray value t'ijNamely:
improved fuzzy membership range from [ mu ]min 1]Expand to [ 01]。
The NSST domain infrared image enhancement method based on the longicorn whisker optimization algorithm can remove noise of an infrared image, improve the edge and detail information of a power equipment part, and improve the overall contrast of an infrared image gray-scale image of the power equipment, and has the following beneficial effects compared with the prior art:
(1) the longicorn stigma search algorithm is improved;
(2) improving a linear enhancement algorithm of a low-frequency component foreground region and a fuzzy enhancement algorithm of a background region;
(3) firstly, an improved longicorn stigma search optimization algorithm, an SANS filtering algorithm, a fuzzy enhancement algorithm and the like are combined, and a new infrared image enhancement method is provided; firstly, decomposing an infrared image into a high-frequency part and a low-frequency part by using NSST (non-subsampled contourlet transform); carrying out threshold segmentation on low-frequency components containing a large amount of target equipment information by using an improved longicorn algorithm to decompose the low-frequency components into a background area and a target area, and then respectively carrying out enhancement processing; for high-frequency components containing noise and image detail information, proper parameters are selected to carry out SANS filtering denoising, and then fuzzy enhancement is carried out. Then enhancing the processed high-frequency sub-band image; finally, NSST inverse transformation is carried out on the processed low-frequency component and the processed high-frequency component to obtain a final enhanced image; and the reliability of the method is proved by comparison and verification through experiments.
Drawings
FIG. 1 is a flow chart of the NSST domain infrared image enhancement method based on the longicorn silk optimization algorithm of the present invention;
FIG. 2 is a schematic diagram of the frequency domain dominance of a shear wave transform and the size of the dominance set;
FIG. 3 is a flow chart of an implementation of a longicorn stigma optimization algorithm;
FIG. 4a is an effect diagram of threshold segmentation of an infrared image of a telegraph pole by adopting a longicorn whisker optimization algorithm;
FIG. 4b is a graph showing the effect of the prior art Ostu algorithm on infrared image segmentation of a utility pole;
fig. 5 is a diagram illustrating the effect of processing the low-frequency component of the infrared image of the utility pole in step S3.
Detailed Description
In order that those skilled in the art will better understand the technical solution of the present invention, the following detailed description is given with reference to the accompanying drawings:
referring to fig. 1, in the preferred embodiment of the present invention, a NSST domain infrared image enhancement method based on a longicorn whisker optimization algorithm includes the following steps:
s1, performing multi-scale and multi-directional transformation on the original power equipment infrared image by adopting NSST transformation, and decomposing the original power equipment infrared image into a high-frequency component and a low-frequency component;
s2, performing threshold segmentation on the low-frequency component obtained in step S1 by using a longicorn whisker optimization algorithm, and segmenting the low-frequency component into two parts, namely a background region and a foreground region (also referred to as a device region or a target region) of the power device main body;
s3, performing linear enhancement on the foreground region obtained by segmentation in the step S2, performing histogram equalization processing on the background region, and performing synthesis processing on the foreground region subjected to linear enhancement and the background region subjected to histogram equalization processing to obtain an enhanced low-frequency component image;
s4, processing the high-frequency component obtained in the step S1 by adopting an SANS algorithm, eliminating vibration and noise of the high-frequency component, and enhancing the contrast and definition of the image of the high-frequency component by using a Pal-King algorithm to obtain an enhanced high-frequency component image;
and S5, performing NSST inverse transformation on the enhanced high-frequency component image and the enhanced low-frequency component image to obtain an enhanced infrared image of the power equipment.
NSST domain image enhancement:
referring to fig. 2, Guo, Labate et al propose a composite wavelet by combining geometric analysis and multiresolution analysis theory, when the dimension n is 2, the composite dilatant affine system is:
in the formula (2-1),
l is a integrable space, R represents a collection and Z represents an integer collection. The moments a and B are both 2 × 2 invertible matrices, and | detB | ═ 1. Suppose for an arbitrary function f ∈ L2(R2) If, if
All satisfy Parseval framework (tight support condition), i.e.
In the formula (2-2), the metal salt,
representing a composite wavelet. Matrix a controls the scaling and matrix B controls the geometric orientation transformation (translation, scaling, shearing, etc.). Thus can let
A Parseval framework is constructed in each direction and scale. When in use
The composite wavelet is a shear wave. In general, a is 4 and b is 1. Shear waves are in fact a special case of composite wavelets. Fig. 2 shows a schematic diagram of the frequency domain contribution of the shear wave transform and the size of the contribution set.
In step S1, the multi-scale and multi-directional transformation includes two processes of multi-scale decomposition and directional localization, where:
a multi-scale decomposition process, namely decomposing an original power equipment infrared image into a high-frequency image and a low-frequency image through a Non-subsampled Pyramid (NSP) filter bank, performing NSP decomposition on the low-frequency image obtained based on the decomposition, and obtaining k high-frequency sub-bands and 1 low-frequency sub-band image after k times of NSP decomposition;
a directional localization process, namely, NSST utilizes improved shear wave transformation to perform localization processing on high-frequency sub-band images and low-frequency sub-band images obtained by multi-scale decomposition, firstly, a standard shear wave (Shearlet) filter is mapped to Cartesian coordinates from a pseudo polar coordinate system, and inverse Fourier transform is performed; and finally, finishing final processing by utilizing two-dimensional convolution, and decomposing the infrared image of the original power equipment into a high-frequency component and a low-frequency component.
And (3) low-frequency component processing:
the standard longhorn searching algorithm is a novel intelligent optimization algorithm based on simulation of foraging longhorns through food smell. The longicorn judges the advancing direction of food smell in the air through beard at two sides of the head, so that the longicorn finds the food. Compared with a Particle Swarm Optimization (PSO), the algorithm only needs at least one longicorn, and the calculation complexity of the algorithm is reduced. The search algorithm in the celestial cow sequence table is improved, the fixed step length algorithm is optimized, the step length is changed from large to small by changing the step length calculation formula and adding the variable step length factor, the approximate solution area is searched in a large range in the search process, then detailed search is carried out, and the precision and the convergence speed are improved.
In step S2, the longicorn stigma optimization algorithm is a new image segmentation processing method formed by combining an adaptive vertex search algorithm (ABAS) and a K-means (K-means) clustering algorithm. Refers to a K-means clustering algorithm based on A BAS. The algorithm consists of two parts
(1) Solving a global optimal solution for a set of image points using ABAS
(2) The output solution of ABAS is applied to the K-means clustering algorithm. By optimizing the initial clustering center, the calculation effect of the K-means clustering algorithm is improved, and the defect that the K-means clustering algorithm is sensitive to the initial clustering center is avoided.
Referring to fig. 3, the longicorn whisker optimization algorithm includes the following steps:
s21, solving the global optimal solution of the image data point set of the low-frequency component by adopting a self-adaptive longicorn whisker search algorithm, firstly calculating the left and right coordinates of the longicorn whisker, calculating the odor intensity (fitness value), calculating the position where the longicorn is moved next according to a variable step length method, and outputting the global optimal solution of the image data point set of the low-frequency component after the iteration times are reached;
s22, taking the global optimal solution obtained in the step S21 as an initial clustering center of a K-means clustering algorithm;
s23, randomly generating K clustering centers, assigning elements in the image data point set to the K classes, and calculating the distance f from each image data point in the image data point set to the clustering centers, wherein the calculation formula is as follows:
in the formula (2-3), XiIs the ith point in the image data point set X, and YjAnd for the jth clustering center, updating the clustering center through the mean value, carrying out algorithm convergence, obtaining the minimum fitness value, and obtaining the optimal initial clustering center of the K-means clustering algorithm.
And (3) foreground area enhancement:
the reason why the infrared image contrast is low is that the gray scale is concentrated in a region with a narrow dynamic range, and a linear enhancement method is used to expand the gray scale range.
The process of linearly enhancing the foreground region of the low frequency component is as follows:
determining a boundary of linear enhancement according to the gray histogram, and determining the range of a concentrated gray area by searching the maximum gray level of the gray histogram; determining left and right boundaries of linear enhancement by maximum gray value, and setting the left boundary as aLThe right boundary is aR；
After the left and right boundaries are determined, the foreground region of the low-frequency component is linearly enhanced, the enhanced image is represented by z (x, y), and the calculation formula is as follows:
by calculating the resulting gray value boundary range [ a ]L,aR]And the existing pixel points are linearly enhanced, the visual effect of the foreground area of the infrared image of the power equipment is improved, the mutual influence of all parts is avoided, and the effect of independent observation is achieved.
Background area enhancement:
the histogram equalization processing for the background region of the low-frequency component is performed as follows:
the probability density of occurrence of a gray value of l is:
in the formula (2-5), N is the total pixel amount; n islThe number of pixels with the gray value of l;
the gray scale distribution function of the image is:
the histogram equalization algorithm multiplies the gray distribution function by (L-1) as a new pixel value, and the conversion relation is as follows:
the low-frequency component subjected to histogram equalization effectively reduces the background brightness, highlights the target pixel value of the main part of the non-electric equipment, and increases the gray difference between the part and the background, so that the identification degree of the original image is improved.
High-frequency component processing:
the high frequency component includes a large amount of target detail information and noise in the power equipment original image. The grey value of the detail information of the target device is lower than that of the noise part, and the visual effect is darker. In order to improve the clearness of the detail presentation of the high-frequency part, firstly, a Spatial Adaptive Noise Smoothing (SANS) algorithm is adopted for processing, vibration or noise which possibly exists in the high-frequency part is eliminated, then a fuzzy enhancement method is used for enhancing the contrast and the definition of the image, and the enhanced high-frequency component is obtained.
In step S4, the SANS algorithm fuses the input blurred image and the restored image by using the local variance of the input blurred image, and a fusion equation and scale parameters calculated according to the local variance are as follows:
in the formulae (2-8) and (2-9),
is the final restored image; alpha (x, y) is a fused weight coefficient, beta is a coefficient which ensures the uniform distribution of alpha between 0 and 1, the smaller the beta value is, the more remarkable the effect of derringing is, and v (x, y) is the local variance of the input image g (x, y);
after the high-frequency component is processed by the SANS algorithm, the edge details of the high-frequency component are kept while vibration and noise are removed, and the method is very suitable for removing the noise of the infrared image of the transformer substation.
The Pal-King algorithm firstly maps the high-frequency component processed by the SANS algorithm from the original space domain to the fuzzy domain, and the classical Pal-King algorithm fuzzy membership function muijComprises the following steps:
in the formula (2-10), tijRefers to the gray value of the image (i, j); l denotes the gray level; kdIs a fuzzy parameter at the denominator; keIs an exponential fuzzy parameter; keAnd KdDetermining a fuzzy membership mapping rule by the combined action;
and further enhancing the fuzzy membership after mapping, and adopting S-type transformation, wherein the S-type transformation formula is as follows:
in the formula (2-11), μijThe fuzzy membership function after S-type transformation is enhanced by the steps and transformed to [0, L-1 according to the inverse function of the fuzzy membership]Within the range, enhanced gray values are obtained, namely:
however, the conventional blur enhancement method has some disadvantages: many pixels with lower gray values are forced to 0 after the transformation, thereby losing edge information. Meanwhile, the transformation form and the formula are complex, and repeated tests are needed, so that the problem of parameter optimization is caused.
Modifying a fuzzy membership function of a classic Pal-King algorithm:
in the formula (2-13), tijFinger image (i, j)) Gray value of tminRefers to the minimum gray value in the image of the high frequency component; t is tmaxRefers to the maximum gray value in the image of the high frequency component;
and then, carrying out flat S-type processing on the modified fuzzy membership function of the Pal-King algorithm by adopting an S-type transformation formula, namely an expression (2-11), and obtaining an enhanced gray value expression as follows:
improved fuzzy membership range from [ mu ]min 1]Expand to [ 01]The improved inverse transformation has no dead zone, thereby effectively preventing the gray scale from flattening, retaining all information of the original image, and simultaneously effectively enhancing the details such as the edge contour in the high-frequency coefficient.
And (3) verification experiment:
in order to verify the NSST domain infrared image enhancement method based on the longicorn whisker optimization algorithm, the infrared image of a certain telegraph pole is segmented and analyzed based on matlab software.
And selecting three layers of Laplacian pyramids to perform multi-scale transformation. The number of directions of each layer is 8, and a low-frequency image (see fig. 5) and three groups of high-frequency images are obtained through NSST conversion. The enhancement method of the invention is used for enhancing the low-frequency image and the high-frequency image, and finally, the NSST inverse transformation is used for synthesis.
Referring to fig. 4a and 4b, the effect of the enhancement method of the present application and the conventional Ostu algorithm (maximum threshold segmentation method) on the infrared image segmentation of the utility pole is shown. Compared with the traditional maximum threshold segmentation method, the adaptive longicorn stigma optimization algorithm has shorter processing speed, can effectively segment a background area (background area) and a foreground area (target area), reduces the background part which is mistakenly segmented to the target area to be detected in the Otsu algorithm, and is more suitable for segmenting the infrared image of the power equipment of the transformer substation in the chaotic environment. And three image algorithms are selected to be compared with the algorithm, and the superiority of the algorithm is verified.
In order to verify the effectiveness of the NSST domain infrared image enhancement method based on the longicorn silk optimization algorithm, a certain transformer substation power device is used as an experimental sample, and the He algorithm, the Pal-King algorithm, the traditional NSST algorithm and the algorithm are compared and analyzed.
Referring to fig. 5, as can be seen from the original ir grayscale image of one pole in fig. 5(a), the image contains noise and has rich fault region targets and backgrounds; if the He algorithm based on the histogram enhanced infrared image is adopted, the integral contrast is improved, but the background brightness is enhanced, and the noise is not inhibited; after Pal-King enhancement, the target brightness is enhanced, but the detail contour is lost; if the enhancement method is adopted, the heat source of the power equipment fault area can be clearly seen, the gray contrast of the measured area and the irrelevant area is improved, the denoising effect is obvious, the detailed outline of the measured area is very clear, and the thermal fault can be conveniently identified by human eyes.
Five objective index evaluations of Edge Strength (ES), Information Entropy (IE), Contrast (CR), Standard Deviation (SD) and peak signal-to-noise ratio (PSNR) are adopted for comparative analysis, and a telegraph pole infrared image enhancement evaluation function table is obtained, and is shown in table 1.
| Evaluation index | ES | IE | CR | SD | PSNR |
| Original drawing | 33..514 | 6.572 | 169.9411 | 58.775 | 16.382 |
| HE | 68.334 | 6.036 | 264.423 | 67.398 | 18.297 |
| Pal-King | 57.689 | 5.897 | 224.798 | 65.485 | 14.263 |
| The invention | 73.991 | 6.893 | 326.260 | 72.885 | 24.543 |
TABLE 1
As can be seen from the experimental results in table 1, compared with the other three algorithms, the edge strength, the information entropy, the contrast, the standard deviation and the peak signal-to-noise ratio of the enhancement method of the present invention are the highest among the three algorithms, and the amplification of the algorithms in the five evaluation indexes is at least 5.38%, 4.24%, 8.57%, 6.68% and 23.46%, which indicates that the image contrast after the optimization of the NSST domain infrared image enhancement method based on the longicorn optimization algorithm of the present invention is improved most obviously, the noise is the smallest, the influence by the monitored area is the smallest, and the infrared target is obvious. Through the analysis, compared with other methods, the NSST domain infrared image enhancement method based on the longicorn whisker optimization algorithm has better performance on both subjective indexes and objective indexes.
As a variation of the present invention, instead of NSST transforms, other frequency domain transform algorithms may be used, such as non-subsampled contours (NSCT); improvement of optimization method, such as optimization of Otsu threshold segmentation using Grey wolf, whale, bat algorithm, etc. And performing enhancement processing on the transformed low-frequency component by adopting other methods, such as Retinex algorithm and gamma transformation. And performing denoising processing on the transformed high-frequency component by adopting other filtering modes, such as a bilateral filter, a guided filter and the like.
In summary, the NSST domain infrared image enhancement method based on the longicorn whisker optimization algorithm can remove noise of the infrared image, improve edge and detail information of the power equipment part in the infrared image, and improve the overall contrast of the power equipment infrared image gray scale image.
It should be understood by those skilled in the art that the above embodiments are only for illustrating the present invention and are not to be used as a limitation of the present invention, and that changes and modifications to the above described embodiments are within the scope of the claims of the present invention as long as they are within the spirit and scope of the present invention.
Claims (5)
1. A NSST domain infrared image enhancement method based on a longicorn whisker optimization algorithm is characterized by comprising the following steps:
s1, performing multi-scale and multi-directional transformation on the original power equipment infrared image by adopting NSST transformation, and decomposing the original power equipment infrared image into a high-frequency component and a low-frequency component;
s2, performing threshold segmentation on the low-frequency component obtained in the step S1 by adopting a longicorn whisker optimization algorithm, and segmenting the low-frequency component into a background region and a foreground region of the power equipment main body;
s3, performing linear enhancement on the foreground region obtained by segmentation in the step S2, performing histogram equalization processing on the background region, and performing synthesis processing on the foreground region and the background region to obtain an enhanced low-frequency component image;
s4, processing the high-frequency component obtained in the step S1 by adopting an SANS algorithm, eliminating vibration and noise of the high-frequency component, and enhancing the contrast and definition of the image of the high-frequency component by using a Pal-King algorithm to obtain an enhanced high-frequency component image;
and S5, performing NSST inverse transformation on the enhanced high-frequency component image and the enhanced low-frequency component image to obtain an enhanced infrared image of the power equipment.
2. The NSST domain infrared image enhancement method based on the longicorn whisker optimization algorithm as claimed in claim 1, wherein in step S1, the multi-scale and multi-directional transformation comprises two processes of multi-scale decomposition and directional localization, wherein:
in the multi-scale decomposition process, an original power equipment infrared image is decomposed into a high-frequency image and a low-frequency image through a non-downsampling pyramid filter bank, NSP decomposition is carried out on the basis of the low-frequency image obtained through decomposition, and k high-frequency sub-bands and 1 low-frequency sub-band image are obtained after k times of NSP decomposition;
the method comprises the following steps of a directional localization process, wherein NSST utilizes improved shear wave transformation to perform localization processing on high-frequency sub-band images and low-frequency sub-band images obtained by multi-scale decomposition, firstly, a standard shear wave filter is mapped to Cartesian coordinates from a pseudo-polar coordinate system, and inverse Fourier transformation is performed; and finally, finishing final processing by utilizing two-dimensional convolution, and decomposing the infrared image of the original power equipment into a high-frequency component and a low-frequency component.
3. The NSST domain infrared image enhancement method based on the longicorn silk optimization algorithm as claimed in claim 1, wherein in step S2, the longicorn silk optimization algorithm is an image segmentation processing method formed by combining a self-adaptive longicorn silk search algorithm and a K-means clustering algorithm, and the method comprises the following steps:
s21, solving a global optimal solution of the image data point set of the low-frequency component by adopting a self-adaptive longicorn whisker search algorithm;
s22, taking the global optimal solution obtained in the step S21 as an initial clustering center of a K-means clustering algorithm;
s23, randomly generating K clustering centers, assigning elements in the image data point set to the K classes, and calculating the distance f from each image data point in the image data point set to the clustering centers, wherein the calculation formula is as follows:
Xiis the ith point in the image data point set X, and YjAnd for the jth clustering center, updating the clustering center through the mean value, carrying out algorithm convergence, obtaining the minimum fitness value, and obtaining the optimal initial clustering center of the K-means clustering algorithm.
4. The NSST domain infrared image enhancement method based on the longicorn whisker optimization algorithm as claimed in claim 1, wherein in step S3, the process of linearly enhancing the foreground region of the low frequency component is as follows:
determining a boundary of linear enhancement according to the gray histogram, and determining the range of a concentrated gray area by searching the maximum gray level of the gray histogram; determining left and right boundaries of linear enhancement by maximum gray value, and setting the left boundary as aLThe right boundary is aR；
After the left and right boundaries are determined, the foreground region of the low-frequency component is linearly enhanced, the enhanced image is represented by z (x, y), and the calculation formula is as follows:
by calculating the resulting gray value boundary range [ a ]L,aR]The existing pixel points are subjected to linear enhancement, and the visual effect of the foreground area of the infrared image of the power equipment is improved;
the histogram equalization processing for the background region of the low-frequency component is performed as follows:
the probability density of occurrence of a gray value of l is:
n is the total amount of pixels; n islThe number of pixels with the gray value of l;
the gray scale distribution function of the image is:
the histogram equalization algorithm multiplies the gray distribution function by (L-1) as a new pixel value, and the conversion relation is as follows:
the low-frequency component subjected to histogram equalization processing reduces the background brightness, highlights the target pixel value of the main part of the non-electric equipment, increases the gray difference between the part and the background, and improves the identification degree of the original image.
5. The NSST domain infrared image enhancement method based on the longicorn whisker optimization algorithm of claim 1, wherein in step S4, the SANS algorithm fuses the input blurred image and the restored image by using the local variance of the input blurred image, and the fusion equation and the scale parameter calculated according to the local variance are as follows:
wherein the content of the first and second substances,
is the final restored image; alpha (x, y) is a fused weight coefficient, beta is a coefficient which ensures the uniform distribution of alpha between 0 and 1, the smaller the beta value is, the more remarkable the effect of derringing is, and v (x, y) is the local variance of the input image g (x, y);
after the high-frequency component is processed by the SANS algorithm, the edge details of the high-frequency component are retained while vibration and noise are removed;
the Pal-King algorithm firstly maps the high-frequency component processed by the SANS algorithm from the original space domain to the fuzzy domain, and the classical Pal-King algorithm fuzzy membership function muijComprises the following steps:
wherein, tijRefers to the gray value of the image (i, j); l denotes the gray level; kdIs a fuzzy parameter at the denominator; keIs an exponential fuzzy parameter; keAnd KdDetermining a fuzzy membership mapping rule by the combined action;
modifying a fuzzy membership function of a classic Pal-King algorithm:
wherein, tijDenotes the gray value, t, of the image (i, j)minRefers to the minimum gray value in the image of the high frequency component; t is tmaxRefers to the maximum gray value in the image of the high frequency component;
and further enhancing the fuzzy membership degree after mapping, and performing flat S-type processing by adopting S-type transformation, wherein the S-type transformation formula is as follows:
μijthe fuzzy membership function after S-type transformation is enhanced by the steps and transformed to [0, L-1 according to the inverse function of the fuzzy membership]Within the range, obtaining an enhanced gray value t'ijNamely:
improved fuzzy membership range from [ mu ]min 1]Expand to [ 01]。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111147083.3A CN113870135A (en) | 2021-09-28 | 2021-09-28 | NSST domain infrared image enhancement method based on longicorn stigma optimization algorithm |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111147083.3A CN113870135A (en) | 2021-09-28 | 2021-09-28 | NSST domain infrared image enhancement method based on longicorn stigma optimization algorithm |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113870135A true CN113870135A (en) | 2021-12-31 |
Family
ID=78992386
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111147083.3A Pending CN113870135A (en) | 2021-09-28 | 2021-09-28 | NSST domain infrared image enhancement method based on longicorn stigma optimization algorithm |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113870135A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114820401A (en) * | 2022-04-19 | 2022-07-29 | 桂林电子科技大学 | Method for enhancing marine backlight infrared image by combining histogram transformation and edge information |
| CN114972074A (en) * | 2022-04-27 | 2022-08-30 | 广东鉴面智能科技有限公司 | Night vision image analysis system based on low-light-level environment |
| CN116433540A (en) * | 2023-06-15 | 2023-07-14 | 武汉高芯科技有限公司 | Infrared image enhancement method and system |
| CN116523798A (en) * | 2023-06-28 | 2023-08-01 | 北京理工大学 | Infrared contrast enhancement method based on local optimization |
| CN117173190A (en) * | 2023-11-03 | 2023-12-05 | 成都中轨轨道设备有限公司 | Insulator infrared damage inspection system based on image processing |
-
2021
- 2021-09-28 CN CN202111147083.3A patent/CN113870135A/en active Pending
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114820401A (en) * | 2022-04-19 | 2022-07-29 | 桂林电子科技大学 | Method for enhancing marine backlight infrared image by combining histogram transformation and edge information |
| CN114820401B (en) * | 2022-04-19 | 2023-03-10 | 桂林电子科技大学 | Method for enhancing marine backlight infrared image by combining histogram transformation and edge information |
| CN114972074A (en) * | 2022-04-27 | 2022-08-30 | 广东鉴面智能科技有限公司 | Night vision image analysis system based on low-light-level environment |
| CN116433540A (en) * | 2023-06-15 | 2023-07-14 | 武汉高芯科技有限公司 | Infrared image enhancement method and system |
| CN116433540B (en) * | 2023-06-15 | 2023-09-29 | 武汉高芯科技有限公司 | Infrared image enhancement method and system |
| CN116523798A (en) * | 2023-06-28 | 2023-08-01 | 北京理工大学 | Infrared contrast enhancement method based on local optimization |
| CN116523798B (en) * | 2023-06-28 | 2023-08-25 | 北京理工大学 | Infrared contrast enhancement method based on local optimization |
| CN117173190A (en) * | 2023-11-03 | 2023-12-05 | 成都中轨轨道设备有限公司 | Insulator infrared damage inspection system based on image processing |
| CN117173190B (en) * | 2023-11-03 | 2024-02-02 | 成都中轨轨道设备有限公司 | Insulator infrared damage inspection system based on image processing |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113870135A (en) | NSST domain infrared image enhancement method based on longicorn stigma optimization algorithm | |
| CN113837974B (en) | NSST domain power equipment infrared image enhancement method based on improved BEEPS filtering algorithm | |
| CN109035166A (en) | Electrical equipment infrared image enhancing method based on non-lower sampling shearing wave conversion | |
| CN114170103A (en) | Electrical equipment infrared image enhancement method | |
| Luo et al. | Infrared and visible image fusion based on visibility enhancement and hybrid multiscale decomposition | |
| Patel et al. | Mammogram of breast cancer detection based using image enhancement algorithm | |
| CN109949256B (en) | Astronomical image fusion method based on Fourier transform | |
| Subramani et al. | Fuzzy contextual inference system for medical image enhancement | |
| Karalı et al. | Adaptive image enhancement based on clustering of wavelet coefficients for infrared sea surveillance systems | |
| CN111179208A (en) | Infrared-visible light image fusion method based on saliency map and convolutional neural network | |
| Patel et al. | Gray level clustering and contrast enhancement (GLC–CE) of mammographic breast cancer images | |
| CN113592729A (en) | Infrared image enhancement method for electrical equipment based on NSCT domain | |
| CN112669249A (en) | Infrared and visible light image fusion method combining improved NSCT (non-subsampled Contourlet transform) transformation and deep learning | |
| Suryavamsi et al. | Comparative analysis of various enhancement methods for astrocytoma MRI images | |
| Hasikin et al. | Adaptive fuzzy intensity measure enhancement technique for non-uniform illumination and low-contrast images | |
| CN113538409B (en) | Cervical cancer image region segmentation method based on fuzzy logic and ANFIS | |
| CN112819739B (en) | Image processing method and system for scanning electron microscope | |
| Liu et al. | Multi-scale saliency measure and orthogonal space for visible and infrared image fusion | |
| Li et al. | A novel medical image fusion approach based on nonsubsampled shearlet transform | |
| Luo et al. | Multi-focus image fusion through pixel-wise voting and morphology | |
| CN114066786A (en) | Infrared and visible light image fusion method based on sparsity and filter | |
| CN113592727A (en) | Infrared image enhancement method for electrical equipment based on NSST domain | |
| Tao et al. | Multimodal image fusion algorithm using dual-tree complex wavelet transform and particle swarm optimization | |
| Li et al. | A novel remote sensing image enhancement method using unsharp masking in NSST domain | |
| TRIVEDI et al. | MOSAICFUSION: MERGING MODALITIES WITH PARTIAL DIFFERENTIAL EQUATION AND DISCRETE COSINE TRANSFORMATION |
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 |