CN116433499A - Image processing method, device and computer readable storage medium - Google Patents
Image processing method, device and computer readable storage medium Download PDFInfo
- Publication number
- CN116433499A CN116433499A CN202211735432.8A CN202211735432A CN116433499A CN 116433499 A CN116433499 A CN 116433499A CN 202211735432 A CN202211735432 A CN 202211735432A CN 116433499 A CN116433499 A CN 116433499A
- Authority
- CN
- China
- Prior art keywords
- image
- frequency
- low
- detail
- sub
- 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
- 238000003672 processing method Methods 0.000 title claims abstract description 15
- 238000001914 filtration Methods 0.000 claims abstract description 135
- 238000000034 method Methods 0.000 claims abstract description 37
- 230000009466 transformation Effects 0.000 claims abstract description 35
- 238000012545 processing Methods 0.000 claims abstract description 10
- 238000004590 computer program Methods 0.000 claims description 6
- 230000001131 transforming effect Effects 0.000 claims description 3
- 230000000717 retained effect Effects 0.000 claims 2
- 238000012935 Averaging Methods 0.000 claims 1
- 238000005516 engineering process Methods 0.000 abstract description 4
- 238000000354 decomposition reaction Methods 0.000 description 17
- 238000004458 analytical method Methods 0.000 description 6
- 238000010586 diagram Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 238000000605 extraction Methods 0.000 description 5
- 230000006870 function Effects 0.000 description 4
- 239000010410 layer Substances 0.000 description 4
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 230000007547 defect Effects 0.000 description 2
- 238000003709 image segmentation Methods 0.000 description 2
- 239000002356 single layer Substances 0.000 description 2
- 238000013519 translation Methods 0.000 description 2
- 230000002159 abnormal effect Effects 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000011284 combination treatment Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010191 image analysis Methods 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000011835 investigation Methods 0.000 description 1
- 238000012423 maintenance Methods 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
- 230000003287 optical effect Effects 0.000 description 1
- 230000003449 preventive effect Effects 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 238000011282 treatment Methods 0.000 description 1
- 238000013024 troubleshooting Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G06T5/70—
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20048—Transform domain processing
- G06T2207/20064—Wavelet transform [DWT]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y04—INFORMATION OR COMMUNICATION TECHNOLOGIES HAVING AN IMPACT ON OTHER TECHNOLOGY AREAS
- Y04S—SYSTEMS INTEGRATING TECHNOLOGIES RELATED TO POWER NETWORK OPERATION, COMMUNICATION OR INFORMATION TECHNOLOGIES FOR IMPROVING THE ELECTRICAL POWER GENERATION, TRANSMISSION, DISTRIBUTION, MANAGEMENT OR USAGE, i.e. SMART GRIDS
- Y04S10/00—Systems supporting electrical power generation, transmission or distribution
- Y04S10/50—Systems or methods supporting the power network operation or management, involving a certain degree of interaction with the load-side end user applications
Abstract
The invention discloses an image processing method, an image processing device and a computer readable storage medium. Wherein the method comprises the following steps: acquiring an original infrared image of target equipment in a transformer substation; performing wavelet transformation on the original infrared image to obtain a low-frequency approximation coefficient and a high-frequency detail coefficient; generating a low frequency approximation sub-image based on the low frequency approximation coefficients and generating a high frequency detail sub-image based on the high frequency detail coefficients; carrying out mean value filtering on the high-frequency detail sub-image to obtain a filtered high-frequency detail sub-image; reconstructing the low-frequency approximate sub-image and the filtered high-frequency detail sub-image to generate a target infrared image. The invention solves the technical problems that in the related technology, the noise is eliminated for the infrared image and the damage and loss are caused to the high-frequency detail component of the image.
Description
Technical Field
The present invention relates to the field of power technology, and in particular, to an image processing method, an image processing device, and a computer readable storage medium.
Background
In the related art, when the operation condition of the transformer substation is monitored remotely, one mode is to analyze the data acquired by the on-site sensor, but many fault characteristics cannot be reflected from the data, and besides the process of dispatching personnel to on-site investigation and inspection, the first visual data is difficult to obtain. The second mode is that when the power grid faults occur, when the running state of the field device is confirmed, the field video is manually selected and called as a main working means, and if the large-scale power grid faults occur, the condition of the field device cannot be accurately judged in a short time by regulatory personnel, so that the fault handling process is seriously affected. Therefore, the infrared camera is required to be combined with the visible light camera for joint monitoring, and the method plays an important role in the development from the traditional preventive trouble shooting to the advanced predictive state maintenance body. As known from the imaging characteristics of the infrared image, the noise of the infrared image mainly comprises gaussian noise derived from background clutter and white noise derived from shot noise and radiation noise, i.e. the noise of the infrared image is mainly gaussian white noise.
In the related art, the problems that the infrared image eliminates noise for the infrared image, and simultaneously damages and loses high-frequency detail components of the image, influences the quality of subsequent image segmentation and has adverse effects on target feature extraction and recognition exist.
In view of the above problems, no effective solution has been proposed at present.
Disclosure of Invention
The embodiment of the invention provides an image processing method, an image processing device and a computer readable storage medium, which at least solve the technical problems that in the related art, noise is eliminated on an infrared image and high-frequency detail components of the image are damaged and lost.
According to an aspect of an embodiment of the present invention, there is provided an image processing method including: acquiring an original infrared image of target equipment in a transformer substation; performing wavelet transformation on the original infrared image to obtain a low-frequency approximation coefficient and a high-frequency detail coefficient; generating a low frequency approximation sub-image based on the low frequency approximation coefficients and a high frequency detail sub-image based on the high frequency detail coefficients; performing mean filtering on the high-frequency detail sub-image to obtain a filtered high-frequency detail sub-image; reconstructing the low-frequency approximate sub-image and the filtered high-frequency detail sub-image to generate a target infrared image.
Optionally, the performing wavelet transformation on the original infrared image to obtain a low-frequency approximation coefficient and a high-frequency detail coefficient includes: under the condition that the wavelet transformation is first-order wavelet transformation, carrying out first-order wavelet transformation on the original infrared image to obtain a low-frequency approximate coefficient and three high-frequency detail coefficients, wherein the low-frequency approximate coefficient is a coefficient for carrying out low-pass filtering and low-pass filtering on the original infrared image, and the three high-frequency detail coefficients comprise: the system comprises a horizontal high-frequency detail coefficient, a vertical high-frequency detail coefficient and a diagonal high-frequency detail coefficient, wherein the horizontal high-frequency detail coefficient is a coefficient for performing row low-pass filtering and column high-pass filtering on the original infrared image, the vertical high-frequency detail coefficient is a coefficient for performing row low-pass filtering and high-pass filtering on the original infrared image, and the diagonal high-frequency detail coefficient is a coefficient for performing high-pass filtering and column high-pass filtering on the original infrared image.
Optionally, the generating a low frequency approximation sub-image based on the low frequency approximation coefficient and generating a high frequency detail sub-image based on the high frequency detail coefficient includes: performing line low-pass filtering and low-pass filtering on the original infrared image by adopting the low-frequency approximation coefficients to obtain the low-frequency approximation sub-image comprising low-frequency information in the horizontal direction and low-frequency information in the vertical direction; adopting the horizontal high-frequency detail coefficient to perform low-pass filtering and column high-pass filtering on the original infrared image to obtain a horizontal detail sub-image comprising horizontal high-frequency information and vertical low-frequency information; adopting the vertical high-frequency detail coefficient to perform line low-pass filtering and high-pass filtering on the original infrared image to obtain a vertical detail sub-image comprising vertical high-frequency information and horizontal low-frequency information; adopting the diagonal high-frequency detail coefficients to perform high-pass filtering and column high-pass filtering on the original infrared image to obtain a diagonal detail sub-image comprising high-frequency information in the horizontal direction and high-frequency information in the vertical direction; wherein the high frequency detail sub-image comprises the horizontal detail sub-image, the vertical detail sub-image and the diagonal detail sub-image.
Optionally, the performing average filtering on the high-frequency detail sub-image to obtain a filtered high-frequency detail sub-image includes: performing mean value filtering on the horizontal detail sub-image by adopting a horizontal linear template to obtain a filtered horizontal detail sub-image; performing mean value filtering on the vertical detail sub-image by adopting a vertical linear template to obtain a filtered vertical detail sub-image; average filtering is carried out on the diagonal detail sub-images by adopting the angular template, so as to obtain filtered diagonal detail sub-images; wherein the filtered high frequency detail sub-image comprises: the filtered horizontal detail sub-image, the filtered vertical detail sub-image and the filtered diagonal detail sub-image. Optionally, the noise of the horizontal detail sub-image in the horizontal direction is removed from the filtered horizontal detail sub-image, the low frequency information in the vertical direction is reserved, the noise of the vertical detail sub-image in the vertical direction is removed from the filtered vertical detail sub-image, and the low frequency information in the horizontal direction is reserved.
Optionally, the target device is a circuit breaker.
Optionally, according to another aspect of the embodiment of the present invention, there is also provided an image processing apparatus including: the acquisition module is used for acquiring an original infrared image of target equipment in the transformer substation; the transformation module is used for carrying out wavelet transformation on the original infrared image to obtain a low-frequency approximation coefficient and a high-frequency detail coefficient; a generation module for generating a low frequency approximation sub-image based on the low frequency approximation coefficients and a high frequency detail sub-image based on the high frequency detail coefficients; the filtering module is used for carrying out mean value filtering on the high-frequency detail sub-image to obtain a filtered high-frequency detail sub-image; and the reconstruction module is used for reconstructing the low-frequency approximate sub-image and the filtered high-frequency detail sub-image to generate a target infrared image.
Optionally, the transformation module includes: the transforming unit is configured to perform first-order wavelet transform on the original infrared image under the condition that the wavelet transform is first-order wavelet transform, so as to obtain a low-frequency approximation coefficient and three high-frequency detail coefficients, where the one low-frequency approximation coefficient is a coefficient for performing low-pass filtering and low-pass filtering on the original infrared image, and the three high-frequency detail coefficients include: the system comprises a horizontal high-frequency detail coefficient, a vertical high-frequency detail coefficient and a diagonal high-frequency detail coefficient, wherein the horizontal high-frequency detail coefficient is a coefficient for performing row low-pass filtering and column high-pass filtering on the original infrared image, the vertical high-frequency detail coefficient is a coefficient for performing row low-pass filtering and high-pass filtering on the original infrared image, and the diagonal high-frequency detail coefficient is a coefficient for performing high-pass filtering and column high-pass filtering on the original infrared image.
According to still another aspect of the embodiments of the present invention, there is also provided a computer-readable storage medium including: the computer-readable storage medium includes a stored program, wherein the program, when executed, controls a device in which the computer-readable storage medium is located to perform the image processing method of any one of the above.
According to still another aspect of the embodiment of the present invention, there is provided a computer apparatus including: a memory and a processor, the memory storing a computer program; the processor is configured to execute a computer program stored in the memory, where the computer program when executed causes the processor to execute any one of the image processing methods described above.
In the embodiment of the invention, the original infrared image of the equipment in the transformer substation is acquired; the method comprises the steps of carrying out wavelet transformation on an original infrared image to obtain a low-frequency approximation coefficient and a high-frequency detail coefficient and generating a sub-image, carrying out mean value filtering on the high-frequency detail sub-image and reconstructing the low-frequency approximation sub-image to generate a target infrared image, so that the technical effects of reducing the noise of the image and keeping the details of the image as far as possible are realized, and further the technical problems that in the related art, the high-frequency detail component of the image is damaged and lost while the noise is eliminated for the infrared image are solved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this application, illustrate embodiments of the invention and together with the description serve to explain the invention and do not constitute a limitation on the invention. In the drawings:
FIG. 1 is a flow chart of an image processing method according to the prior art;
FIG. 2 is an exploded view of an alternative first and second order wavelet transform according to an embodiment of the present invention;
FIG. 3 is an exemplary diagram of an alternative breaker image wavelet analysis in accordance with an embodiment of the present invention;
FIG. 4 is a schematic diagram of an alternative mean filtering template according to an embodiment of the present invention;
fig. 5 is a schematic diagram of an alternative image processing apparatus according to an embodiment of the present invention.
Detailed Description
In order that those skilled in the art will better understand the present invention, a technical solution in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in which 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 present invention without making any inventive effort, shall fall within the scope of the present invention.
It should be noted that the terms "first," "second," and the like in the description and the claims of the present invention and the above figures are used for distinguishing between similar objects and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used may be interchanged where appropriate such that the embodiments of the invention described herein may be implemented in sequences other than those illustrated or otherwise described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.
According to an embodiment of the present invention, there is provided a method embodiment of a data processing method, it being noted that the steps shown in the flowcharts of the figures may be performed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowcharts, in some cases the steps shown or described may be performed in an order different from that herein.
Fig. 1 is a flowchart of an image processing method according to an embodiment of the present invention, as shown in fig. 1, including the steps of:
step S102, acquiring an original infrared image of target equipment in a transformer substation;
as an alternative embodiment, the execution subject of the image processing method may be a terminal or a server. The terminals may be various types of terminals, for example, computer terminals, mobile terminals, virtual terminals, etc., but regardless of the type of terminals, a certain computing power is required to meet the computing requirements. The server may be in various forms, for example, may be a single computer device, may be a computer cluster including a plurality of computers, may be a local computing unit, may be a remote cloud server, or the like.
As an alternative embodiment, the target device may be a circuit breaker, and the circuit breaker may be a switching device capable of switching on, carrying, and off a current under normal loop conditions and capable of switching on, carrying, and off a current under abnormal loop conditions for a prescribed time.
As an alternative embodiment, there may be a variety of ways to obtain the original infrared image of the target device in the substation, for example, a camera may be used to capture the target device to obtain the original infrared image, or a part of the original infrared image may be processed by capturing in the obtained original infrared image. When shooting is performed by adopting a camera, the camera can be on any device with the capability of acquiring infrared images or computing capability. In addition, when the existing image is intercepted, the image can be obtained by enlarging or reducing the image acquired from the image acquisition equipment.
Step S104, carrying out wavelet transformation on the original infrared image to obtain a low-frequency approximation coefficient and a high-frequency detail coefficient;
as an alternative embodiment, in the case of wavelet transformation into first-order wavelet transformation, the first-order wavelet transformation is performed on the original infrared image to obtain a low-frequency approximation coefficient and three high-frequency detail coefficients, where one low-frequency approximation coefficient is a coefficient for performing low-pass filtering and low-pass filtering on the original infrared image, and the three high-frequency detail coefficients include: the system comprises a horizontal high-frequency detail coefficient, a vertical high-frequency detail coefficient and a diagonal high-frequency detail coefficient, wherein the horizontal high-frequency detail coefficient is a coefficient for performing low-pass filtering and column high-pass filtering on an original infrared image, the vertical high-frequency detail coefficient is a coefficient for performing column low-pass filtering and high-pass filtering on the original infrared image, and the diagonal high-frequency detail coefficient is a coefficient for performing high-pass filtering and column high-pass filtering on the original infrared image.
As an alternative embodiment, the above wavelet transformation may be to scale the width of the mother wavelet to make the frequency characteristic of the signal more obvious, translate the mother wavelet to obtain the time information of the signal, and calculate the wavelet coefficient better through the scaling and translating operation, where the wavelet coefficient reflects the correlation degree between the wavelet and the local signal, so that the frequency characteristic of the signal is more easily obtained, and the basis is laid for extracting the characteristic of the infrared image and removing the interference noise.
As an alternative embodiment, the above wavelet transform is a localized analysis of the temporal (spatial) frequency, the signal (function) is multi-scale refined by scaling and translation operations, since the wavelet transform has the characteristics of: for small frequency values, the frequency domain resolution is high and the time domain resolution is low; for large frequency values, the frequency domain resolution is low and the time domain resolution is high. Therefore, the method can achieve the effects of time subdivision at high frequency and frequency subdivision at low frequency and meet the requirements of time-frequency signal analysis, so that a feature extraction target can be focused on any details in the signals, the feature extraction precision is effectively improved through purposeful detail extraction, and the influence of noise on the original infrared image features is indirectly eliminated.
As an alternative embodiment, the original infrared image is decomposed into a series of components with different details by a first order wavelet transform, wherein the series of components with different details may include: one low frequency approximation coefficient component obtained by low pass filtering and column low pass filtering the original infrared image, and three high frequency detail coefficients. The three high frequency detail coefficients are respectively as follows, and the horizontal high frequency detail coefficients are as follows: the method comprises the steps of performing low-pass filtering and high-pass filtering on an original infrared image; vertical high frequency detail coefficients: the method comprises the steps of performing low-pass line filtering and high-pass line filtering on an original infrared image; diagonal high frequency detail coefficients: the method is obtained by performing high-pass filtering and column high-pass filtering on an original infrared image. According to the high-pass filtering method, the part with higher frequency is reserved, the part with lower filtering frequency can be used for detecting the sharp part and the part with obvious change in the image, the part with lower frequency is reserved in the low-pass filtering mode, so that the original image becomes smooth, noise in the original image can be effectively filtered, and the technical effects of filtering the noise well and extracting the image characteristics can be achieved by carrying out different filtering combination treatment on the rows and the columns of the original infrared image.
Step S106, generating a low-frequency approximate sub-image based on the low-frequency approximate coefficient, and generating a high-frequency detail sub-image based on the high-frequency detail coefficient;
as an alternative embodiment, the low-frequency approximation coefficients are adopted to perform low-pass filtering and low-pass filtering on the original infrared image, so as to obtain a low-frequency approximation sub-image comprising low-frequency information in the horizontal direction and low-frequency information in the vertical direction; performing low-pass filtering and column high-pass filtering on an original infrared image by adopting a horizontal high-frequency detail coefficient to obtain a horizontal detail sub-image comprising horizontal high-frequency information and vertical low-frequency information; adopting a vertical high-frequency detail coefficient to perform row low-pass filtering and high-pass filtering on an original infrared image to obtain a vertical detail sub-image comprising high-frequency information in a vertical direction and low-frequency information in a horizontal direction; the diagonal high-frequency detail coefficients are adopted to carry out high-pass filtering and column high-pass filtering on the original infrared image, so that a diagonal detail sub-image comprising high-frequency information in the horizontal direction and high-frequency information in the vertical direction is obtained; wherein the high frequency detail sub-images include a horizontal detail sub-image, a vertical detail sub-image and a diagonal detail sub-image.
As an optional embodiment, the low-frequency approximation coefficient characterizes the low-frequency partial information of wavelet decomposition reconstruction of the signal, the high-frequency detail coefficient characterizes the high-frequency partial information of the signal, and the low-frequency approximation coefficient and the high-frequency detail coefficient are used for respectively acquiring the low-frequency approximation sub-image and the high-frequency detail sub-image of the original infrared image in different combined filtering modes.
Step S108, carrying out mean value filtering on the high-frequency detail sub-image to obtain a filtered high-frequency detail sub-image;
as an alternative embodiment, a horizontal line-shaped template may be used to perform mean filtering on the horizontal detail sub-image, so as to obtain a filtered horizontal detail sub-image; performing mean value filtering on the vertical detail sub-images by adopting a vertical linear template to obtain filtered vertical detail sub-images; average filtering is carried out on the diagonal detail sub-images by adopting the angular template, so as to obtain filtered diagonal detail sub-images; wherein the filtered high frequency detail sub-image comprises: a filtered horizontal detail sub-image, a filtered vertical detail sub-image and a filtered diagonal detail sub-image.
As an alternative embodiment, the average filtering is a linear filtering algorithm, a template is set on the image for the target pixel, the horizontal linear template, the vertical linear template and the diagonal linear template comprise adjacent pixel points around the target pixel in the high-frequency detail sub-image and pixel points of the target pixel, and the average value of all pixels in the template is used for replacing the original pixel value.
And step S110, reconstructing the low-frequency approximate sub-image and the filtered high-frequency detail sub-image to generate a target infrared image.
As an alternative embodiment, the image reconstruction is performed based on the obtained low-frequency approximate sub-image and the filtered high-frequency detail sub-image, the original signal is restored by utilizing the wavelet decomposition coefficient of the signal, the restored signal and the original signal still keep the same length in length, and the image capable of retaining the characteristic detail of the original infrared image after reducing the image noise is obtained.
Through the steps, the obtained original infrared image is subjected to wavelet transformation to obtain the low-frequency approximation coefficient and the high-frequency detail coefficient, a series of sub-images of the original infrared image are respectively obtained based on the low-frequency approximation coefficient and the high-frequency detail coefficient, and the series of sub-images are reconstructed after noise is removed by mean filtering of the high-frequency detail sub-images, so that the infrared image which reduces the noise of the image and simultaneously retains the details of the image is obtained, the technical effect of well denoising the original infrared image is achieved, and the technical problem that denoising treatment of the infrared image in place in the related technology is solved.
Based on the above examples and preferred embodiments, an alternative implementation is provided.
In the related art, the conventional mean filtering is a common image filtering denoising method, and the method is simple in operation and has good denoising capability on Gaussian noise. However, the mean filtering is a low-pass filtering method in nature, and the method can destroy and lose high-frequency detail components of the image and blur the image while eliminating noise, which is an inherent defect of the mean filtering method. By adopting the method to process the acquired original infrared image of the target equipment, the image segmentation and the image feature extraction of the infrared image are seriously affected, and whether the fault of the target equipment occurs can not be accurately found in time.
Based on the above-mentioned problems, in this alternative embodiment, there is provided an infrared image denoising method based on a combination of wavelet transform and mean filtering, and fig. 2 is an exploded view of first-order and second-order wavelet transform provided according to an embodiment of the present invention, as shown in fig. 2. The relevant functional processing in the method includes the following.
The average filtering is a common image filtering denoising method, and the method is simple in operation and has good denoising capability on Gaussian noise. However, the mean filtering is a low-pass filtering method in nature, and the method can destroy and lose high-frequency detail components of the image and blur the image while eliminating noise, which is an inherent defect of the mean filtering method. Therefore, wavelet transformation and mean filtering are combined, a noisy image is subjected to wavelet decomposition to obtain detail sub-images with frequency distribution characteristics, and then, according to the frequency distribution characteristics of each detail sub-image, templates with different shapes are adopted for mean filtering, so that the noise of the image is reduced, and meanwhile, the details of the image are kept as far as possible.
Wavelet transformation is a very useful tool in image analysis. Essentially, the method filters the signal by using a band-pass filter with different scales, decomposes the signal onto different frequency bands and then analyzes the signal. Wavelet decomposition is based on two-dimensional discrete wavelet transforms and corresponding multi-resolution analysis. In the decomposition process, the row and column signals of the image are always decomposed at the same time, the sub-image obtained after decomposition of each layer is a value in four directions of LL, LH, HL, HH, and the decomposition of the next layer is repeated on the LL sub-image of the previous layer. As shown in fig. 2, fig. 2 is an exploded view of the first and second order wavelet transforms.
LL describes the low-pass filtered row and column diagonal components of the sub-image, which are one-fourth of the input image, contains the basic information of the original image, contains most of the energy, and constitutes the next decomposed image; LH is the horizontal detail component of the image consisting of low-pass filtered rows and high-pass filtered columns; HL is the vertical detail component of an image consisting of high-pass filtered rows and low-pass filtered columns; HH is the diagonal detail component of the high-pass filtered row and column composition.
Figure 3 is a diagram of an example wavelet analysis of a circuit breaker image provided by an embodiment of the present invention,
the wavelet decomposition of the image needs to select a proper wavelet basis, the wavelet basis is a series of functions obtained by scaling and translation transformation of a parent wavelet, and as the antisymmetric biorthogonal wavelet has an approximate differential operator function, when the antisymmetric biorthogonal wavelet is used for image decomposition, gradient information can be directly obtained by the decomposition low-frequency coefficient of the antisymmetric biorthogonal wavelet, which means that the decomposition coefficient of the antismoke biorthogonal wavelet can more effectively distinguish the low-frequency component and the high-frequency component of the image. Therefore, the embodiment of the invention adopts the anti-symmetric biorthogonal wavelet bin 1.5. FIG. 3 is a result of a layer of wavelet decomposition of a two-dimensional image in a substation infrared image library using a bin 1.5 wavelet basis.
By performing a single-layer wavelet decomposition image on the noisy image f, a low-frequency approximation coefficient cA1 and three high-frequency detail coefficients cH1, cV1, cD1 can be obtained. The low-frequency approximation sub-image A1, the horizontal detail sub-image H1, the vertical detail sub-image V1, and the diagonal detail sub-image D1 are reconstructed from cA1, cH1, cV1, and cD1, respectively (each decomposition sub-image is respectively corresponding in fig. 3 (b)). And (3) keeping A1 unchanged, respectively adopting different templates for average filtering on H1, V1 and D1, and reconstructing the filtered images into 1, 1 and 1 to obtain the denoised image.
FIG. 4 is an exemplary diagram of a mean filtering template, and further analysis in conjunction with the relevant content of FIG. 3 described above, may yield: by performing a single-layer wavelet decomposition image on the noisy image f, a low-frequency approximation coefficient cA1 and three high-frequency detail coefficients cH1, cV1, cD1 can be obtained. The low-frequency approximation sub-image A1, the horizontal detail sub-image H1, the vertical detail sub-image V1, and the diagonal detail sub-image D1 are reconstructed from cA1, cH1, cV1, and cD1, respectively (each decomposition sub-image is respectively corresponding in fig. 3 (b)). And (3) keeping A1 unchanged, respectively adopting different templates for average filtering on H1, V1 and D1, and reconstructing the filtered images into 1, 1 and 1 to obtain the denoised image.
As can be seen from wavelet decomposition, H1 is a sub-image of an image after line low-pass and column high-pass filtering, and comprises high-frequency information of an image signal in the horizontal direction and low-frequency information in the vertical direction, and the horizontal linear template shown in the figure 4 (a) is adopted for mean value filtering, so that noise in the horizontal direction is removed, and meanwhile, the low-frequency information in the vertical direction is well reserved; v1 is a sub-image of the image after column low-pass and high-pass filtering, which contains high-frequency information of the image signal in the vertical direction and low-frequency information in the horizontal direction, and the vertical linear template shown in FIG. 4 (b) is adopted for mean filtering, so that noise in the vertical direction is removed, and meanwhile, the low-frequency information in the horizontal direction is well reserved; d1 is a sub-image of the image after high-pass and column high-pass filtering, which contains high-frequency information of signals in two directions of horizontal and vertical, namely diagonal high-frequency information, and the X-shaped template shown in fig. 4 (c) is adopted for mean filtering.
In an embodiment of the present invention, there is further provided an image processing apparatus, and fig. 5 is a schematic diagram of the image processing apparatus provided according to an embodiment of the present invention, as shown in fig. 3, including: the system comprises an acquisition module 50, a transformation module 52, a generation module 54, a filtering module 56 and a reconstruction module 58. The device will be described below.
The acquisition module 50 is used for acquiring an original infrared image of target equipment in the transformer substation; the transformation module 52 is connected to the acquisition module 50 and is used for performing wavelet transformation on the original infrared image to obtain a low-frequency approximation coefficient and a high-frequency detail coefficient; a generation module 54, coupled to the transformation module 52, for generating a low frequency approximation sub-image based on the low frequency approximation coefficients and a high frequency detail sub-image based on the high frequency detail coefficients; the filtering module 56 is linked to the generating module 54, and is configured to perform mean filtering on the high-frequency detail sub-image to obtain a filtered high-frequency detail sub-image; the reconstruction module 58 is connected to the filtering module 56, and is configured to reconstruct the low-frequency approximate sub-image and the filtered high-frequency detail sub-image to generate the target infrared image.
Optionally, the transformation module includes: the transformation unit is used for carrying out first-order wavelet transformation on the original infrared image under the condition that the wavelet transformation is the first-order wavelet transformation, so as to obtain a low-frequency approximate coefficient and three high-frequency detail coefficients, wherein one low-frequency approximate coefficient is a coefficient for carrying out low-pass filtering and low-pass filtering on the original infrared image, and the three high-frequency detail coefficients comprise: the system comprises a horizontal high-frequency detail coefficient, a vertical high-frequency detail coefficient and a diagonal high-frequency detail coefficient, wherein the horizontal high-frequency detail coefficient is a coefficient for performing low-pass filtering and column high-pass filtering on an original infrared image, the vertical high-frequency detail coefficient is a coefficient for performing column low-pass filtering and high-pass filtering on the original infrared image, and the diagonal high-frequency detail coefficient is a coefficient for performing high-pass filtering and column high-pass filtering on the original infrared image.
The foregoing embodiment numbers of the present invention are merely for the purpose of description, and do not represent the advantages or disadvantages of the embodiments.
In the foregoing embodiments of the present invention, the descriptions of the embodiments are emphasized, and for a portion of this disclosure that is not described in detail in this embodiment, reference is made to the related descriptions of other embodiments.
In the several embodiments provided in the present application, it should be understood that the disclosed technology content may be implemented in other manners. The above-described embodiments of the apparatus are merely exemplary, and the division of the units, for example, may be a logic function division, and may be implemented in another manner, for example, a plurality of units or components may be combined or may be integrated into another system, or some features may be omitted, or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with each other may be through some interfaces, units or modules, or may be in electrical or other forms.
The units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in the embodiments of the present invention may be integrated in one processing unit, or each unit may exist alone physically, or two or more units may be integrated in one unit. The integrated units may be implemented in hardware or in software functional units.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present invention may be embodied essentially or in part or all of the technical solution or in part in the form of a software product stored in a storage medium, including instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a removable hard disk, a magnetic disk, or an optical disk, or other various media capable of storing program codes.
The foregoing is merely a preferred embodiment of the present invention and it should be noted that modifications and adaptations to those skilled in the art may be made without departing from the principles of the present invention, which are intended to be comprehended within the scope of the present invention.
Claims (10)
1. An image processing method, comprising:
acquiring an original infrared image of target equipment in a transformer substation;
performing wavelet transformation on the original infrared image to obtain a low-frequency approximation coefficient and a high-frequency detail coefficient;
generating a low frequency approximation sub-image based on the low frequency approximation coefficients and a high frequency detail sub-image based on the high frequency detail coefficients;
performing mean filtering on the high-frequency detail sub-image to obtain a filtered high-frequency detail sub-image;
reconstructing the low-frequency approximate sub-image and the filtered high-frequency detail sub-image to generate a target infrared image.
2. The method of claim 1, wherein said wavelet transforming said original infrared image to obtain low frequency approximation coefficients and high frequency detail coefficients comprises:
under the condition that the wavelet transformation is first-order wavelet transformation, carrying out first-order wavelet transformation on the original infrared image to obtain a low-frequency approximate coefficient and three high-frequency detail coefficients, wherein the low-frequency approximate coefficient is a coefficient for carrying out low-pass filtering and low-pass filtering on the original infrared image, and the three high-frequency detail coefficients comprise: the system comprises a horizontal high-frequency detail coefficient, a vertical high-frequency detail coefficient and a diagonal high-frequency detail coefficient, wherein the horizontal high-frequency detail coefficient is a coefficient for performing row low-pass filtering and column high-pass filtering on the original infrared image, the vertical high-frequency detail coefficient is a coefficient for performing row low-pass filtering and high-pass filtering on the original infrared image, and the diagonal high-frequency detail coefficient is a coefficient for performing high-pass filtering and column high-pass filtering on the original infrared image.
3. The method of claim 1, wherein the generating a low frequency approximation sub-image based on the low frequency approximation coefficients and generating a high frequency detail sub-image based on the high frequency detail coefficients comprises:
performing line low-pass filtering and low-pass filtering on the original infrared image by adopting the low-frequency approximation coefficients to obtain the low-frequency approximation sub-image comprising low-frequency information in the horizontal direction and low-frequency information in the vertical direction;
adopting the horizontal high-frequency detail coefficient to perform low-pass filtering and column high-pass filtering on the original infrared image to obtain a horizontal detail sub-image comprising horizontal high-frequency information and vertical low-frequency information;
adopting the vertical high-frequency detail coefficient to perform line low-pass filtering and high-pass filtering on the original infrared image to obtain a vertical detail sub-image comprising vertical high-frequency information and horizontal low-frequency information;
adopting the diagonal high-frequency detail coefficients to perform high-pass filtering and column high-pass filtering on the original infrared image to obtain a diagonal detail sub-image comprising high-frequency information in the horizontal direction and high-frequency information in the vertical direction;
wherein the high frequency detail sub-image comprises the horizontal detail sub-image, the vertical detail sub-image and the diagonal detail sub-image.
4. A method according to claim 3, wherein said averaging said high frequency detail sub-images to obtain filtered high frequency detail sub-images comprises:
performing mean value filtering on the horizontal detail sub-image by adopting a horizontal linear template to obtain a filtered horizontal detail sub-image;
performing mean value filtering on the vertical detail sub-image by adopting a vertical linear template to obtain a filtered vertical detail sub-image;
average filtering is carried out on the diagonal detail sub-images by adopting the angular template, so as to obtain filtered diagonal detail sub-images;
wherein the filtered high frequency detail sub-image comprises: the filtered horizontal detail sub-image, the filtered vertical detail sub-image and the filtered diagonal detail sub-image.
5. The method of claim 4, wherein noise in a horizontal direction of the horizontal detail sub-image is removed from the filtered horizontal detail sub-image, low frequency information in a vertical direction is retained, noise in a vertical direction of the vertical detail sub-image is removed from the filtered vertical detail sub-image, and low frequency information in a horizontal direction is retained.
6. The method according to any one of claims 1 to 5, wherein the target device is a circuit breaker.
7. An image processing apparatus, comprising:
the acquisition module is used for acquiring an original infrared image of target equipment in the transformer substation;
the transformation module is used for carrying out wavelet transformation on the original infrared image to obtain a low-frequency approximation coefficient and a high-frequency detail coefficient;
a generation module for generating a low frequency approximation sub-image based on the low frequency approximation coefficients and a high frequency detail sub-image based on the high frequency detail coefficients;
the filtering module is used for carrying out mean value filtering on the high-frequency detail sub-image to obtain a filtered high-frequency detail sub-image;
and the reconstruction module is used for reconstructing the low-frequency approximate sub-image and the filtered high-frequency detail sub-image to generate a target infrared image.
8. The apparatus of claim 7, wherein the transformation module comprises:
the transforming unit is configured to perform first-order wavelet transform on the original infrared image under the condition that the wavelet transform is first-order wavelet transform, so as to obtain a low-frequency approximation coefficient and three high-frequency detail coefficients, where the one low-frequency approximation coefficient is a coefficient for performing low-pass filtering and low-pass filtering on the original infrared image, and the three high-frequency detail coefficients include: the system comprises a horizontal high-frequency detail coefficient, a vertical high-frequency detail coefficient and a diagonal high-frequency detail coefficient, wherein the horizontal high-frequency detail coefficient is a coefficient for performing row low-pass filtering and column high-pass filtering on the original infrared image, the vertical high-frequency detail coefficient is a coefficient for performing row low-pass filtering and high-pass filtering on the original infrared image, and the diagonal high-frequency detail coefficient is a coefficient for performing high-pass filtering and column high-pass filtering on the original infrared image.
9. A computer-readable storage medium, characterized in that the computer-readable storage medium comprises a stored program, wherein the program, when run, controls a device in which the computer-readable storage medium is located to perform the image processing method of any one of claims 1 to 6.
10. A computer device, comprising: a memory and a processor, wherein the memory is configured to store,
the memory stores a computer program;
the processor configured to execute a computer program stored in the memory, the computer program when executed causing the processor to perform the image processing method of any one of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211735432.8A CN116433499A (en) | 2022-12-30 | 2022-12-30 | Image processing method, device and computer readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211735432.8A CN116433499A (en) | 2022-12-30 | 2022-12-30 | Image processing method, device and computer readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116433499A true CN116433499A (en) | 2023-07-14 |
Family
ID=87087913
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211735432.8A Pending CN116433499A (en) | 2022-12-30 | 2022-12-30 | Image processing method, device and computer readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116433499A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117830125A (en) * | 2024-03-04 | 2024-04-05 | 厦门微图软件科技有限公司 | Quick multi-focus fusion algorithm |
-
2022
- 2022-12-30 CN CN202211735432.8A patent/CN116433499A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117830125A (en) * | 2024-03-04 | 2024-04-05 | 厦门微图软件科技有限公司 | Quick multi-focus fusion algorithm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111798396A (en) | Multifunctional image processing method based on wavelet transformation | |
| KR20190031295A (en) | Background suppression method and detection device during display panel automatic optical detection | |
| CN110796615B (en) | Image denoising method, device and storage medium | |
| JP5027757B2 (en) | Moving image denoising device, method and program thereof | |
| CN116433499A (en) | Image processing method, device and computer readable storage medium | |
| KR101405435B1 (en) | Method and apparatus for blending high resolution image | |
| CN109447903A (en) | A kind of method for building up of half reference type super-resolution reconstruction image quality evaluation model | |
| Jiao et al. | Infrared dim small target detection method based on background prediction and high-order statistics | |
| Wang et al. | A unified framework of source camera identification based on features | |
| Pang | Improved image denoising based on Haar wavelet transform | |
| Bhonsle et al. | Medical Image De-Noising Using Combined Bayes Shrink and Total Variation Techniques | |
| Ezekiel et al. | No-reference blur metric using double-density and dual-tree two-dimensional wavelet transformation | |
| Lakshman et al. | Image interpolation using shearlet based sparsity priors | |
| CN107680063B (en) | Method for enhancing direct digital image | |
| Abdulah et al. | Comparison of Image Restoration using Median, Wiener, and Gaussian Filtering Techniques based on Electrical Tree | |
| Arif et al. | Super-resolution using edge modification through stationary wavelet transform | |
| Kaur et al. | Study of Image Denoising and Its Techniques | |
| JP5022400B2 (en) | Moving image noise removing device, moving region image noise removing device, moving image noise removing program, and moving region image noise removing program | |
| Verma et al. | Analysis of Hybrid Wavelet Transformed Images to Improve Image Pixel Quality | |
| Shree et al. | A new fusion-based agricultural synthetic aperture radar image despeckling by using anisotropic diffusion and discrete wavelet transform methods | |
| Tajima et al. | Chromatic interpolation based on anisotropy-scale-mixture statistics | |
| Foucher et al. | Unsupervised multiscale speckle filtering | |
| Biswas et al. | Medical image fusion using regional statistics of shift-invariant shearlet domain | |
| Talbi et al. | A hybrid technique of image denoising using the curvelet transform based denoising method and two-stage image denoising by PCA with local pixel grouping | |
| CN102801928A (en) | Image processing method and image processing equipment |
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 |