CN116109501A - Low-illumination image sequence enhancement method, device, electronic equipment and storage medium - Google Patents
Low-illumination image sequence enhancement method, device, electronic equipment and storage medium Download PDFInfo
- Publication number
- CN116109501A CN116109501A CN202211638694.2A CN202211638694A CN116109501A CN 116109501 A CN116109501 A CN 116109501A CN 202211638694 A CN202211638694 A CN 202211638694A CN 116109501 A CN116109501 A CN 116109501A
- Authority
- CN
- China
- Prior art keywords
- vector
- decomposition
- feature
- value
- characteristic
- 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
- 238000005286 illumination Methods 0.000 title claims abstract description 58
- 238000000034 method Methods 0.000 title claims abstract description 54
- 239000013598 vector Substances 0.000 claims abstract description 335
- 239000011159 matrix material Substances 0.000 claims abstract description 177
- 238000000354 decomposition reaction Methods 0.000 claims abstract description 151
- 238000012545 processing Methods 0.000 claims abstract description 41
- 238000005070 sampling Methods 0.000 claims abstract description 29
- 238000004364 calculation method Methods 0.000 claims description 20
- 238000004590 computer program Methods 0.000 claims description 11
- 230000006870 function Effects 0.000 claims description 11
- 230000003416 augmentation Effects 0.000 claims description 9
- 230000002708 enhancing effect Effects 0.000 claims description 6
- 230000009466 transformation Effects 0.000 claims description 3
- 238000004891 communication Methods 0.000 description 8
- 238000004458 analytical method Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 238000007726 management method Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 238000010191 image analysis Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 230000008859 change Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 3
- 238000003672 processing method Methods 0.000 description 3
- 238000003384 imaging method Methods 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000005457 optimization Methods 0.000 description 2
- 230000002093 peripheral effect Effects 0.000 description 2
- 238000003491 array Methods 0.000 description 1
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000000295 complement effect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000000802 evaporation-induced self-assembly Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012163 sequencing technique Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 230000000007 visual effect Effects 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
- G06T5/90—Dynamic range modification of images or parts thereof
- G06T5/92—Dynamic range modification of images or parts thereof based on global image properties
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Theoretical Computer Science (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Mathematical Analysis (AREA)
- Data Mining & Analysis (AREA)
- Computational Mathematics (AREA)
- Computing Systems (AREA)
- Algebra (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Image Processing (AREA)
Abstract
The invention relates to the field of image enhancement, and discloses a low-illumination image sequence enhancement method, a device, electronic equipment and a storage medium, wherein the method comprises the following steps: calculating a component image matrix of the sampling image, constructing a data matrix of the component image matrix, calculating a covariance matrix of the data matrix, and performing feature decomposition on the covariance matrix; determining a feature value recursion form and a feature vector recursion form of the decomposed feature vector, calculating an updated reference value of the decomposed feature value, and calculating an updated reference vector of the decomposed feature vector; updating the characteristic value of the decomposed characteristic value to obtain an updated characteristic value, and updating the characteristic vector of the decomposed characteristic vector to obtain an updated characteristic vector; and identifying a local vector space of the updated feature vector, performing vector orthogonal projection on the updated feature vector by utilizing the local vector space to obtain an orthogonal projection vector, and performing illumination enhancement processing on the sampled image to obtain an enhanced illumination image. The invention can realize the rapid processing of single frame images.
Description
Technical Field
The present invention relates to the field of image enhancement, and in particular, to a method and apparatus for enhancing a low-luminance image sequence, an electronic device, and a storage medium.
Background
The low-illumination image sequence enhancement refers to the omnibearing enhancement of the low-illumination intensity image in terms of color, saturation, measurement and the like.
At present, the single-frame image processing low-illumination image is affected by illumination conditions, so that the imaging quality is lower, the imaging of an object is more fuzzy, further use of the image is affected, and the common RGB (red, green and blue) representation method for image enhancement cannot accurately reflect the illumination condition of the image, so that the brightness of an original image is not improved by image processing; the image enhancement method of HSV can reflect the brightness of the image, but the calculation complexity is often too high, and the single-frame image cannot be rapidly processed. Therefore, a scheme for enhancing a low-illumination image is needed to realize rapid processing of a single frame image by reducing the complexity of image analysis and calculation.
Disclosure of Invention
In order to solve the problems, the invention provides a low-illumination image sequence enhancement method, a device, an electronic device and a storage medium, which can realize the rapid processing of a single frame image by reducing the complexity of image analysis and calculation.
In a first aspect, the present invention provides a method for enhancing a low-luminance image sequence, including:
acquiring a sampling image, calculating a component image matrix of the sampling image, constructing a data matrix of the component image matrix, calculating a covariance matrix of the data matrix, and carrying out feature decomposition on the covariance matrix to obtain a decomposition feature value and a decomposition feature vector;
determining a feature value recursion form of the decomposition feature value and a feature vector recursion form of the decomposition feature vector, calculating an update reference value of the decomposition feature value, and calculating an update reference vector of the decomposition feature vector;
according to the feature value recursion form and the updating reference value, updating the feature value of the decomposition feature value to obtain an updating feature value, and according to the feature vector recursion form and the updating reference vector, updating the feature vector of the decomposition feature vector to obtain an updating feature vector;
and identifying a local vector space of the updated feature vector according to the updated feature value, carrying out vector orthogonal projection on the updated feature vector by utilizing the local vector space to obtain an orthogonal projection vector, and carrying out illumination enhancement processing on the sampled image according to the orthogonal projection vector to obtain an enhanced illumination image.
In one possible implementation manner of the first aspect, the performing feature decomposition on the covariance matrix to obtain a decomposition feature value and a decomposition feature vector includes:
identifying an initial covariance matrix of the covariance matrices;
performing feature decomposition on the initial covariance matrix by using the following formula to obtain an initial decomposition value and an initial decomposition vector:
wherein ,σi (1) Representing the ith eigenvalue, u, of the initial decomposition values i (1) Representing an ith eigenvector in the initial decomposition vector, wherein C (1) represents a covariance matrix at the 1 st moment, namely the initial covariance matrix, L represents the total number of eigenvalues, i represents the serial numbers of the eigenvalues and the eigenvectors, and H represents vector complex conjugate transpose;
based on the initial decomposition value and the initial decomposition vector, determining a covariance matrix to be updated of the initial covariance matrix by using the following formula:
wherein C (k) represents the covariance matrix to be updated, k represents the time, alpha represents a weight between 0 and 1, x k Column vectors of the component image matrix at the kth time are represented, and H represents vector complex conjugate transpose;
and determining the decomposition characteristic value and the decomposition characteristic vector according to the covariance matrix to be updated.
In a possible implementation manner of the first aspect, the determining the feature value recursion form of the decomposition feature value and the feature vector recursion form of the decomposition feature vector includes:
and constructing the characteristic relation between the decomposition characteristic value and the decomposition characteristic vector by using the following formula:
wherein ,
representing the characteristic relation, sigma i (k) The ith eigenvalue symbol representing the kth moment covariance matrix C (k), u i (k) The i-th eigenvector symbol representing the k-th covariance matrix C (k), H representing the vector complex conjugate transpose, i representing the eigenvalue and the eigenvector number, j representing the eigenvector number distinguished from i, δ ij A kronecker function, C (k) representing the covariance matrix to be updated;
and carrying out Taylor expansion processing on the characteristic relation by using the following formula to obtain the characteristic value recursion form and the characteristic vector recursion form:
wherein ,
representing a recursive form of the characteristic values,
representing the recursive form of the feature vector, O (ε) 2 ) Representing epsilon 2 Is infinitely small in the higher order of (c),
represents +.>
From sigma at time k-1 i Obtained (I)>
Represents +.>
From sigma at time k-1 i Obtained (I)>
Represents +.>
From u at time k-1 i Obtained (I) >
Representing an updated reference value->
The update reference vector is represented, i represents the eigenvalue and the serial number of the eigenvector.
In a possible implementation manner of the first aspect, the calculating the updated reference value of the decomposition feature value includes:
calculating an updated reference value for the decomposition feature value using the formula:
wherein ,
representing the updated reference value, C (k-1) representing the covariance matrix at time k-1, H representing the complex conjugate transpose of the vector, i representing the eigenvalues and the serial numbers of the eigenvectors，x k Column vector of the component image matrix representing the kth moment, k representing the moment, +.>
Represents +.>
From u at time k-1 i Obtained (I)>
Representing the time of k
In a possible implementation manner of the first aspect, the calculating the updated reference vector of the decomposition feature vector includes:
inquiring a current data matrix and a preamble moment space base corresponding to the decomposition feature vector;
calculating the Euclidean distance between the current data matrix and the preamble moment space base;
selecting a target space base from the preamble moment space bases based on the Euclidean distance;
calculating a vector inner product between the current data matrix and the target space base;
according to the target space base and the vector inner product, the updated reference vector is calculated by using the following formula:
wherein ,
representing the updated reference vector, u i (k-1) represents a feature vector at time k-1, k represents time, Σ j∈{g,h,k,l,m} β j u j U in (b) g ,u h ,u k ,u l ,u m Representing the 5 basis vectors closest to x (k) Euclidean distance in U (k-1) space, i.e. the target space basis, beta j Representing the vector inner product.
In a possible implementation manner of the first aspect, the identifying, according to the updated feature value, a local vector space of the updated feature vector includes:
sorting the updated characteristic values to obtain a characteristic value sequence;
selecting a target characteristic value from the characteristic value sequence;
inquiring a target feature vector from the updated feature vector according to the target feature value;
and constructing the local vector space by utilizing the target feature vector.
In a possible implementation manner of the first aspect, the performing, according to the orthogonal projection vector, an illumination enhancement process on the sampled image to obtain an enhanced illumination image includes:
performing dimension augmentation treatment on the orthogonal projection vector to obtain a dimension augmentation vector;
performing inverse column vectorization processing on the dimension augmentation vector to obtain an inverse column vectorization vector;
and carrying out inverse color transformation on the inverse column vectorization vector to obtain the enhanced illumination image.
In a second aspect, the present invention provides a low-luminance image sequence enhancement apparatus, the apparatus comprising:
the feature decomposition module is used for acquiring a sampling image, calculating a component image matrix of the sampling image, constructing a data matrix of the component image matrix, calculating a covariance matrix of the data matrix, and performing feature decomposition on the covariance matrix to obtain a decomposition feature value and a decomposition feature vector;
the vector calculation module is used for determining a characteristic value recursion form of the decomposition characteristic value and a characteristic vector recursion form of the decomposition characteristic vector, calculating an update reference value of the decomposition characteristic value and calculating an update reference vector of the decomposition characteristic vector;
the vector updating module is used for updating the characteristic value of the decomposition characteristic value according to the characteristic value recurrence form and the updating reference value to obtain an updating characteristic value, and updating the characteristic vector of the decomposition characteristic vector according to the characteristic vector recurrence form and the updating reference vector to obtain an updating characteristic vector;
and the illumination enhancement module is used for identifying a local vector space of the updating feature vector according to the updating feature value, carrying out vector orthogonal projection on the updating feature vector by utilizing the local vector space to obtain an orthogonal projection vector, and carrying out illumination enhancement processing on the sampling image according to the orthogonal projection vector to obtain an enhanced illumination image.
In a third aspect, the present invention provides an electronic device comprising:
at least one processor; and a memory communicatively coupled to the at least one processor;
wherein the memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the low-light image sequence enhancement method as described in any one of the first aspects above.
In a fourth aspect, the present invention provides a computer readable storage medium storing a computer program which, when executed by a processor, implements the low-intensity image sequence enhancement method according to any one of the first aspects above.
Compared with the prior art, the technical principle and beneficial effect of this scheme lie in:
the embodiment of the invention firstly calculates the component image matrix of the sampled image to distinguish from the traditional enhancement processing method for carrying out three components of H component, S component and V component on the image, adopts the method for carrying out enhancement processing on only the V component of the image, improves the simplicity of image enhancement processing, and further, the embodiment of the invention constructs the data matrix of the component image matrix to be used for arranging the column vectors of the component image matrix at different moments into the data matrix based on the data matrix With the generation of new data of an image sequence, the characteristic of dynamic change along with time change is often adopted to realize high-efficiency analysis of the data at each sampling moment, further, the embodiment of the invention calculates the covariance matrix of the data matrix to obtain the characteristic representation of the data matrix by utilizing the covariance matrix, further, the embodiment of the invention reduces the calculation complexity of the characteristic decomposition of the covariance matrix at the moment by carrying out characteristic decomposition on the covariance matrix, and carries out recursive calculation on the characteristic data of the characteristic space of the image covariance matrix at the later moment according to the time sequence of image sampling according to the previous moment, thereby avoiding the complexity of the characteristic decomposition at each sampling moment; secondly, the embodiment of the invention realizes the recursive calculation and updating of the characteristic data at the next moment based on the characteristic data at the previous moment by determining the characteristic value recursive form of the decomposition characteristic value and the characteristic vector recursive form of the decomposition characteristic vector, further, the embodiment of the invention calculates the updating reference value of the decomposition characteristic value for reversely updating the characteristic value at the next moment by using the updating reference value, further, the embodiment of the invention calculates the updating reference vector of the decomposition characteristic vector for improving the processing efficiency of the low-illumination image sequence and rapidly calculates the characteristic space of the covariance matrix of the data matrix at the current moment by establishing the local basis vector, namely, calculates 
Thereby improving the processing efficiency of the low-illumination image sequence; further, in the embodiment of the present invention, the feature value is updated for the decomposed feature value according to the feature value recursion form and the updated reference value, and the feature vector is updated for the decomposed feature vector according to the feature vector recursion form and the updated reference vector, so as to be used for calculating new feature values and feature vectors through a simpler mathematical optimization method and a numerical calculation method, and in the embodiment of the present invention, the feature value is identified according to the updated feature valueAnd updating the local vector space of the feature vector to perform orthogonal projection on the data in the feature space after performing recursive analysis on the feature space of the data matrix, extracting a low-noise image of the data, and performing projection on the image data in the feature space to strongly suppress noise by using noise, namely information of interference illumination. Therefore, the low-illumination image sequence enhancement method, the low-illumination image sequence enhancement device, the electronic equipment and the storage medium can realize rapid processing of a single frame image by reducing complexity of image analysis and calculation.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description, serve to explain the principles of the invention.
In order to more clearly illustrate the embodiments of the invention or the technical solutions of the prior art, the drawings which are used in the description of the embodiments or the prior art will be briefly described, and it will be obvious to a person skilled in the art that other drawings can be obtained from these drawings without inventive effort.
Fig. 1 is a flowchart of a low-illumination image sequence enhancement method according to an embodiment of the present invention;
FIG. 2 is a flowchart illustrating one of the steps of the low-intensity image sequence enhancement method of FIG. 1 according to an embodiment of the present invention;
FIG. 3 is a flowchart illustrating another step of the method for enhancing a low-intensity image sequence according to the embodiment of the present invention;
fig. 4 is a schematic block diagram of a low-light image sequence enhancement device according to an embodiment of the present invention;
fig. 5 is a schematic diagram of an internal structure of an electronic device for implementing a low-light image sequence enhancement method according to an embodiment of the present invention.
Detailed Description
It should be understood that the detailed description is presented by way of example only and is not intended to limit the invention.
The embodiment of the invention provides a low-illumination image sequence enhancement method, and an execution subject of the low-illumination image sequence enhancement method comprises, but is not limited to, at least one of a server, a terminal and the like which can be configured to execute the method provided by the embodiment of the invention. In other words, the low-light image sequence enhancement method may be performed by software or hardware installed in a terminal device or a server device, and the software may be a blockchain platform. The service end includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, and the like. The server may be an independent server, or may be a cloud server that provides cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communications, middleware services, domain name services, security services, content delivery networks (Content Delivery Network, CDN), and basic cloud computing services such as big data and artificial intelligence platforms.
Referring to fig. 1, a flowchart of a method for enhancing a low-light image sequence according to an embodiment of the invention is shown. The low-illumination image sequence enhancement method depicted in fig. 1 includes:
S1, acquiring a sampling image, calculating a component image matrix of the sampling image, constructing a data matrix of the component image matrix, calculating a covariance matrix of the data matrix, and carrying out feature decomposition on the covariance matrix to obtain a decomposition feature value and a decomposition feature vector.
The sampled image in the embodiment of the invention refers to an image sequence, and is composed of single-frame images obtained by sampling at each moment.
Further, the embodiment of the invention calculates the component image matrix of the sampled image to distinguish from the traditional enhancement processing method for carrying out three components of H component, S component and V component on the image, and adopts a method for carrying out enhancement processing on only the V component of the image, thereby improving the simplicity of image enhancement processing. The H component, the S component and the V component respectively represent the hue (H), the saturation (S) and the brightness (V) of the image, the hue H is measured by an angle, the value range is 0-360 degrees, the red, the green and the blue are respectively separated by 120 degrees, and the complementary colors are respectively separated by 180 degrees; the saturation S represents the purity degree of the color, the value range is 0.0-1.0, and when S=0, only the gray scale exists; the brightness V represents the brightness of the color, and the range of the brightness is 0.0 (black) to 1.0 (white). The component image matrix refers to a V component image matrix of the sampled image, for example, a sequence of V components at different moments: p (t=k), P (t=k-1), P (t=1), P (t=k) representing the V-component image matrix at the kth time, abbreviated as P (k), P (t=k-1) representing the image at the kth-1, abbreviated as P (k-1), and P (t=1) representing the image up to the 1 st sampling time.
In one embodiment of the present invention, referring to fig. 2, the calculating the component image matrix of the sampled image includes:
s201, identifying an initial color model of the sampling image;
s202, performing color model conversion on the initial color model to obtain a converted color model;
s203, selecting the component image matrix from the conversion color model.
Wherein the initial color model is an RGB color model, and the conversion color model is an HSV color model.
Furthermore, the embodiment of the invention realizes the efficient analysis of the data at each sampling moment by constructing the data matrix of the component image matrix so as to arrange the column vectors of the component image matrix at different moments into the data matrix based on the characteristic that the data matrix is dynamically changed along with the generation of new data of the image sequence, which is often changed along with the time change. The data matrix refers to a data matrix formed by vectorizing component image matrix arrays at different moments, for example, image matrices P (t=k), P (t=k-1) at k sampling moments, and P (t=1) are vectorized and then arranged into a data matrix d= [ x ] 1 ,x 2 ,…,x k], wherein xk For the column vector corresponding to the image at time k, it is assumed that the sliding window length of the data matrix is T, that is, the number of columns of the data matrix is T. Data moment with the arrival of new images The array always holds the nearest T column vectors, while t+1 and earlier data are discarded.
In an embodiment of the present invention, the constructing the data matrix of the component image matrix includes: performing column vectorization processing on the component image matrix to obtain column vectorized components; identifying a temporal order of the column-vectorized components; and constructing a data matrix of the component image matrix by utilizing the column vectorized components according to the time sequence.
Further, the embodiment of the invention calculates the covariance matrix of the data matrix to be used for obtaining the characteristic representation of the data matrix by using the covariance matrix.
Further, the embodiment of the invention is used for reducing the calculation complexity of the covariance matrix feature decomposition at the moment by carrying out feature decomposition on the covariance matrix, and carrying out recursive calculation on the feature data of the feature space of the image covariance matrix according to the previous moment at the later moment according to the time sequence of image sampling, so as to avoid the complexity of carrying out feature decomposition at each sampling moment.
In an embodiment of the present invention, the performing feature decomposition on the covariance matrix to obtain a decomposition feature value and a decomposition feature vector includes: identifying an initial covariance matrix of the covariance matrices; performing feature decomposition on the initial covariance matrix by using the following formula to obtain an initial decomposition value and an initial decomposition vector:
wherein ,σi (1) Representing the ith eigenvalue, u, of the initial decomposition values i (1) Representing an ith eigenvector in the initial decomposition vector, wherein C (1) represents a covariance matrix at the 1 st moment, namely the initial covariance matrix, L represents the total number of eigenvalues, i represents the serial numbers of the eigenvalues and the eigenvectors, and H represents vector complex conjugate transpose;
based on the initial decomposition value and the initial decomposition vector, determining a covariance matrix to be updated of the initial covariance matrix by using the following formula:
wherein C (k) represents the covariance matrix to be updated, k represents the time, alpha represents a weight between 0 and 1, x k Column vectors of the component image matrix at the kth time are represented, and H represents vector complex conjugate transpose;
and determining the decomposition characteristic value and the decomposition characteristic vector according to the covariance matrix to be updated.
The covariance matrix to be updated refers to a matrix waiting for subsequent updating of the feature value and the feature vector obtained by new calculation in turn. Optionally, the principle of determining the decomposition feature value and the decomposition feature vector according to the covariance matrix to be updated is similar to the principle of performing feature decomposition on the initial covariance matrix to obtain an initial decomposition value and an initial decomposition vector, which is not described in detail herein.
S2, determining a characteristic value recursion form of the decomposition characteristic value and a characteristic vector recursion form of the decomposition characteristic vector, calculating an update reference value of the decomposition characteristic value, and calculating an update reference vector of the decomposition characteristic vector.
The embodiment of the invention determines the characteristic value recursion form of the decomposition characteristic value and the characteristic vector recursion form of the decomposition characteristic vector, so as to realize the recursion calculation and update of the characteristic data at the later moment based on the characteristic data at the former moment by utilizing the characteristic value recursion form and the characteristic vector recursion form.
In an embodiment of the present invention, the determining the feature value recursion form of the decomposition feature value and the feature vector recursion form of the decomposition feature vector includes: and constructing the characteristic relation between the decomposition characteristic value and the decomposition characteristic vector by using the following formula:
wherein ,
representing the characteristic relation, sigma i (k) The ith eigenvalue symbol representing the kth moment covariance matrix C (k), u i (k) The i-th eigenvector symbol representing the k-th covariance matrix C (k), H representing the vector complex conjugate transpose, i representing the eigenvalue and the eigenvector number, j representing the eigenvector number distinguished from i, δ ij A kronecker function, C (k) representing the covariance matrix to be updated;
and carrying out Taylor expansion processing on the characteristic relation by using the following formula to obtain the characteristic value recursion form and the characteristic vector recursion form:
wherein ,
representing a recursive form of the characteristic values,
representing the recursive form of the feature vector, O (ε) 2 ) Representing epsilon 2 Is infinitely small in the higher order of (c),
represents +.>
From sigma at time k-1 i Obtained (I)>
Represents +.>
From sigma at time k-1 i Obtained (I)>
Represents +.>
From u at time k-1 i Obtained (I)>
Indicating that the reference value is to be updated,
the update reference vector is represented, i represents the eigenvalue and the serial number of the eigenvector.
Further, the embodiment of the invention calculates the updated reference value of the decomposition feature value to be used for updating the feature value at the next moment in turn by using the updated reference value.
In an embodiment of the present invention, the calculating the updated reference value of the decomposition feature value includes: calculating an updated reference value for the decomposition feature value using the formula:
wherein ,
representing the updated reference value, C (k-1) representing the covariance matrix at time k-1, H representing the complex conjugate transpose of the vector, i representing the eigenvalues and the serial numbers of the eigenvectors, x k Column vector of the component image matrix representing the kth moment, k representing the moment, +.>
Represents +.>
From u at time k-1 i Obtained (I)>
Represents +.>
/>
Further, the embodiment of the invention calculates the update reference vector of the decomposition feature vector to be used for improving the processing efficiency of the low-illumination image sequence, and rapidly calculates the feature space of the covariance matrix of the data matrix at the current moment by establishing the local basis vector, namely, calculates
Thereby improving the processing efficiency of the low-illumination image sequence.
In an embodiment of the present invention, the calculating the updated reference vector of the decomposition feature vector includes: inquiring a current data matrix and a preamble moment space base corresponding to the decomposition feature vector; calculating the Euclidean distance between the current data matrix and the preamble moment space base; selecting a target space base from the preamble moment space bases based on the Euclidean distance; calculating a vector inner product between the current data matrix and the target space base; according to the target space base and the vector inner product, the updated reference vector is calculated by using the following formula:
wherein ,
representing the updated reference vector, u i (k-1) represents a feature vector at time k-1, k represents time, Σ j∈{g,h,k,l,m} β j u j U in (b) g ,u h ,u k ,u l ,u m Representing the 5 basis vectors closest to x (k) Euclidean distance in U (k-1) space, i.e. the target space basis, beta j Representing the vector inner product.
Illustratively, the matrix U (k-1) that is the basis of the covariance matrix eigenspace at time k-1 is U (k-1) = [ U ] 1 (k-1),u 2 (k-1),...,u L (k-1)]Let the number of vectors contained in the local basis vector set be a small constant, e.g. 5, i.e. 5 local basis vectors, then, according to the definition of 0 norm, the approximate approximation is used to derive an approximate solution of y (k) as y (k) =β g u g +β h u h +β k u k +β l u l +β m u m, wherein ug ,u h ,u k ,u l ,u m The method comprises the steps of representing 5 base vectors closest to the Euclidean distance of x (k) in the U (k-1) space, calculating the Euclidean distance between each base vector in the x (k) and the U (k-1), sequencing, and taking the 5 base vectors with the smallest distance. Beta represents the corresponding linear combination coefficient, and the calculation method is x (k) and u g ,u h ,u k ,u l ,u m Respectively, inner products, e.g. beta g ＝x H* u g Then
And S3, updating the characteristic value of the decomposed characteristic value according to the characteristic value recurrence form and the updated reference value to obtain an updated characteristic value, and updating the characteristic vector of the decomposed characteristic vector according to the characteristic vector recurrence form and the updated reference vector to obtain an updated characteristic vector.
According to the embodiment of the invention, the characteristic value of the decomposed characteristic value is updated according to the characteristic value recurrence form and the updated reference value, and the characteristic vector of the decomposed characteristic vector is updated according to the characteristic vector recurrence form and the updated reference vector, so that the method is used for calculating new characteristic values and characteristic vectors through a simpler mathematical optimization method and a numerical value calculation method.
And S4, identifying a local vector space of the updated feature vector according to the updated feature value, carrying out vector orthogonal projection on the updated feature vector by utilizing the local vector space to obtain an orthogonal projection vector, and carrying out illumination enhancement processing on the sampled image according to the orthogonal projection vector to obtain an enhanced illumination image.
According to the embodiment of the invention, the local vector space of the updated feature vector is identified according to the updated feature value, so that after recursive analysis is performed on the feature space of the data matrix, data is orthogonally projected in the feature space, a low-noise image of the data is extracted, noise, namely information of interference illumination, is extracted, and the image data is projected in the feature space, so that the noise is strongly suppressed.
In an embodiment of the present invention, referring to fig. 3, the identifying, according to the updated feature value, a local vector space of the updated feature vector includes:
s301, sorting the updated characteristic values to obtain a characteristic value sequence;
s302, selecting a target characteristic value from the characteristic value sequence;
s303, inquiring a target feature vector from the updated feature vector according to the target feature value;
s304, constructing the local vector space by utilizing the target feature vector.
Illustratively, the feature values are arranged from large to small according to the size of the feature values, and a proper number of feature values is taken, for example, the first 10% of feature values are taken from the number, or the feature values are summed to obtain the first n feature values corresponding to the first 90% of the sum; the matrix formed by the corresponding eigenvectors is marked as U', the column vectors are the corresponding eigenvectors obtained by selecting eigenvalues, the eigenvectors Zhang Chengju are linear spaces, namely vector spaces where signals are located are obtained, and the rest eigenvectors are tensed, namely noise vector spaces.
Further, the embodiment of the invention performs vector orthogonal projection on the updated feature vector by utilizing the local vector space, so as to be used for projecting the image data in the feature space, thereby strongly suppressing the noise of the image.
In an embodiment of the present invention, the performing vector orthogonal projection on the updated feature vector by using the local vector space to obtain an orthogonal projection vector includes: according to the local vector space, vector orthogonal projection is carried out on the updated feature vector by using the following formula to obtain the orthogonal projection vector:
X′(k)＝U′*U′ H *X(k)
wherein X '(k) represents the orthogonal projection vector, X (k) represents the update feature vector, U' represents the local vector space, H represents a vector complex conjugate transpose, and k represents time.
In an embodiment of the present invention, the performing, according to the orthogonal projection vector, illumination enhancement processing on the sampled image to obtain an enhanced illumination image includes: performing dimension augmentation treatment on the orthogonal projection vector to obtain a dimension augmentation vector; performing inverse column vectorization processing on the dimension augmentation vector to obtain an inverse column vectorization vector; and carrying out inverse color transformation on the inverse column vectorization vector to obtain the enhanced illumination image.
For example, the other dimensions of the orthogonal projection U '×u' h×x (k) are filled with 0 until the dimensions of the orthogonal projection are complemented with n, i.e., a vector extending into n dimensions is denoted as X "(k), the X" (k) is reversely vectorized and transformed into a matrix form denoted as P "(k), a V component image matrix after low-illumination enhancement is obtained, denoted as V" (k), and the V "(k) is inversely HSV transformed with the H and S components obtained in the initial HSV decomposition process to obtain a new image after low-illumination enhancement.
It can be seen that embodiments of the present invention first calculate the sampled imageThe component image matrix is used for distinguishing from the traditional enhancement processing method for carrying out H component, S component, V component and the like on an image, the simplicity of the image enhancement processing is improved by adopting a method for carrying out enhancement processing on only the V component of the image, further, the embodiment of the invention is used for reducing the computation complexity of the feature decomposition of the covariance matrix at the moment by constructing the data matrix of the component image matrix, arranging column vectors of the component image matrix at different moments into the data matrix, and carrying out the recursive computation according to the previous moment on the feature data of the feature space of the covariance matrix at the moment according to the time sequence of image sampling so as to realize the efficient analysis on the data at each sampling moment, and further, the embodiment of the invention is used for obtaining the feature representation of the data matrix by calculating the covariance matrix; secondly, the embodiment of the invention realizes the recursive calculation and updating of the characteristic data at the next moment based on the characteristic data at the previous moment by determining the characteristic value recursive form of the decomposition characteristic value and the characteristic vector recursive form of the decomposition characteristic vector, further, the embodiment of the invention calculates the updating reference value of the decomposition characteristic value for reversely updating the characteristic value at the next moment by using the updating reference value, further, the embodiment of the invention calculates the updating reference vector of the decomposition characteristic vector for improving the processing efficiency of the low-illumination image sequence and rapidly calculates the characteristic space of the covariance matrix of the data matrix at the current moment by establishing the local basis vector, namely, calculates 
Thereby improving the processing efficiency of the low-illumination image sequence; further, the embodiment of the invention is generalThe embodiment of the invention identifies the local vector space of the updated feature vector according to the updated feature value, so as to be used for carrying out orthogonal projection on data in the feature space after carrying out recursive analysis on the feature space of the data matrix, extracting a low-noise image of the data, and projecting noise, namely information of interference illumination, in the feature space by carrying out projection on the image data. Therefore, the low-illumination image sequence enhancement method provided by the embodiment of the invention can realize the rapid processing of a single frame image by reducing the complexity of image analysis and calculation.
As shown in fig. 4, a functional block diagram of the low-light image sequence enhancement device of the present invention is shown.
The low-illuminance image sequence enhancement apparatus 400 of the present invention may be installed in an electronic device. Depending on the implemented functionality, the low-light image sequence enhancement means may comprise a feature decomposition module 401, a vector calculation module 402, a vector update module 403, and a light enhancement module 404. The module of the invention, which may also be referred to as a unit, refers to a series of computer program segments, which are stored in the memory of the electronic device, capable of being executed by the processor of the electronic device and of performing a fixed function.
In the embodiment of the present invention, the functions of each module/unit are as follows:
the feature decomposition module 401 is configured to obtain a sampled image, calculate a component image matrix of the sampled image, construct a data matrix of the component image matrix, calculate a covariance matrix of the data matrix, and perform feature decomposition on the covariance matrix to obtain a decomposition feature value and a decomposition feature vector;
the vector calculation module 402 is configured to determine a feature value recursive form of the decomposed feature value and a feature vector recursive form of the decomposed feature vector, calculate an updated reference value of the decomposed feature value, and calculate an updated reference vector of the decomposed feature vector;
The vector update module 403 is configured to update the feature value of the decomposed feature value according to the feature value recurrence form and the update reference value to obtain an updated feature value, and update the feature vector of the decomposed feature vector according to the feature vector recurrence form and the update reference vector to obtain an updated feature vector;
the illuminance enhancement module 404 is configured to identify a local vector space of the updated feature vector according to the updated feature value, perform vector orthogonal projection on the updated feature vector by using the local vector space to obtain an orthogonal projection vector, and perform illuminance enhancement processing on the sampled image according to the orthogonal projection vector to obtain an enhanced illuminance image.
In detail, the modules in the low-illumination image sequence enhancement device 400 in the embodiment of the present invention use the same technical means as the low-illumination image sequence enhancement method described in fig. 1 to 3 and can produce the same technical effects, which are not described herein.
Fig. 5 is a schematic structural diagram of an electronic device for implementing the low-illumination image sequence enhancement method according to the present invention.
The electronic device may comprise a processor 50, a memory 51, a communication bus 52 and a communication interface 53, and may further comprise a computer program, such as a low-light image sequence enhancement program, stored in the memory 51 and executable on the processor 50.
The processor 50 may be formed by an integrated circuit in some embodiments, for example, a single packaged integrated circuit, or may be formed by a plurality of integrated circuits packaged with the same function or different functions, including one or more central processing units (Central Processing unit, CPU), a microprocessor, a digital processing chip, a graphics processor, a combination of various control chips, and so on. The processor 50 is a Control Unit (Control Unit) of the electronic device, connects various components of the entire electronic device using various interfaces and lines, and executes various functions of the electronic device and processes data by running or executing programs or modules (e.g., executing a low-light image sequence enhancement program, etc.) stored in the memory 51, and calling data stored in the memory 51.
The memory 51 includes at least one type of readable storage medium including flash memory, a removable hard disk, a multimedia card, a card type memory (e.g., SD or DX memory, etc.), a magnetic memory, a magnetic disk, an optical disk, etc. The memory 51 may in some embodiments be an internal storage unit of the electronic device, such as a mobile hard disk of the electronic device. The memory 51 may also be an external storage device of the electronic device in other embodiments, for example, a plug-in mobile hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card) or the like. Further, the memory 51 may also include both an internal storage unit and an external storage device of the electronic device. The memory 51 may be used not only for storing application software installed in an electronic device and various types of data, such as codes of a database-configured connection program, but also for temporarily storing data that has been output or is to be output.
The communication bus 52 may be a peripheral component interconnect standard (peripheral component interconnect, PCI) bus, or an extended industry standard architecture (extended industry standard architecture, EISA) bus, among others. The bus may be classified as an address bus, a data bus, a control bus, etc. The bus is arranged to enable a connection communication between the memory 51 and at least one processor 50 etc.
The communication interface 53 is used for communication between the electronic device 5 and other devices, including a network interface and a user interface. Optionally, the network interface may include a wired interface and/or a wireless interface (e.g., WI-FI interface, bluetooth interface, etc.), typically used to establish a communication connection between the electronic device and other electronic devices. The user interface may be a Display (Display), an input unit such as a Keyboard (Keyboard), or alternatively a standard wired interface, a wireless interface. Alternatively, in some embodiments, the display may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, an OLED (Organic Light-Emitting Diode) touch, or the like. The display may also be referred to as a display screen or display unit, as appropriate, for displaying information processed in the electronic device and for displaying a visual user interface.
Fig. 5 shows only an electronic device with components, and it will be understood by those skilled in the art that the structure shown in fig. 5 is not limiting of the electronic device and may include fewer or more components than shown, or may combine certain components, or a different arrangement of components.
For example, although not shown, the electronic device may further include a power source (such as a battery) for supplying power to the respective components, and the power source may be logically connected to the at least one processor 50 through a power management device, so that functions of charge management, discharge management, and power consumption management are implemented through the power management device. The power supply may also include one or more of any of a direct current or alternating current power supply, recharging device, power failure detection circuit, power converter or inverter, power status indicator, etc. The electronic device may further include various sensors, bluetooth modules, wi-Fi modules, etc., which are not described herein.
It should be understood that the embodiments described are for illustrative purposes only and are not limited in scope by this configuration.
The database-configured connection program stored in the memory 51 in the electronic device is a combination of a plurality of computer programs, which, when run in the processor 50, can implement:
Acquiring a sampling image, calculating a component image matrix of the sampling image, constructing a data matrix of the component image matrix, calculating a covariance matrix of the data matrix, and carrying out feature decomposition on the covariance matrix to obtain a decomposition feature value and a decomposition feature vector;
determining a feature value recursion form of the decomposition feature value and a feature vector recursion form of the decomposition feature vector, calculating an update reference value of the decomposition feature value, and calculating an update reference vector of the decomposition feature vector;
according to the feature value recursion form and the updating reference value, updating the feature value of the decomposition feature value to obtain an updating feature value, and according to the feature vector recursion form and the updating reference vector, updating the feature vector of the decomposition feature vector to obtain an updating feature vector;
and identifying a local vector space of the updated feature vector according to the updated feature value, carrying out vector orthogonal projection on the updated feature vector by utilizing the local vector space to obtain an orthogonal projection vector, and carrying out illumination enhancement processing on the sampled image according to the orthogonal projection vector to obtain an enhanced illumination image.
In particular, the specific implementation method of the processor 50 on the computer program may refer to the description of the relevant steps in the corresponding embodiment of fig. 1, which is not repeated herein.
Further, the electronic device integrated modules/units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a non-volatile computer readable storage medium. The storage medium may be volatile or nonvolatile. For example, the computer readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer Memory, a Read-Only Memory (ROM).
The present invention also provides a storage medium storing a computer program which, when executed by a processor of an electronic device, can implement:
acquiring a sampling image, calculating a component image matrix of the sampling image, constructing a data matrix of the component image matrix, calculating a covariance matrix of the data matrix, and carrying out feature decomposition on the covariance matrix to obtain a decomposition feature value and a decomposition feature vector;
Determining a feature value recursion form of the decomposition feature value and a feature vector recursion form of the decomposition feature vector, calculating an update reference value of the decomposition feature value, and calculating an update reference vector of the decomposition feature vector;
according to the feature value recursion form and the updating reference value, updating the feature value of the decomposition feature value to obtain an updating feature value, and according to the feature vector recursion form and the updating reference vector, updating the feature vector of the decomposition feature vector to obtain an updating feature vector;
and identifying a local vector space of the updated feature vector according to the updated feature value, carrying out vector orthogonal projection on the updated feature vector by utilizing the local vector space to obtain an orthogonal projection vector, and carrying out illumination enhancement processing on the sampled image according to the orthogonal projection vector to obtain an enhanced illumination image.
In the several embodiments provided in the present invention, it should be understood that the disclosed apparatus, device and method may be implemented in other manners. For example, the above-described apparatus embodiments are merely illustrative, and for example, the division of the modules is merely a logical function division, and there may be other manners of division when actually implemented.
The modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical units, may be located in one place, or may be distributed over multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional module 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 can be realized in a form of hardware or a form of hardware and a form of software functional modules.
It will be evident to those skilled in the art that the invention is not limited to the details of the foregoing illustrative embodiments, and that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof.
The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference signs in the claims shall not be construed as limiting the claim concerned.
It should be noted that in this document, relational terms such as "first" and "second" and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element.
The foregoing is only a specific embodiment of the invention to enable those skilled in the art to understand or practice the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (10)
1. A method of enhancing a sequence of low-intensity images, the method comprising:
acquiring a sampling image, calculating a component image matrix of the sampling image, constructing a data matrix of the component image matrix, calculating a covariance matrix of the data matrix, and carrying out feature decomposition on the covariance matrix to obtain a decomposition feature value and a decomposition feature vector;
determining a feature value recursion form of the decomposition feature value and a feature vector recursion form of the decomposition feature vector, calculating an update reference value of the decomposition feature value, and calculating an update reference vector of the decomposition feature vector;
according to the feature value recursion form and the updating reference value, updating the feature value of the decomposition feature value to obtain an updating feature value, and according to the feature vector recursion form and the updating reference vector, updating the feature vector of the decomposition feature vector to obtain an updating feature vector;
and identifying a local vector space of the updated feature vector according to the updated feature value, carrying out vector orthogonal projection on the updated feature vector by utilizing the local vector space to obtain an orthogonal projection vector, and carrying out illumination enhancement processing on the sampled image according to the orthogonal projection vector to obtain an enhanced illumination image.
2. The method of claim 1, wherein the performing feature decomposition on the covariance matrix to obtain a decomposition feature value and a decomposition feature vector comprises:
identifying an initial covariance matrix of the covariance matrices;
performing feature decomposition on the initial covariance matrix by using the following formula to obtain an initial decomposition value and an initial decomposition vector:
wherein ,σi (1) Representing the ith eigenvalue, u, of the initial decomposition values i (1) Representing an ith eigenvector in the initial decomposition vector, wherein C (1) represents a covariance matrix at the 1 st moment, namely the initial covariance matrix, L represents the total number of eigenvalues, i represents the serial numbers of the eigenvalues and the eigenvectors, and H represents vector complex conjugate transpose;
based on the initial decomposition value and the initial decomposition vector, determining a covariance matrix to be updated of the initial covariance matrix by using the following formula:
wherein C (k) represents the covariance matrix to be updated, k represents the time, alpha represents a weight between 0 and 1, x k Column vectors of the component image matrix at the kth time are represented, and H represents vector complex conjugate transpose;
and determining the decomposition characteristic value and the decomposition characteristic vector according to the covariance matrix to be updated.
3. The method of claim 1, wherein said determining a eigenvalue recursion version of the decomposed eigenvalue and an eigenvector recursion version of the decomposed eigenvector comprises:
and constructing the characteristic relation between the decomposition characteristic value and the decomposition characteristic vector by using the following formula:
wherein ,
representing the characteristic relation, sigma i (k) The ith eigenvalue symbol representing the kth moment covariance matrix C (k)Number u i (k) The i-th eigenvector symbol representing the k-th covariance matrix C (k), H representing the vector complex conjugate transpose, i representing the eigenvalue and the eigenvector number, j representing the eigenvector number distinguished from i, δ ij A kronecker function, C (k) representing the covariance matrix to be updated;
and carrying out Taylor expansion processing on the characteristic relation by using the following formula to obtain the characteristic value recursion form and the characteristic vector recursion form:
the quantity i represents the sequence numbers of the eigenvalues and eigenvectors.
4. The method of claim 1, wherein said calculating an updated reference value for the decomposition feature value comprises:
calculating an updated reference value for the decomposition feature value using the formula:
wherein ,
representing the updated reference value, C (k-1) representing the covariance matrix at time k-1, H representing the complex conjugate transpose of the vector, i representing the eigenvalues and the serial numbers of the eigenvectors, x k A column vector of the component image matrix representing the kth time, k representing the time,
represents +.>
From u at time k-1 i Obtained (I)>
Represents +.>
5. The method of claim 1, wherein said calculating an updated reference vector for the decomposed feature vector comprises:
inquiring a current data matrix and a preamble moment space base corresponding to the decomposition feature vector;
calculating the Euclidean distance between the current data matrix and the preamble moment space base;
selecting a target space base from the preamble moment space bases based on the Euclidean distance;
calculating a vector inner product between the current data matrix and the target space base;
according to the target space base and the vector inner product, the updated reference vector is calculated by using the following formula:
wherein ,
representing the updated reference vector, u i (k-1) represents a feature vector at time k-1, k represents time, Σ j∈{g,h,k,l,m} β j u j U in (b) g ,u h ,u k ,u l ,u m Representing the 5 basis vectors closest to x (k) Euclidean distance in U (k-1) space, i.e. the target space basis, beta j Representing the vector inner product.
6. The method of claim 1, wherein the identifying the local vector space of the updated feature vector from the updated feature value comprises:
Sorting the updated characteristic values to obtain a characteristic value sequence;
selecting a target characteristic value from the characteristic value sequence;
inquiring a target feature vector from the updated feature vector according to the target feature value;
and constructing the local vector space by utilizing the target feature vector.
7. The method according to claim 1, wherein the performing illumination enhancement processing on the sampled image according to the orthogonal projection vector to obtain an enhanced illumination image includes:
performing dimension augmentation treatment on the orthogonal projection vector to obtain a dimension augmentation vector;
performing inverse column vectorization processing on the dimension augmentation vector to obtain an inverse column vectorization vector;
and carrying out inverse color transformation on the inverse column vectorization vector to obtain the enhanced illumination image.
8. A low-light image sequence enhancement device, the device comprising:
the feature decomposition module is used for acquiring a sampling image, calculating a component image matrix of the sampling image, constructing a data matrix of the component image matrix, calculating a covariance matrix of the data matrix, and performing feature decomposition on the covariance matrix to obtain a decomposition feature value and a decomposition feature vector;
The vector calculation module is used for determining a characteristic value recursion form of the decomposition characteristic value and a characteristic vector recursion form of the decomposition characteristic vector, calculating an update reference value of the decomposition characteristic value and calculating an update reference vector of the decomposition characteristic vector;
the vector updating module is used for updating the characteristic value of the decomposition characteristic value according to the characteristic value recurrence form and the updating reference value to obtain an updating characteristic value, and updating the characteristic vector of the decomposition characteristic vector according to the characteristic vector recurrence form and the updating reference vector to obtain an updating characteristic vector;
and the illumination enhancement module is used for identifying a local vector space of the updating feature vector according to the updating feature value, carrying out vector orthogonal projection on the updating feature vector by utilizing the local vector space to obtain an orthogonal projection vector, and carrying out illumination enhancement processing on the sampling image according to the orthogonal projection vector to obtain an enhanced illumination image.
9. An electronic device, the electronic device comprising:
at least one processor; the method comprises the steps of,
a memory communicatively coupled to the at least one processor; wherein,
The memory stores a computer program executable by the at least one processor to enable the at least one processor to perform the low-light image sequence enhancement method according to any one of claims 1 to 7.
10. A computer readable storage medium storing a computer program, wherein the computer program when executed by a processor implements the low-light image sequence enhancement method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211638694.2A CN116109501A (en) | 2022-12-19 | 2022-12-19 | Low-illumination image sequence enhancement method, device, electronic equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211638694.2A CN116109501A (en) | 2022-12-19 | 2022-12-19 | Low-illumination image sequence enhancement method, device, electronic equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116109501A true CN116109501A (en) | 2023-05-12 |
Family
ID=86255240
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211638694.2A Pending CN116109501A (en) | 2022-12-19 | 2022-12-19 | Low-illumination image sequence enhancement method, device, electronic equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116109501A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105184747A (en) * | 2015-09-09 | 2015-12-23 | 天津光电高斯通信工程技术股份有限公司 | Low illumination image contrast improvement method |
| KR20160056729A (en) * | 2014-11-12 | 2016-05-20 | 고려대학교 산학협력단 | Video quality enhancement device and method for extremely low-light video |
| CN114187222A (en) * | 2021-12-13 | 2022-03-15 | 安徽大学 | Low-illumination image enhancement method and system and storage medium |
| CN114897753A (en) * | 2022-05-13 | 2022-08-12 | 安徽工程大学 | Low-illumination image enhancement method |
| CN115482169A (en) * | 2022-09-26 | 2022-12-16 | 深圳信息职业技术学院 | Low-illumination image enhancement method and device, electronic equipment and storage medium |
-
2022
- 2022-12-19 CN CN202211638694.2A patent/CN116109501A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20160056729A (en) * | 2014-11-12 | 2016-05-20 | 고려대학교 산학협력단 | Video quality enhancement device and method for extremely low-light video |
| CN105184747A (en) * | 2015-09-09 | 2015-12-23 | 天津光电高斯通信工程技术股份有限公司 | Low illumination image contrast improvement method |
| CN114187222A (en) * | 2021-12-13 | 2022-03-15 | 安徽大学 | Low-illumination image enhancement method and system and storage medium |
| CN114897753A (en) * | 2022-05-13 | 2022-08-12 | 安徽工程大学 | Low-illumination image enhancement method |
| CN115482169A (en) * | 2022-09-26 | 2022-12-16 | 深圳信息职业技术学院 | Low-illumination image enhancement method and device, electronic equipment and storage medium |
Non-Patent Citations (2)
| Title |
|---|
| XIAOJIE GUO等: "LIME: Low-Light Image Enhancement via Illumination Map Estimation", IEEE TRANSACTIONS ON IMAGE PROCESSING, vol. 26, no. 2, 14 December 2016 (2016-12-14), pages 982 * |
| 陈勇等: "基于物理模型与边界约束的低照度图像增强算法", 电子与信息学报, no. 12, 15 December 2017 (2017-12-15), pages 2962 - 2969 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112446839B (en) | Image enhancement method, image enhancement device, electronic equipment and computer readable storage medium | |
| CN110033026B (en) | Target detection method, device and equipment for continuous small sample images | |
| WO2022247005A1 (en) | Method and apparatus for identifying target object in image, electronic device and storage medium | |
| CN110717953B (en) | Coloring method and system for black-and-white pictures based on CNN-LSTM (computer-aided three-dimensional network-link) combination model | |
| CN113705462B (en) | Face recognition method, device, electronic equipment and computer readable storage medium | |
| CN111355941A (en) | Image color real-time correction method, device and system | |
| CN112308802A (en) | Image analysis method and system based on big data | |
| CN104202448A (en) | System and method for solving shooting brightness unevenness of mobile terminal camera | |
| CN114757854B (en) | Night vision image quality improving method, device and equipment based on multispectral analysis | |
| CN112364916A (en) | Image classification method based on transfer learning, related equipment and storage medium | |
| CN114723044B (en) | Error compensation method, device, chip and equipment for in-memory computing chip | |
| CN113012068B (en) | Image denoising method, image denoising device, electronic equipment and computer-readable storage medium | |
| Zhan et al. | Computational colour constancy based on convolutional neural networks with a cross‐level architecture | |
| CN113792672A (en) | Public place health code acquisition method, device, equipment and medium | |
| CN112686365B (en) | Method, device and computer equipment for operating neural network model | |
| Zhou et al. | A two‐stage hue‐preserving and saturation improvement color image enhancement algorithm without gamut problem | |
| CN116109501A (en) | Low-illumination image sequence enhancement method, device, electronic equipment and storage medium | |
| CN111179222A (en) | Intelligent detection method and device for cerebral hemorrhage point and computer readable storage medium | |
| CN113139650A (en) | Tuning method and computing device of deep learning model | |
| CN115908175A (en) | Low-illumination image multi-level enhancement method and device, electronic equipment and storage medium | |
| LU500193B1 (en) | Low-illumination image enhancement method and system based on multi-expression fusion | |
| CN113191376A (en) | Image processing method, image processing device, electronic equipment and readable storage medium | |
| CN111899239A (en) | Image processing method and device | |
| CN117953811B (en) | LED huge curtain point-by-point correction system, method, equipment and medium | |
| CN113159081B (en) | Image processing method and related 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 |