CN103559715B - A kind of method for detecting abnormality of high spectrum image and device - Google Patents

Abstract

This application provides a kind of method for detecting abnormality and device of high spectrum image, wherein, method comprises: obtain high spectrum image to be detected, RX algorithm is adopted to carry out abnormality detection to high spectrum image to be detected, obtain abnormality detection result, Threshold segmentation is carried out to abnormality detection result, obtain goal pels collection and backdrop pels collection, calculate mean vector and the covariance matrix of goal pels collection, and the mean vector of backdrop pels collection and covariance matrix, for the pixel each to be detected in high spectrum image, the distance of pixel to be detected and backdrop pels is calculated by the mean vector of backdrop pels collection and covariance matrix, the distance of pixel to be detected and goal pels is calculated by the mean vector of goal pels collection and covariance matrix, the distance of pixel to be detected and goal pels is deducted by the distance of pixel to be detected and backdrop pels, obtain abnormality detection result.The method that the application provides and device, can improve abnormal object recall rate, reduce false alarm rate.

Description

A kind of method for detecting abnormality of high spectrum image and device

Technical field

The present invention relates to abnormality detection technical field, particularly relate to a kind of method for detecting abnormality and device of high spectrum image.

Background technology

Abnormality detection is a hot issue in EO-1 hyperion field.Abnormal object refers to the interested atural object of people.It has two major features, one, has obvious SPECTRAL DIVERSITY, its two, the probability occurred in the picture is lower.

In prior art, usually use RX algorithm realization abnormality detection.RX algorithm hypothesis image background obeys multivariate normal distribution, describes background information by the mean vector and covariance matrix calculating backdrop pels collection.Inventor finds in the process realizing the invention: RX algorithm cannot accurate estimated background information, causes testing result to have higher false alarm rate.

Summary of the invention

In view of this, the invention provides a kind of method for detecting abnormality and device of high spectrum image, cannot accurate estimated background information in order to solve in prior art RX algorithm, cause testing result to have the problem of higher false alarm rate, its technical scheme is as follows:

On the one hand, a kind of method for detecting abnormality of high spectrum image, comprising:

Obtain high spectrum image to be detected;

Adopt RX algorithm to carry out abnormality detection to described high spectrum image to be detected, obtain abnormality detection result;

Threshold segmentation is carried out to described abnormality detection result, obtains goal pels collection and backdrop pels collection;

Calculate mean vector and the covariance matrix of described goal pels collection, and the mean vector of described backdrop pels collection and covariance matrix;

For the pixel each to be detected in described high spectrum image, by the mean vector of described backdrop pels collection and the distance of the covariance matrix described pixel to be detected of calculating and backdrop pels, by the mean vector of described goal pels collection and the distance of the covariance matrix described pixel to be detected of calculating and goal pels, deduct the distance of described pixel to be detected and goal pels by the distance of described pixel to be detected and backdrop pels, obtain abnormality detection result.

Wherein, use RX algorithm to carry out abnormality detection to described high spectrum image to be detected, obtain abnormality detection result matrix, comprising:

Utilize mean vector and the covariance matrix of described high spectrum image background to be detected, by A (i, j)=[r (i, j)-μ] ^tΣ ^-1[r (i, j)-μ] calculate described abnormality detection result matrix, wherein, A (i, j) is described abnormality detection result matrix, r (i, j) in described high spectrum image to be detected at position (i, j) pixel, μ is the mean vector of background in described high spectrum image to be detected, and Σ is the covariance matrix of background in described high spectrum image to be detected.

Wherein, Threshold segmentation is carried out to described abnormality detection result, obtains goal pels collection and backdrop pels collection, comprising:

Obtain confidence factor γ _lowerand γ _upper, and the grey level histogram of described abnormality detection result;

Determined and described confidence factor γ by the grey level histogram of described abnormality detection result _lowercorresponding gray threshold TH _lower, and with described confidence factor γ _uppercorresponding gray threshold TH _upper;

From interval [TH _lower, TH _upper] in determine optimum gradation threshold value TH _best;

Described optimum gradation threshold value TH will be greater than in described abnormality detection result _bestpixel be defined as goal pels, obtain described goal pels collection, the pixel except the goal pels determined be defined as backdrop pels, obtain described backdrop pels collection.

Wherein, described from interval [TH _lower, TH _upper] in determine optimum gradation threshold value TH _best, comprising:

By described interval [TH _lower, TH _upper] be divided into multiple isometric interval;

Determine the quantity of the abnormality detection result value falling into each interval in described abnormality detection result;

By the quantity falling into the abnormality detection result value in each interval determined, determine described optimum gradation threshold value TH by preset rules _best;

Wherein, described preset rules is: t _ifor falling into the quantity of i-th interval abnormality detection result value, T _i+1for falling into the quantity of the i-th+1 interval abnormality detection result value.

Wherein, calculate by the mean vector of described backdrop pels collection and covariance matrix the distance that described pixel to be detected is backdrop pels, comprising:

Utilize mean vector and the covariance matrix of described backdrop pels collection, pass through calculate the distance of described pixel to be detected and backdrop pels, wherein, μ ₀for the mean vector of described backdrop pels collection, Σ ₀for the covariance matrix of described backdrop pels collection, D ₀(i, j) is described pixel r (i, j) to be detected and the distance of backdrop pels;

Calculating described pixel to be detected by the mean vector of described goal pels collection and covariance matrix is that the distance of goal pels comprises:

Utilize mean vector and the covariance matrix of described goal pels collection, pass through calculate the distance of described pixel to be detected and goal pels, wherein, r (i, j) is pixel to be detected, μ ₁for the mean vector of described goal pels collection, Σ ₁for the covariance matrix of described goal pels collection, D ₁(i, j) is described pixel r (i, j) to be detected and the distance of goal pels.

An abnormal detector for high spectrum image, comprising:

Image collection module, for obtaining high spectrum image to be detected;

First detection module, for adopting RX algorithm to carry out abnormality detection to described high spectrum image to be detected, obtains abnormality detection result;

Threshold segmentation module, for carrying out Threshold segmentation to described abnormality detection result, obtains goal pels collection and backdrop pels collection;

Computing module, for calculating mean vector and the covariance matrix of described goal pels collection, and the mean vector of described backdrop pels collection and covariance matrix;

Second detection module, for for the pixel each to be detected in described high spectrum image, by the mean vector of described backdrop pels collection and the distance of the covariance matrix described pixel to be detected of calculating and backdrop pels, by the mean vector of described goal pels collection and the distance of the covariance matrix described pixel to be detected of calculating and goal pels, deduct the distance of described pixel to be detected and goal pels by the distance of described pixel to be detected and backdrop pels, obtain abnormality detection result.

Wherein, described first detection module comprises:

First calculating sub module, for utilizing mean vector and the covariance matrix of described high spectrum image background to be detected, by A (i, j)=[r (i, j)-μ] ^tΣ ^-1[r (i, j)-μ] calculate described abnormality detection result matrix, wherein, A (i, j) is described abnormality detection result matrix, r (i, j) in described high spectrum image to be detected at position (i, j) pixel, μ is the mean vector of background in described high spectrum image to be detected, and Σ is the covariance matrix of background in described high spectrum image to be detected.

Wherein, described Threshold segmentation module comprises:

Obtain submodule, for obtaining confidence factor γ _lowerand γ _upper, and, the grey level histogram of described abnormality detection result;

First determines submodule, for being determined and described confidence factor γ by the grey level histogram of described abnormality detection result _lowercorresponding gray threshold TH _lower, and with described confidence factor γ _uppercorresponding gray threshold TH _upper;

Second determines submodule, for from interval [TH _lower, TH _upper] in determine optimum gradation threshold value TH _best;

3rd determines submodule, for being greater than described optimum gradation threshold value TH in described abnormality detection result _bestpixel be defined as goal pels, obtain described goal pels collection, the pixel except the goal pels determined be defined as backdrop pels, obtain described backdrop pels collection.

Wherein, described second determines that submodule comprises:

Interval division unit, for by described interval [TH _lower, TH _upper] be divided into multiple isometric interval;

First determining unit, for determining the quantity of the abnormality detection result value falling into each interval in described abnormality detection result;

Second determining unit, for the quantity falling into the abnormality detection result value in each interval by determining, determines described optimum gradation threshold value TH by preset rules _best;

Wherein, described preset rules is: t _ifor falling into the quantity of i-th interval abnormality detection result value, T _i+1for falling into the quantity of the i-th+1 interval abnormality detection result value.

Wherein, described second detection module comprises:

Second calculating sub module, for utilizing mean vector and the covariance matrix of described backdrop pels collection, passes through calculating described pixel to be detected is the distance with backdrop pels, wherein, and μ ₀for the mean vector of described backdrop pels collection, Σ ₀for the covariance matrix of described backdrop pels collection, D ₀(i, j) is described pixel r (i, j) to be detected and the distance of backdrop pels;

3rd calculating sub module, for utilizing mean vector and the covariance matrix of described goal pels collection, passes through calculate the distance of described pixel to be detected and goal pels, wherein, r (i, j) is pixel to be detected, μ ₁for the mean vector of described goal pels collection, Σ ₁for the covariance matrix of described goal pels collection, D ₁(i, j) is described pixel r (i, j) to be detected and the distance of goal pels.

Technique scheme has following beneficial effect:

The method for detecting abnormality of high spectrum image provided by the invention and device, after getting high spectrum image to be detected, first RX algorithm is adopted to carry out abnormality detection to high spectrum image to be detected, then Threshold segmentation is carried out to abnormality detection result, obtain goal pels collection and backdrop pels collection, last for each pixel to be detected, the distance of pixel to be detected and goal pels is determined by goal pels collection and backdrop pels collection, and the distance of pixel to be detected and backdrop pels, the distance of pixel to be detected and backdrop pels is utilized to deduct the distance of pixel to be detected and goal pels, obtain the abnormality detection result value of this pixel to be detected, thus obtain the abnormality detection result of high spectrum image to be detected.The method for detecting abnormality of high spectrum image provided by the invention and device, the background information that contains in high spectrum image to be detected and target information is utilized to instruct follow-up abnormality detection work, the method can estimated background information more accurately, thus make background information meet the hypothesis of RX algorithm to background information, and then improve abnormal object recall rate, reduction false alarm rate.

Accompanying drawing explanation

In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, be briefly described to the accompanying drawing used required in embodiment or description of the prior art below, apparently, accompanying drawing in the following describes is only embodiments of the invention, for those of ordinary skill in the art, under the prerequisite not paying creative work, other accompanying drawing can also be obtained according to the accompanying drawing provided.

The schematic flow sheet of the method for detecting abnormality of a kind of high spectrum image that Fig. 1 provides for the embodiment of the present invention;

The schematic flow sheet of the method for detecting abnormality of the another kind of high spectrum image that Fig. 2 provides for the embodiment of the present invention;

The structural representation of the abnormal detector of a kind of high spectrum image that Fig. 3 provides for the embodiment of the present invention;

The structural representation of the abnormal detector of the another kind of high spectrum image that Fig. 4 provides for the embodiment of the present invention.

Embodiment

Below in conjunction with the accompanying drawing in the embodiment of the present invention, be clearly and completely described the technical scheme in the embodiment of the present invention, obviously, described embodiment is only the present invention's part embodiment, instead of whole embodiments.Based on the embodiment in the present invention, those of ordinary skill in the art, not making the every other embodiment obtained under creative work prerequisite, belong to the scope of protection of the invention.

Refer to Fig. 1, the schematic flow sheet of the method for detecting abnormality of a kind of high spectrum image provided for the embodiment of the present invention, the method can comprise:

Step S101: obtain high spectrum image to be detected.

Wherein, high spectrum image is a three-dimensional array, concrete, and for M is capable, the image of N row, a L wave band, high-spectrum image set image information and spectral information are.High spectrum image is made up of M × N number of pixel, and each pixel r (i, j) is a L dimensional vector, is made up of the numerical value of L the wave band of high spectrum image in (i, j) position.

Step S102: adopt RX algorithm to carry out abnormality detection to high spectrum image to be detected, obtain abnormality detection result.

Step S103: carry out Threshold segmentation to abnormality detection result, obtains goal pels collection and backdrop pels collection.

Carrying out image threshold segmentation is a kind of traditional the most frequently used image partition method, because of its realize simple, calculated amount is little, performance is comparatively stable and to become in Iamge Segmentation the most most widely used cutting techniques of fundamental sum.It is specially adapted to the image that target and background occupies different grey-scale scope.

Due to the blind detection process that abnormality detection of the prior art is a unknown object information and background information, RX algorithm lacks available information and makes accurate estimation to background information, and inaccurate background estimating causes testing result to have higher false alarm rate.In view of this, the embodiment of the present invention obtains the background information and target information that contain in high spectrum image to be detected by Threshold segmentation, the background information obtained and target information are used for the abnormality detection of aid in later, accurate estimation can be made to background by background information and target information, thus improve recall rate, the reduction false alarm rate of follow-up abnormality detection.

Step S104: the mean vector and the covariance matrix that calculate goal pels collection, and, the mean vector of backdrop pels collection and covariance matrix.

Step S105: for the pixel each to be detected in high spectrum image, the distance of pixel to be detected and backdrop pels is calculated by the mean vector of backdrop pels collection and covariance matrix, the distance of pixel to be detected and goal pels is calculated by the mean vector of goal pels collection and covariance matrix, deduct the distance of pixel to be detected and goal pels by the distance of pixel to be detected and backdrop pels, obtain abnormality detection result.

The method for detecting abnormality of the high spectrum image that the embodiment of the present invention provides, after getting high spectrum image to be detected, first RX algorithm is adopted to carry out abnormality detection to high spectrum image to be detected, then Threshold segmentation is carried out to abnormality detection result, obtain goal pels collection and backdrop pels collection, last for each pixel to be detected, the distance of pixel to be detected and goal pels is determined by goal pels collection and backdrop pels collection, and the distance of pixel to be detected and backdrop pels, the distance of pixel to be detected and goal pels is deducted by the distance of pixel to be detected and backdrop pels, obtain the abnormality detection result value of this pixel to be detected, thus obtain the abnormality detection result of high spectrum image to be detected.The method for detecting abnormality of the high spectrum image that the embodiment of the present invention provides, the background information that contains in high spectrum image to be detected and target information is utilized to instruct follow-up abnormality detection work, the method can estimated background information more accurately, thus make background information meet the hypothesis of RX algorithm to background information, and then improve abnormal object recall rate, reduction false alarm rate.

Refer to Fig. 2, the schematic flow sheet of the method for detecting abnormality of the another kind of high spectrum image provided for the embodiment of the present invention, the method can comprise:

Step S201: obtain high spectrum image R to be detected.

Wherein, high spectrum image R is a three-dimensional array, concrete, and for M is capable, the image of N row, a L wave band, high-spectrum image set image information and spectral information are.High spectrum image is made up of M × N number of pixel, and each pixel r (i, j) is a L dimensional vector, is made up of the numerical value of L the wave band of high spectrum image in (i, j) position.

Step S202: the mean vector μ calculating background in high spectrum image R to be detected.

Wherein, in high spectrum image R to be detected, the mean vector μ of background calculates by formula (1):

$\mathrm{\μ}=\frac{1}{M\·N}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}r(i,j)---\left(1\right)$

Step S203: utilize the mean vector μ of background to calculate the covariance matrix Σ of background in high spectrum image R to be detected.

Wherein, in high spectrum image R to be detected, the covariance matrix Σ of background calculates by formula (2):

$\mathrm{\Σ}=\frac{1}{M\·N-1}\underset{i=1}{\overset{M}{\mathrm{\Σ}}}\underset{j=1}{\overset{N}{\mathrm{\Σ}}}{[r(i,j)-\mathrm{\μ}][r(i,j)-\mathrm{\μ}]}^T---\left(2\right)$

Step S204: the mean vector μ and the covariance matrix Σ that utilize background in high spectrum image R to be detected, adopts RX algorithm to carry out abnormality detection to high spectrum image R to be detected, obtains abnormality detection result matrix A.

Wherein, abnormality detection result matrix A is that M is capable, the gray level image of N row, 1 wave band.

Adopt RX algorithm to carry out abnormality detection to high spectrum image R to be detected to be specially: calculate testing result matrix A by formula (3):

A(i,j)=[r(i,j)-μ] ^TΣ ^-1[r(i,j)-μ](3)

Step S205: obtain confidence factor γ _lowerand γ _upper, and the grey level histogram of abnormality detection result, determined and confidence factor γ by the grey level histogram of abnormality detection result _lowercorresponding gray threshold TH _lower, and with confidence factor γ _uppercorresponding gray threshold TH _upper.

Wherein, confidence factor represents the probability that in image, background occurs.Concrete, confidence factor γ _lowerrepresent the minimum probability that in image, background occurs, confidence factor γ _upperrepresent the maximum probability that in image, background occurs.

Step S206: from interval [TH _lower, TH _upper] in determine optimum gradation threshold value TH _best.

In a kind of possible implementation, from interval [TH _lower, TH _upper] in determine optimum gradation threshold value TH _bestcan comprise: by interval [TH _lower, TH _upper] be divided into multiple isometric interval; Determine the quantity of the abnormality detection result value falling into each interval in abnormality detection result; By the quantity falling into the abnormality detection result value in each interval determined, by preset rules determination optimum gradation threshold value TH _best.Wherein, preset rules can be: t _ifor falling into the quantity of i-th interval abnormality detection result value, T _i+1for falling into the quantity of the i-th+1 interval abnormality detection result value, i and for being more than or equal to the integer of 1.

Step S207: optimum gradation threshold value TH will be greater than in abnormality detection result _bestpixel be defined as goal pels, obtain goal pels collection H ₁, the pixel except the goal pels determined is defined as backdrop pels, obtains backdrop pels collection H ₀.

Carrying out image threshold segmentation is a kind of traditional the most frequently used image partition method, because of its realize simple, calculated amount is little, performance is comparatively stable and to become in Iamge Segmentation the most most widely used cutting techniques of fundamental sum.It is specially adapted to the image that target and background occupies different grey-scale scope.

Due to the blind detection process that abnormality detection of the prior art is a unknown object information and background information, RX algorithm lacks available information and makes accurate estimation to background information, and inaccurate background estimating causes testing result to have higher false alarm rate.In view of this, the embodiment of the present invention obtains the background information and target information that contain in high spectrum image to be detected by Threshold segmentation, and the background information of acquisition and target information are used for the abnormality detection of aid in later.On the one hand, background extraction information and target information are as priori, with the pixel estimated background more similar to background, abnormal object and noise can be suppressed, background is made to better meet algorithm hypothesis, on the other hand, the introducing of background information and target information can highlight target, thus improves abnormal object recall rate, reduction false alarm rate.

Step S208: calculate goal pels collection H ₁mean vector μ ₁with covariance matrix Σ ₁, and, backdrop pels collection H ₀mean vector μ ₀with covariance matrix Σ ₀.

Step S209: for the pixel each to be detected in high spectrum image R to be detected, by backdrop pels collection H ₀mean vector μ ₀with covariance matrix Σ ₀calculate the distance of pixel to be detected and backdrop pels, by goal pels collection H ₁mean vector μ ₁with covariance matrix Σ ₁calculate the distance of pixel to be detected and goal pels, deduct the distance of pixel to be detected and goal pels by the distance of pixel to be detected and backdrop pels, obtain abnormality detection result.

For pixel r (i, j) to be detected, by backdrop pels collection H ₀mean vector μ ₀with covariance matrix Σ ₀calculate pixel r (i, j) to be detected to be specially with the distance of backdrop pels: utilize backdrop pels collection H ₀mean vector μ ₀with covariance matrix Σ ₀through type (4) calculates the distance D of pixel r (i, j) to be detected and backdrop pels ₀(i, j):

$D_0(i,j)=\frac{1}{2}(r(i,j)-{\mathrm{\μ}}_0{)}^T{\mathrm{\Σ}}_0^{-1}(r(i,j)-{\mathrm{\μ}}_0)---\left(4\right)$

Same, for pixel r (i, j) to be detected, by goal pels collection H ₁mean vector μ ₁with covariance matrix Σ ₁calculate pixel r (i, j) to be detected to be specially with the distance of goal pels: utilize goal pels collection H ₁mean vector μ ₁with covariance matrix Σ ₁, through type (5) calculates the distance D of pixel r (i, j) to be detected and goal pels ₁(i, j):

$D_1(i,j)=\frac{1}{2}{(r(i,j)-{\mathrm{\μ}}_1)}^T{\mathrm{\Σ}}_1^{-1}(r(i,j)-{\mathrm{\μ}}_1)---\left(5\right)$

Calculating the distance P of pixel r (i, j) to be detected with backdrop pels ₀(i, j), and the distance P of pixel r (i, j) to be detected and goal pels ₁after (i, j), through type (6) calculates the abnormality detection result value Δ D of pixel r (i, j) _0-1(i, j):

ΔD _0-1(i,j)=D ₀(i,j)-D ₁(i,j)(6)

Aforesaid operations is performed for each pixel in high spectrum image to be detected, just obtains the abnormality detection result Δ D of high spectrum image to be detected _0-1.

The method for detecting abnormality of the high spectrum image that the embodiment of the present invention provides, after getting high spectrum image to be detected, first RX algorithm is adopted to carry out abnormality detection to high spectrum image to be detected, then Threshold segmentation is carried out to abnormality detection result, obtain goal pels collection and backdrop pels collection, last for each pixel to be detected, the distance of pixel to be detected and goal pels is determined by goal pels collection and backdrop pels collection, and the distance of pixel to be detected and backdrop pels, the distance of pixel to be detected and backdrop pels is utilized to deduct the distance of pixel to be detected and goal pels, obtain the abnormality detection result value of this pixel to be detected, thus obtain the abnormality detection result of high spectrum image to be detected.The method for detecting abnormality of the high spectrum image that the embodiment of the present invention provides, the background information that contains in high spectrum image to be detected and target information is utilized to instruct follow-up abnormality detection work, the method can estimated background information more accurately, thus make background information meet the hypothesis of RX algorithm to background information, and then improve abnormal object recall rate, reduction false alarm rate.

Refer to Fig. 3, the structural representation of the abnormal detector of a kind of high spectrum image provided for the embodiment of the present invention, this device can comprise: image collection module 301, first detection module 302, Threshold segmentation module 303, computing module 304, second detection module 305.Wherein:

Image collection module 301, for obtaining high spectrum image to be detected.

First detection module 302, for adopting RX algorithm to carry out abnormality detection to high spectrum image to be detected, obtains abnormality detection result.

Threshold segmentation module 303, for carrying out Threshold segmentation to abnormality detection result, obtains goal pels collection and backdrop pels collection.

Computing module 304, for calculating mean vector and the covariance matrix of goal pels collection, and the mean vector of backdrop pels collection and covariance matrix.

Second detection module 305, for for the pixel each to be detected in high spectrum image to be detected, the distance of pixel to be detected and backdrop pels is calculated by the mean vector of backdrop pels collection and covariance matrix, the distance of pixel to be detected and goal pels is calculated by the mean vector of goal pels collection and covariance matrix, deduct the distance of pixel to be detected and goal pels by the distance of pixel to be detected and backdrop pels, obtain abnormality detection result.

The abnormal detector of the high spectrum image that the embodiment of the present invention provides, after getting high spectrum image to be detected, first RX algorithm is adopted to carry out abnormality detection to high spectrum image to be detected, then Threshold segmentation is carried out to abnormality detection result, obtain goal pels collection and backdrop pels collection, last for each pixel to be detected, the distance of pixel to be detected and goal pels is determined by goal pels collection and backdrop pels collection, and the distance of pixel to be detected and backdrop pels, the distance of pixel to be detected and backdrop pels is utilized to deduct the distance of pixel to be detected and goal pels, obtain the abnormality detection result value of this pixel to be detected, thus obtain the abnormality detection result of high spectrum image to be detected.The abnormal detector of the high spectrum image that the embodiment of the present invention provides, the background information that contains in high spectrum image to be detected and target information is utilized to instruct follow-up abnormality detection work, can estimated background information more accurately, thus make background information meet the hypothesis of RX algorithm to background information, and then improve abnormal object recall rate, reduction false alarm rate.

Refer to Fig. 4, the structural representation of the abnormal detector of the another kind of high spectrum image provided for the embodiment of the present invention, this device can comprise: image collection module 401, first detection module 402, Threshold segmentation module 403, computing module 404, second detection module 405.Wherein:

Image collection module 401, for obtaining high spectrum image to be detected.

First detection module 402, for adopting RX algorithm to carry out abnormality detection to high spectrum image to be detected, obtains abnormality detection result.

Further, first detection module comprises 402: the first calculating sub module 4021.

First calculating sub module, for utilizing mean vector and the covariance matrix of high spectrum image background to be detected, by A (i, j)=[r (i, j)-μ] ^tΣ ^-1[r (i, j)-μ] calculate abnormality detection result matrix, wherein, A (i, j) is abnormality detection result matrix, r (i, j) in high spectrum image to be detected at position (i, j) pixel, μ is the mean vector of background in high spectrum image to be detected, and Σ is the covariance matrix of background in high spectrum image to be detected.

Threshold segmentation module 403, for carrying out Threshold segmentation to abnormality detection result, obtains goal pels collection and backdrop pels collection.

Further, Threshold segmentation module 403 can comprise: obtain submodule 4031, first and determine that submodule 4032, second determines that submodule 4033 and the 3rd determines submodule 4034.Wherein:

Obtain submodule 4031, for obtaining confidence factor γ _lowerand γ _upper, and the grey level histogram of abnormality detection result.

First determines submodule 4032, for being determined and confidence factor γ by the grey level histogram of abnormality detection result _lowercorresponding gray threshold TH _lower, and with confidence factor γ _uppercorresponding gray threshold TH _upper.

Second determines submodule 4033, for from interval [TH _lower, TH _upper] in determine optimum gradation threshold value TH _best.

3rd determines submodule 4034, for being greater than optimum gradation threshold value TH in abnormality detection result _bestpixel be defined as goal pels, obtain goal pels collection, the pixel except the goal pels determined be defined as backdrop pels, obtain backdrop pels collection.

Further, second determines that submodule 4033 can comprise: interval division unit, the first determining unit and the second determining unit.Wherein, interval division unit, for by described interval [TH _lower, TH _upper] be divided into multiple isometric interval; First determining unit, for determining the quantity of the abnormality detection result value falling into each interval in described abnormality detection result; Second determining unit, for the quantity falling into the abnormality detection result value in each interval by determining, determines described optimum gradation threshold value TH by preset rules _best, preset rules can be: t _ifor falling into the quantity of i-th interval abnormality detection result value, T _i+1for falling into the quantity of the i-th+1 interval abnormality detection result value.

Computing module 404, for calculating mean vector and the covariance matrix of goal pels collection, and the mean vector of backdrop pels collection and covariance matrix.

Second detection module 405, for for the pixel each to be detected in high spectrum image to be detected, the distance of pixel to be detected and backdrop pels is calculated by the mean vector of backdrop pels collection and covariance matrix, the distance of pixel to be detected and goal pels is calculated by the mean vector of goal pels collection and covariance matrix, deduct the distance of pixel to be detected and goal pels by the distance of pixel to be detected and backdrop pels, obtain abnormality detection result.

Further, the second detection module 405 comprises: the second calculating sub module 4051 and the 3rd calculating sub module 4052.Wherein:

Second calculating sub module 4051, for utilizing mean vector and the covariance matrix of backdrop pels collection, passes through calculate the distance of pixel to be detected and backdrop pels, wherein, μ ₀for the mean vector of backdrop pels collection, Σ ₀for the covariance matrix of backdrop pels collection, P ₀(i, j) is pixel r (i, j) to be detected and the distance of backdrop pels.

3rd calculating sub module 4052, for utilizing mean vector and the covariance matrix of backdrop pels collection, passes through calculate the distance of pixel to be detected and goal pels, wherein, r (i, j) is pixel to be detected, μ ₁for the mean vector of goal pels collection, Σ ₁for the covariance matrix of goal pels collection, D ₁(i, j) is pixel r (i, j) to be detected and the distance of goal pels.

The abnormal detector of the high spectrum image that the embodiment of the present invention provides, after getting high spectrum image to be detected, first RX algorithm is adopted to carry out abnormality detection to high spectrum image to be detected, then Threshold segmentation is carried out to abnormality detection result, obtain goal pels collection and backdrop pels collection, last for each pixel to be detected, the distance of pixel to be detected and goal pels is determined by goal pels collection and backdrop pels collection, and the distance of pixel to be detected and backdrop pels, the distance of pixel to be detected and backdrop pels is utilized to deduct the distance of pixel to be detected and goal pels, obtain the abnormality detection result value of this pixel to be detected, thus obtain the abnormality detection result of high spectrum image to be detected.The abnormal detector of the high spectrum image that the embodiment of the present invention provides, the background information that contains in high spectrum image to be detected and target information is utilized to instruct follow-up abnormality detection work, can estimated background information more accurately, thus make background information meet the hypothesis of RX algorithm to background information, and then improve abnormal object recall rate, reduction false alarm rate.

It should be noted that, each embodiment in this instructions all adopts the mode of going forward one by one to describe, and what each embodiment stressed is the difference with other embodiments, between each embodiment identical similar part mutually see.For device or system class embodiment, due to itself and embodiment of the method basic simlarity, so description is fairly simple, relevant part illustrates see the part of embodiment of the method.

Also it should be noted that, in this article, the such as relational terms of first and second grades and so on is only used for an entity or operation to separate with another entity or operational zone, and not necessarily requires or imply the relation that there is any this reality between these entities or operation or sequentially.And, term " comprises ", " comprising " or its any other variant are intended to contain comprising of nonexcludability, thus make to comprise the process of a series of key element, method, article or equipment and not only comprise those key elements, but also comprise other key elements clearly do not listed, or also comprise by the intrinsic key element of this process, method, article or equipment.When not more restrictions, the key element limited by statement " comprising ... ", and be not precluded within process, method, article or the equipment comprising described key element and also there is other identical element.

The software module that the method described in conjunction with embodiment disclosed herein or the step of algorithm can directly use hardware, processor to perform, or the combination of the two is implemented.Software module can be placed in the storage medium of other form any known in random access memory (RAM), internal memory, ROM (read-only memory) (ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technical field.

To the above-mentioned explanation of the disclosed embodiments, professional and technical personnel in the field are realized or uses the present invention.To be apparent for those skilled in the art to the multiple amendment of these embodiments, General Principle as defined herein can without departing from the spirit or scope of the present invention, realize in other embodiments.Therefore, the present invention can not be restricted to these embodiments shown in this article, but will meet the widest scope consistent with principle disclosed herein and features of novelty.

Claims (8)

1. a method for detecting abnormality for high spectrum image, is characterized in that, comprising:

Obtain high spectrum image to be detected;

Adopt RX algorithm to carry out abnormality detection to described high spectrum image to be detected, obtain abnormality detection result;

Threshold segmentation is carried out to described abnormality detection result, obtains goal pels collection and backdrop pels collection;

Calculate mean vector and the covariance matrix of described goal pels collection, and the mean vector of described backdrop pels collection and covariance matrix;

For the pixel each to be detected in described high spectrum image, by the mean vector of described backdrop pels collection and the distance of the covariance matrix described pixel to be detected of calculating and backdrop pels, by the mean vector of described goal pels collection and the distance of the covariance matrix described pixel to be detected of calculating and goal pels, by the distance deducting described pixel to be detected and goal pels of the distance of described pixel to be detected and backdrop pels, obtain abnormality detection result;

Wherein, Threshold segmentation is carried out to described abnormality detection result, obtains goal pels collection and backdrop pels collection, comprising:

Obtain confidence factor γ _lowerand γ _upper, and the grey level histogram of described abnormality detection result;

Determined and described confidence factor γ by the grey level histogram of described abnormality detection result _lowercorresponding gray threshold TH _lower, and with described confidence factor γ _uppercorresponding gray threshold TH _upper;

From interval [TH _lower, TH _upper] in determine optimum gradation threshold value TH _best;

Described optimum gradation threshold value TH will be greater than in described abnormality detection result _bestpixel be defined as goal pels, obtain described goal pels collection, the pixel except the goal pels determined be defined as backdrop pels, obtain described backdrop pels collection.

2. method according to claim 1, is characterized in that, uses RX algorithm to carry out abnormality detection to described high spectrum image to be detected, obtains abnormality detection result matrix, comprising:

Utilize mean vector and the covariance matrix of described high spectrum image background to be detected, by A (i, j)=[r (i, j)-μ] ^tΣ ^-1[r (i, j)-μ] calculate described abnormality detection result matrix, wherein, A (i, j) is described abnormality detection result matrix, r (i, j) in described high spectrum image to be detected at position (i, j) pixel, μ is the mean vector of background in described high spectrum image to be detected, and Σ is the covariance matrix of background in described high spectrum image to be detected.

3. method according to claim 1, is characterized in that, described from interval [TH _lower, TH _upper] in determine optimum gradation threshold value TH _best, comprising:

By described interval [TH _lower, TH _upper] be divided into multiple isometric interval;

Determine the quantity of the abnormality detection result value falling into each interval in described abnormality detection result;

By the quantity falling into the abnormality detection result value in each interval determined, determine described optimum gradation threshold value TH by preset rules _best;

Wherein, described preset rules is: ${\mathrm{TH}}_{Best}={\mathrm{TH}}_{\hat{i}+1},\hat{i}={\mathrm{argmin}}_i\left\{\frac{T_{i+1}}{T_i}\right\},$ T _ifor falling into the quantity of i-th interval abnormality detection result value, T _i+1for falling into the quantity of the i-th+1 interval abnormality detection result value.

4. according to the method in claim 1-3 described in any one, it is characterized in that, by the mean vector of described backdrop pels collection and the distance of the covariance matrix described pixel to be detected of calculating and backdrop pels, comprising:

Utilize mean vector and the covariance matrix of described backdrop pels collection, pass through calculate the distance of described pixel to be detected and backdrop pels, wherein, μ ₀for the mean vector of described backdrop pels collection, Σ ₀for the covariance matrix of described backdrop pels collection, D ₀(i, j) is described pixel r (i, j) to be detected and the distance of backdrop pels;

Comprised by the mean vector of described goal pels collection and the distance of the covariance matrix described pixel to be detected of calculating and goal pels:

Utilize mean vector and the covariance matrix of described goal pels collection, pass through calculate the distance of described pixel to be detected and goal pels, wherein, r (i, j) is pixel to be detected, μ ₁for the mean vector of described goal pels collection, Σ ₁for the covariance matrix of described goal pels collection, D ₁(i, j) is described pixel r (i, j) to be detected and the distance of goal pels.

5. an abnormal detector for high spectrum image, is characterized in that, comprising:

Image collection module, for obtaining high spectrum image to be detected;

First detection module, for adopting RX algorithm to carry out abnormality detection to described high spectrum image to be detected, obtains abnormality detection result;

Threshold segmentation module, for carrying out Threshold segmentation to described abnormality detection result, obtains goal pels collection and backdrop pels collection;

Computing module, for calculating mean vector and the covariance matrix of described goal pels collection, and the mean vector of described backdrop pels collection and covariance matrix;

Second detection module, for for the pixel each to be detected in described high spectrum image, by the mean vector of described backdrop pels collection and the distance of the covariance matrix described pixel to be detected of calculating and backdrop pels, by the mean vector of described goal pels collection and the distance of the covariance matrix described pixel to be detected of calculating and goal pels, deduct the distance of described pixel to be detected and goal pels by the distance of described pixel to be detected and backdrop pels, obtain abnormality detection result;

Described Threshold segmentation module comprises:

Obtain submodule, for obtaining confidence factor γ _lowerand γ _upper, and, the grey level histogram of described abnormality detection result;

First determines submodule, for being determined and described confidence factor γ by the grey level histogram of described abnormality detection result _lowercorresponding gray threshold TH _lower, and with described confidence factor γ _uppercorresponding gray threshold TH _upper;

Second determines submodule, for from interval [TH _lower, TH _upper] in determine optimum gradation threshold value TH _best;

3rd determines submodule, for being greater than described optimum gradation threshold value TH in described abnormality detection result _bestpixel be defined as goal pels, obtain described goal pels collection, the pixel except the goal pels determined be defined as backdrop pels, obtain described backdrop pels collection.

6. device according to claim 5, is characterized in that, described first detection module comprises:

First calculating sub module, for utilizing mean vector and the covariance matrix of described high spectrum image background to be detected, by A (i, j)=[r (i, j)-μ] ^tΣ ^-1[r (i, j)-μ] calculate described abnormality detection result matrix, wherein, A (i, j) is described abnormality detection result matrix, r (i, j) in described high spectrum image to be detected at position (i, j) pixel, μ is the mean vector of background in described high spectrum image to be detected, and Σ is the covariance matrix of background in described high spectrum image to be detected.

7. device according to claim 5, is characterized in that, described second determines that submodule comprises:

Interval division unit, for by described interval [TH _lower, TH _upper] be divided into multiple isometric interval;

First determining unit, for determining the quantity of the abnormality detection result value falling into each interval in described abnormality detection result;

Second determining unit, for the quantity falling into the abnormality detection result value in each interval by determining, determines described optimum gradation threshold value TH by preset rules _best;

Wherein, described preset rules is: ${\mathrm{TH}}_{Best}={\mathrm{TH}}_{\hat{i}+1},\hat{i}={\mathrm{argmin}}_i\left\{\frac{T_{i+1}}{T_i}\right\},$ T _ifor falling into the quantity of i-th interval abnormality detection result value, T _i+1for falling into the quantity of the i-th+1 interval abnormality detection result value.

8. according to the device in claim 5-7 described in any one, it is characterized in that, described second detection module comprises:

Second calculating sub module, for utilizing mean vector and the covariance matrix of described backdrop pels collection, passes through calculate the distance of described pixel to be detected and backdrop pels, wherein, μ ₀for the mean vector of described backdrop pels collection, Σ ₀for the covariance matrix of described backdrop pels collection, D ₀(i, j) is described pixel r (i, j) to be detected and the distance of backdrop pels;

3rd calculating sub module, for utilizing mean vector and the covariance matrix of described goal pels collection, passes through calculate the distance of described pixel to be detected and goal pels, wherein, r (i, j) is pixel to be detected, μ ₁for the mean vector of described goal pels collection, Σ ₁for the covariance matrix of described goal pels collection, D ₁(i, j) is described pixel r (i, j) to be detected and the distance of goal pels.