CN103971334A

CN103971334A - Hyperspectral remote sensing image correcting method and device

Info

Publication number: CN103971334A
Application number: CN201410171313.3A
Authority: CN
Inventors: 吴远峰; 张�浩; 陈正超; 孙旭; 高连如; 张兵
Original assignee: Institute of Remote Sensing and Digital Earth of CAS
Current assignee: Institute of Remote Sensing and Digital Earth of CAS
Priority date: 2014-04-25
Filing date: 2014-04-25
Publication date: 2014-08-06
Anticipated expiration: 2034-04-25

Abstract

The embodiment of the invention discloses a hyperspectral remote sensing image correcting method and device. The method includes the steps that one waveband in a hyperspectral remote sensing image to be corrected is determined as a reference waveband, the relative offset of each row is determined through the minimum error of the adjacent rows, and the absolute offset (namely the offset of each row relative to the pixel of a target row) of each row is determined through the relative offset of each row, so that corresponding rows in the hyperspectral remote sensing image to be corrected are moved waveband by waveband according to the absolute offset of each row to obtain the corrected hyperspectral remote sensing image. According to the hyperspectral remote sensing image correcting method and device, starting from own features of the hyperspectral remote sensing image, related data between the rows in the hyperspectral remote sensing image are counted, the image is corrected through minimum row related error and row-by-row position translation methods, and the method and device correct geometric distortion of the hyperspectral remote sensing image without depending on external posture data.

Description

High-spectrum remote sensing bearing calibration and device

Technical field

The present invention relates to technical field of image processing, more particularly, relate to a kind of high-spectrum remote sensing bearing calibration and device.

Background technology

Along with developing rapidly of science and technology, remote sensing technology more and more receives people's concern, according to remote-sensing flatform, classify, substantially can be divided into spacer remote sensing, airborne remote sensing and ground remote sensing, wherein, airborne remote sensing general reference is carried the Remote Sensing Technical System of imaging sensor earth observation by aerial platforms such as aircraft (including man-machine and unmanned plane), dirigible, balloons.

But, the impact of the easy wind-engaging of aerial platform, unstable air-flow and himself mechanical part vibrations etc. and the phenomenon of side-sway, rolling or vibrations occurs, and imaging sensor generally adopts push-scanning image mode to obtain the multispectral data (i.e. imaging line by line) of atural object in high spectrum resolution remote sensing technique, the unstable of aerial platform will greatly be lowered into image quality, make the high-spectrum remote sensing that obtains occur the geometric distortion phenomenon of " burr " or distortion, as shown in Figure 1.Therefore, be necessary occurring that the high-spectrum remote sensing of geometric distortion proofreaies and correct.

At present, conventional is a kind of to occurring that the method that the high-spectrum remote sensing of geometric distortion is proofreaied and correct is: by aerial platform, carry attitude recording unit, utilize the data of these attitude recording unit records to carry out proofreading and correct by pixel to the high-spectrum remote sensing obtaining.But the general small volume of unmanned plane, consideration for weight and cost factor, unmanned plane is difficult to carry the attitude recording unit of high precision (its volume and weight of more accurate equipment is larger), therefore, when aerial platform is unmanned plane, by carrying the mode of attitude recording unit, be difficult to proofread and correct there is the high-spectrum remote sensing of geometric distortion.

Therefore, how the geometric distortion of the high-spectrum remote sensing obtaining by unmanned plane mode is proofreaied and correct and become problem demanding prompt solution.

Summary of the invention

The object of this invention is to provide a kind of high-spectrum remote sensing bearing calibration and device, so that the geometric distortion of the high-spectrum remote sensing obtaining by unmanned plane mode is proofreaied and correct.

For achieving the above object, the invention provides following technical scheme:

A high-spectrum remote sensing bearing calibration, comprising:

Determine with reference to wave band, described is to meet pre-conditioned wave band in all wave bands of high-spectrum remote sensing to be corrected with reference to wave band;

According to default movement rule, the described capable pixel of i+1 with reference to wave band is moved by pixel with respect to the described capable pixel of i with reference to wave band, location of pixels of every movement, the error amount that calculates the described capable pixel of i+1 and the described capable pixel of i with reference to wave band with reference to wave band according to the first formula, described the first formula is:

E＝[∑(R _p-C _q) ²]/m ²

Wherein, E is the error amount of the described capable pixel of i+1 and the described capable pixel of i with reference to wave band with reference to wave band, R _pbe the value of p pixel in the capable pixel of i, C _qbe the value of q pixel in the capable pixel of i+1, in the capable pixel of described i+1, q pixel alignd with p pixel column in the capable pixel of described i; M is the number of the pixel of column alignment with the capable pixel of i in the capable pixel of i+1;

By error amount hour, the capable pixel of described i+1 is defined as the relative displacement of the capable pixel of described i+1 with respect to the side-play amount of the capable pixel of described i, wherein, the capable pixel of described i+1 is positive side-play amount or negative side-play amount with respect to the side-play amount of the capable pixel of described i, the just skew orientation determination that the capable pixel of described i+1 is default with respect to the positive and negative foundation of the value of the side-play amount of the capable pixel of described i;

The first absolute offset values of calculating described each row pixel with reference to wave band, comprising: calculate the relative displacement sum of all row between the capable pixel of i and predetermined target line pixel, obtain first and value; The first absolute offset values of the capable pixel of described i is described first and the relative displacement sum of value and the capable pixel of described i;

According to the first absolute offset values of described each row pixel with reference to wave band, by wave band, move row corresponding in high-spectrum remote sensing to be corrected, obtain the high-spectrum remote sensing after the first correction; Wherein, the pixel count that the capable pixel of i moves is the absolute value of the first absolute offset values of the described capable pixel of i with reference to wave band, and the direction that the capable pixel of described each wave band i moves is identical with the first corresponding offset direction of absolute offset values of the described capable pixel of i with reference to wave band; Wherein, i is more than or equal to 1 positive integer.

Said method, preferred, describedly according to default movement rule, the described capable pixel of i+1 with reference to wave band is moved and comprised by pixel with respect to the described capable pixel of i with reference to wave band:

Determine initial column position and position, end column that the capable pixel of described i+1 moves with respect to the capable pixel of described i;

The pixel of the column position maximum of the capable pixel of described i+1 is alignd with the pixel column at the initial column position place of the capable pixel of described i;

The capable pixel of described i+1 is moved by pixel to described end column locality, until the pixel of the column position minimum of the capable pixel of described i+1 is alignd with the pixel column of the position, end column of the capable pixel of described i.

Said method, preferred, described initial column position and the position, end column of determining that the capable pixel of described i+1 moves with respect to the capable pixel of described i comprises:

The pixel column position of determining the column position minimum of the capable pixel of described i is the initial column position that the capable pixel of described i+1 moves with respect to the capable pixel of described i;

The pixel column position of determining the column position maximum of the capable pixel of described i is the position, end column that the capable pixel of described i+1 moves with respect to the capable pixel of described i.

Determine that the first row position that user inputs is the initial column position that the capable pixel of described i+1 moves with respect to the capable pixel of described i;

Determine that the secondary series position that user inputs is the position, end column that the capable pixel of described i+1 moves with respect to the capable pixel of described i;

Wherein, described first row position is different from the pixel column position of the column position minimum of the capable pixel of described i, described secondary series position is different from the pixel column position of the column position maximum of the capable pixel of described i, and described first row position is different from described secondary series position.

Degree of overlapping according to predetermined adjacent lines is determined initial column position and the position, end column that the capable pixel of described i+1 moves with respect to the capable pixel of described i, wherein,

The initial column position that the capable pixel of described i+1 moves with respect to the capable pixel of described i is:

The position, end column that the capable pixel of described i+1 moves with respect to the capable pixel of described i is:

Wherein, n is the number of pixels of the capable pixel of described i; R is the degree of overlapping of described predetermined adjacent lines.

Said method, preferred, after the high-spectrum remote sensing after obtaining the first correction, also comprise:

According to first user, operate in the distortion contour curve of render target atural object in the reference wave band of the described first high-spectrum remote sensing after proofreading and correct;

According to the second user, operate in the correct contour curve of drawing described Target scalar in the reference wave band of the described first high-spectrum remote sensing after proofreading and correct;

Determine described the 3rd column position with reference to the capable pixel of i of wave band and the intersection point of described distortion contour curve, and determine described the 4th column position with reference to the capable pixel of i of wave band and the intersection point of described correct contour curve;

Obtain the second absolute offset values of the described capable pixel of i with reference to wave band, the second absolute offset values of the described capable pixel of i with reference to wave band is the difference of described the 3rd column position and described the 4th column position;

According to the second absolute offset values of the capable pixel of described i, the capable pixel of i of the high-spectrum remote sensing after proofreading and correct described first by wave band moves to the direction of described correct contour curve, obtains the high-spectrum remote sensing after the second correction; The capable pixel of described i moves to the direction of described correct contour curve the absolute value that amount of movement is described the second absolute offset values.

A high-spectrum remote sensing means for correcting, comprising:

The first determination module, for determining with reference to wave band, described is to meet pre-conditioned wave band in all wave bands of high-spectrum remote sensing to be corrected with reference to wave band;

The first computing module, for the described capable pixel of i+1 with reference to wave band being moved by pixel with respect to the described capable pixel of i with reference to wave band according to default movement rule, location of pixels of every movement, the error amount that calculates the described capable pixel of i+1 and the described capable pixel of i with reference to wave band with reference to wave band according to the first formula, described the first formula is:

E＝[∑(R _p-C _q) ²]/m ²

The second determination module, for by error amount hour, the capable pixel of described i+1 is defined as the relative displacement of the capable pixel of described i+1 with respect to the side-play amount of the capable pixel of described i, wherein, the capable pixel of described i+1 is positive side-play amount or negative side-play amount with respect to the side-play amount of the capable pixel of described i, the just skew orientation determination that the capable pixel of described i+1 is default with respect to the positive and negative foundation of the value of the side-play amount of the capable pixel of described i;

The second computing module, for calculating the first absolute offset values of described each row pixel with reference to wave band, comprising: calculate the relative displacement sum of all row between the capable pixel of i and predetermined target line pixel, obtain first and value; The first absolute offset values of the capable pixel of described i is described first and the relative displacement sum of value and the capable pixel of described i;

The first correction module, for according to the first absolute offset values of described each row pixel with reference to wave band, moves row corresponding in high-spectrum remote sensing to be corrected by wave band, obtains the high-spectrum remote sensing after the first correction; Wherein, the pixel count that the capable pixel of i moves is the absolute value of the first absolute offset values of the described capable pixel of i with reference to wave band, and the direction that the capable pixel of described each wave band i moves is identical with the first corresponding offset direction of absolute offset values of the described capable pixel of i with reference to wave band; Wherein, i is more than or equal to 1 positive integer.

Said apparatus, preferred, described the first computing module comprises: mover module and calculating sub module; Wherein,

Described mover module is for moving with respect to the described capable pixel of i with reference to wave band the described capable pixel of i+1 with reference to wave band according to default movement rule by pixel; Comprise:

The first determining unit, for initial column position and the position, end column of determining that the capable pixel of described i+1 moves with respect to the capable pixel of described i;

Mobile unit, for the pixel of the column position maximum of the capable pixel of described i+1 is alignd with the pixel column at the initial column position place of the capable pixel of described i, the capable pixel of described i+1 is moved by pixel to described end column locality, until the pixel of the column position minimum of the capable pixel of described i+1 is alignd with the pixel column of the position, end column of the capable pixel of described i;

Described calculating sub module, at described location of pixels of the every movement of the capable pixel of i+1 with reference to wave band, is calculated the error amount of the described capable pixel of i+1 and the described capable pixel of i with reference to wave band with reference to wave band according to the first formula, described the first formula is:

E＝[∑(R _p-C _q) ²]/m ²

Wherein, E is the error amount of the described capable pixel of i+1 and the described capable pixel of i with reference to wave band with reference to wave band, R _pbe the value of p pixel in the capable pixel of i, C _qbe the value of q pixel in the capable pixel of i+1, in the capable pixel of described i+1, q pixel alignd with p pixel column in the capable pixel of described i; M is the number of the pixel of column alignment with the capable pixel of i in the capable pixel of i+1.

Said apparatus, preferred, described the first determining unit comprises:

First determines subelement, for determining that the pixel column position of the column position minimum of the capable pixel of described i is the initial column position that the capable pixel of described i+1 moves with respect to the capable pixel of described i;

Second determines subelement, for determining that the pixel column position of the column position maximum of the capable pixel of described i is the position, end column that the capable pixel of described i+1 moves with respect to the capable pixel of described i.

Said apparatus, preferred, described the first determining unit comprises:

The 3rd determines subelement, for determining that the first row position of user's input is the initial column position that the capable pixel of described i+1 moves with respect to the capable pixel of described i;

The 4th determines subelement, for determining that the secondary series position of user's input is the position, end column that the capable pixel of described i+1 moves with respect to the capable pixel of described i;

Said apparatus, preferred, described the first determining unit comprises:

The 5th determines subelement, for the degree of overlapping according to predetermined adjacent lines, determines initial column position and the position, end column that the capable pixel of described i+1 moves with respect to the capable pixel of described i, wherein,

Said apparatus, preferred, also comprise:

The first drafting module, for operating in the distortion contour curve of the reference wave band render target atural object of the high-spectrum remote sensing after described the first correction according to first user;

The second drafting module, draws the correct contour curve of described Target scalar for operate in the reference wave band of the described first high-spectrum remote sensing after proofreading and correct according to the second user;

The 3rd determination module, for determining described the 3rd column position with reference to the capable pixel of i of wave band and the intersection point of described distortion contour curve, and determines described the 4th column position with reference to the capable pixel of i of wave band and the intersection point of described correct contour curve;

Acquisition module, for obtaining the second absolute offset values of the described capable pixel of i with reference to wave band, the second absolute offset values of the described capable pixel of i with reference to wave band is the difference of described the 3rd column position and described the 4th column position;

The second correction module, for the second absolute offset values according to the capable pixel of described i, the capable pixel of i of the high-spectrum remote sensing after proofreading and correct described first by wave band moves to the direction of described correct contour curve, obtains the high-spectrum remote sensing after the second correction; The capable pixel of described i moves to the direction of described correct contour curve the absolute value that amount of movement is described the second absolute offset values.

Known by above scheme, a kind of high-spectrum remote sensing bearing calibration and device that the application provides, from high-spectrum remote sensing to be corrected, determine that a wave band is with reference to wave band, by the least error of adjacent lines, determine the relative displacement of each row, and by the relative displacement of each row, determine the absolute offset values (being that each row is with respect to the side-play amount of target line pixel) of each row, thereby by wave band, move row corresponding in high-spectrum remote sensing to be corrected, the high-spectrum remote sensing after being proofreaied and correct according to the absolute offset values of each row.Hence one can see that, high-spectrum remote sensing bearing calibration and device that the embodiment of the present application provides, from the feature of high-spectrum remote sensing own, the row of statistics high-spectrum remote sensing inside and the related data between row, by line correlation error minimum and line by line the method for position translation proofread and correct, do not rely on external attitude data, realized the correction for high-spectrum remote sensing geometric distortion.

Accompanying drawing explanation

In order to be illustrated more clearly in the embodiment of the present invention or technical scheme of the prior art, to the accompanying drawing of required use in embodiment or description of the Prior Art be briefly described below, apparently, accompanying drawing in the following describes is only some embodiments of the present invention, for those of ordinary skills, do not paying under the prerequisite of creative work, can also obtain according to these accompanying drawings other accompanying drawing.

For there is the exemplary plot of the unmanned plane high-spectrum remote sensing of geometric distortion in Fig. 1;

A kind of realization flow figure of the high-spectrum remote sensing bearing calibration that Fig. 2 provides for the embodiment of the present application;

The exemplary plot of the capable pixel of i+1 that Fig. 3 a provides for the embodiment of the present application and a kind of relative position of the capable pixel of i;

The exemplary plot of the capable pixel of i+1 that Fig. 3 b provides for the embodiment of the present application and the another kind of relative position of the capable pixel of i;

The schematic diagram that the capable pixel of i+1 that Fig. 4 provides for the embodiment of the present application moves by pixel with respect to the capable pixel of i;

The another kind of realization flow figure of the high-spectrum remote sensing bearing calibration that Fig. 5 provides for the embodiment of the present application;

A kind of structural representation of the high-spectrum remote sensing means for correcting that Fig. 6 provides for the embodiment of the present application;

A kind of structural representation of the first computing module that Fig. 7 provides for the embodiment of the present application;

A kind of structural representation of the first determining unit that Fig. 8 provides for the embodiment of the present application;

The another kind of structural representation of the first determining unit that Fig. 9 provides for the embodiment of the present application;

Another structural representation of the first determining unit that Figure 10 provides for the embodiment of the present application;

The another kind of structural representation of the high-spectrum remote sensing means for correcting that Figure 11 provides for the embodiment of the present application;

The first design sketch after proofreading and correct that passes through to obtain after the embodiment of the present application is proofreaied and correct high-spectrum remote sensing shown in Fig. 1 that Figure 12 provides for the embodiment of the present application;

The distortion contour curve that the user that Figure 13 provides for the embodiment of the present application sketches in high-spectrum remote sensing shown in Figure 12 and correct contour curve;

Design sketch after the second correction that the high-spectrum remote sensing to shown in Figure 12 that Figure 14 provides for the embodiment of the present application obtains after proofreading and correct.

Term " first " in instructions and claims and above-mentioned accompanying drawing, " second ", " the 3rd " " 4th " etc. (if existence) are for distinguishing similar part, and needn't be for describing specific order or precedence.The data that should be appreciated that such use suitably can exchanged in situation, so that the application's described herein embodiment can be with the order enforcement except here illustrated.

Embodiment

Below in conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly and completely described, obviously, described embodiment is only the present invention's part embodiment, rather than whole embodiment.Embodiment based in the present invention, those of ordinary skills, not making the every other embodiment obtaining under creative work prerequisite, belong to the scope of protection of the invention.

Refer to Fig. 2, a kind of realization flow figure of the high-spectrum remote sensing bearing calibration that Fig. 2 provides for the embodiment of the present application, can comprise:

Step S21: determine with reference to wave band, described is to meet pre-conditioned wave band in all wave bands of high-spectrum remote sensing to be corrected with reference to wave band;

High-spectrum remote sensing comprises tens conventionally, the data of up to a hundred wave bands.In the embodiment of the present application, select a wave band as with reference to wave band from all wave bands of high-spectrum remote sensing, wherein, described can be the wave band that signal to noise ratio (S/N ratio) meets default signal to noise ratio (S/N ratio) condition with reference to wave band;

Preferably, the image of each wave band can be shown, by artificial visual, determined with reference to wave band, fewer by noise of artificial selection, and more clearly wave band as with reference to wave band, according to experience, corresponding high-spectrum remote sensing, before several wave bands and the signal to noise ratio (S/N ratio) of several wave bands is generally poor below, so some wave bands are as with reference to wave band in the middle of conventionally can selecting; Also can be by calculative determination with reference to wave band, concrete, can calculate by wave band the signal to noise ratio (S/N ratio) of each wave band of high-spectrum remote sensing to be corrected, the wave band of selecting signal to noise ratio (S/N ratio) maximum is as with reference to wave band, when the wave band of signal to noise ratio (S/N ratio) maximum exists when a plurality of, can from the wave band of signal to noise ratio (S/N ratio) maximum, select at random a wave band as with reference to wave band.

Step S22: the described capable pixel of i+1 with reference to wave band is moved by pixel with respect to the described capable pixel of i with reference to wave band according to default movement rule, location of pixels of every movement, the error amount that calculates the described capable pixel of i+1 and the described capable pixel of i with reference to wave band with reference to wave band according to the first formula, described the first formula is:

E＝[∑(R _p-C _q) ²]/m ²

When calculating, the value of pixel can be original gray-scale value, can be also the spoke brightness value obtaining after radiant correction, can be also the reflectance value obtaining through atmospheric correction.

In the embodiment of the present application, the described capable pixel of i+1 with reference to wave band is moved and refers to that the position of the capable pixel of i remains unchanged by pixel with respect to the described capable pixel of i with reference to wave band, the capable pixel of mobile i+1, the capable pixel of i+1 is when moving, and mobile step-length is once a pixel.

It should be noted that, when the described capable pixel of i+2 with reference to wave band is moved by pixel with respect to the described capable pixel of i+1 with reference to wave band, the position of the capable pixel of i+1 be the capable pixel of described i+1 described with reference to the original position in wave band, rather than the position of the capable pixel of i+1 after moving.

For the better computing method of the error amount of the explanation capable pixel of i+1 and the capable pixel of i, please refer to Fig. 3 a, the exemplary plot of the capable pixel of i+1 that Fig. 3 a provides for the embodiment of the present application and a kind of relative position of the capable pixel of i;

In Fig. 3 a, in the capable pixel of i+1, have 3 pixels and the capable pixel of i is column alignment, wherein, R1 column alignment in C6 and the capable pixel of i in the capable pixel of i+1, R2 column alignment in C7 and the capable pixel of i in the capable pixel of i+1, R3 column alignment in C8 and the capable pixel of i in the capable pixel of i+1; , the capable pixel of i+1 shown in Fig. 3 a and the error value E of the capable pixel of i _example1computing method be:

E _example1＝[(R1-C6) ²+(R2-C7) ²+(R3-C8) ²]/3 ²。

On the basis of Fig. 3 a, when the capable pixel of i+1 moves right after a location of pixels, the exemplary plot of the another kind of relative position of the capable pixel of i+1 and the capable pixel of i as shown in Figure 3 b;

In Fig. 3 b, in the capable pixel of i+1, have 4 pixels and the capable pixel of i is column alignment, wherein, R1 column alignment in C5 and the capable pixel of i in the capable pixel of i+1, R2 column alignment in C6 and the capable pixel of i in the capable pixel of i+1, R3 column alignment in C7 and the capable pixel of i in the capable pixel of i+1, R4 column alignment in C8 and the capable pixel of i in the capable pixel of i+1; The capable pixel of i+1 shown in Fig. 3 b and the error value E of the capable pixel of i _example2computing method be:

E _example2＝[(R1-C5) ²+(R2-C6) ²+(R3-C7) ²+(R4-C8) ²]/4 ²。

Step S23: by error amount hour, the capable pixel of described i+1 is defined as the relative displacement of the capable pixel of described i+1 with respect to the side-play amount of the capable pixel of described i, wherein, the capable pixel of described i+1 is positive side-play amount or negative side-play amount with respect to the side-play amount of the capable pixel of described i, the just skew orientation determination that the capable pixel of described i+1 is default with respect to the positive and negative foundation of the value of the side-play amount of the capable pixel of described i;

Due to the capable pixel of every movement one order i+1, calculate the error amount of an order capable pixel of i+1 and the capable pixel of i, so, the error amount of the capable pixel of i+1 and the capable pixel of i has a plurality of, in the embodiment of the present application, from the capable pixel of i+1 of acquisition and a plurality of error amounts of the capable pixel of i, determine the error amount of value minimum, and the capable pixel of i+1 corresponding to the error amount of this value minimum is defined as to the relative displacement of the capable pixel of described i+1 with respect to the side-play amount of the capable pixel of described i.

In the embodiment of the present application, the capable pixel of i+1 comprises that with respect to the side-play amount of the capable pixel of described i the L row of the capable pixel of i+1 depart from pixel count and the offset direction of the L row of the capable pixel of described i.Therefore, in the embodiment of the present application, the capable pixel of described i+1 is positive side-play amount or negative side-play amount with respect to the side-play amount of the capable pixel of described i, the just skew orientation determination that the capable pixel of described i+1 is default with respect to the positive and negative foundation of the value of the side-play amount of the capable pixel of described i; Concrete, when the L of the capable pixel of i+1 is listed as the Wei Zheng offset direction, offset direction of the L row that depart from the capable pixel of described i, the capable pixel of i+1 is positive side-play amount with respect to the side-play amount of the capable pixel of described i, otherwise the capable pixel of i+1 is negative side-play amount with respect to the side-play amount of the capable pixel of described i.

Concrete, when the L of the capable pixel of i+1 is listed as the first side of the L row that are positioned at the capable pixel of described i, can be defined as positive offset direction; When the L of the capable pixel of i+1 is listed as the second side of the L row that are positioned at the capable pixel of described i, be defined as negative offset direction.For example, as shown in Figure 3 a, can define the Shi Weizheng offset direction, right side that is positioned at the 1st row (being the row at R1 place) of the capable pixel of described i when the 1st row (being the row at C1 place) of the capable pixel of i+1, so, now, the capable pixel of i+1 is-5 with respect to the relative displacement of the capable pixel of described i.

It should be noted that, in the embodiment of the present application, the capable pixel of i is 0 with respect to the relative displacement of the capable pixel of i.

Step S24: calculate the first absolute offset values of described each row pixel with reference to wave band, comprising: calculate the relative displacement sum of all row between the capable pixel of i and predetermined target line pixel, obtain first and value; The first absolute offset values of the capable pixel of described i is described first and the relative displacement sum of value and the capable pixel of described i;

Described predetermined target line pixel can be for described with reference to any one-row pixels in wave band; In the embodiment of the present application, the first absolute drift quality entity of the capable pixel of i is that the capable pixel of i is with respect to the relative displacement of described predetermined target line pixel.

In the embodiment of the present application, the absolute offset values of predetermined target line pixel is 0.

For example, suppose that predetermined target line pixel is for the 6th row pixel with reference to wave band, so, the absolute offset values of the 6th row pixel is 0; The absolute offset values of the 5th row pixel is the relative displacement of the 5th row pixel, the absolute offset values of the 4th row pixel is the relative displacement of the 4th row pixel and the relative displacement sum of the 5th row pixel, and the absolute offset values of the 3rd row pixel is that the relative displacement of the 3rd row pixel is, the relative displacement sum of the relative displacement of the 4th row pixel and the 5th row pixel; In like manner, the absolute offset values of the 7th row pixel is the relative displacement of the 7th row pixel, the absolute offset values of eighth row pixel is the relative displacement of eighth row pixel and the relative displacement sum of the 7th row pixel, and the absolute offset values of the 9th row pixel is that the relative displacement of the 9th row pixel is, the relative displacement sum of the relative displacement of eighth row pixel and the 7th row pixel.

Preferably, for ease of calculating, described target line pixel is the described the first row pixel with reference to wave band, the relative displacement sum of all row before the relative displacement that the first absolute offset values of the capable pixel of i is the capable pixel of i and the capable pixel of i.

Step S25: according to the first absolute offset values of described each row pixel with reference to wave band, move row corresponding in high-spectrum remote sensing to be corrected by wave band, obtain the high-spectrum remote sensing after the first correction; Wherein, the pixel count that the capable pixel of i moves is the absolute value of the first absolute offset values of the described capable pixel of i with reference to wave band, and the direction that the capable pixel of described each wave band i moves is identical with the first corresponding offset direction of absolute offset values of the described capable pixel of i with reference to wave band; Wherein, i is more than or equal to 1 positive integer, i.e. i=1, and 2,3 ..., M-1; M is the line number of the pixel of high-spectrum remote sensing.

After the capable pixel of mobile i, can be 0 by the pixel assignment that removes position of the capable pixel of i.

In the embodiment of the present application, in each wave band of high-spectrum remote sensing to be corrected, the move mode of the capable pixel of i is all identical with the move mode of the capable pixel of i with reference in wave band.For example, the first absolute offset values of supposing the capable pixel of i is-3, the move mode with reference to the capable pixel of i of wave band is: to predefined negative offset direction, move 3 location of pixels, in like manner, in other wave band in high-spectrum remote sensing to be corrected, the capable pixel of i all moves 3 location of pixels to described predefined negative offset direction.

That is to say, in the embodiment of the present application, first determine the absolute offset values with reference to each row pixel of wave band, then according to the side-play amount of each row pixel with reference to wave band, the corresponding row of each wave band of high-spectrum remote sensing to be corrected is moved, thereby realized the correction to high-spectrum remote sensing geometric distortion.

A kind of high-spectrum remote sensing bearing calibration that the embodiment of the present application provides, from high-spectrum remote sensing to be corrected, determine that a wave band is with reference to wave band, by the least error of adjacent lines, determine the relative displacement of each row, and by the relative displacement of each row, determine the absolute offset values (being that each row is with respect to the side-play amount of target line pixel) of each row, thereby by wave band, move row corresponding in high-spectrum remote sensing to be corrected, the high-spectrum remote sensing after being proofreaied and correct according to the absolute offset values of each row.Hence one can see that, the high-spectrum remote sensing bearing calibration that the embodiment of the present application provides, from the feature of high-spectrum remote sensing own, the row of statistics high-spectrum remote sensing inside and the related data between row, by line correlation error minimum and line by line the method for position translation proofread and correct, do not rely on external attitude data, realized the correction for high-spectrum remote sensing geometric distortion.

In embodiment illustrated in fig. 2, preferred, describedly according to default movement rule, the described capable pixel of i+1 with reference to wave band is moved and can be comprised by pixel with respect to the described capable pixel of i with reference to wave band:

Wherein, the pixel column position that can determine the column position minimum of the capable pixel of described i is the initial column position that the capable pixel of described i+1 moves with respect to the capable pixel of described i;

That is to say, the column position that can determine first pixel place of the capable pixel of i is the initial column position that the capable pixel of i+1 moves with respect to the capable pixel of described i; The column position of determining last pixel place of the capable pixel of i is the position, end column that the capable pixel of i+1 moves with respect to the capable pixel of described i.

In the embodiment of the present application, after the pixel of the column position maximum of the capable pixel of described i+1 is alignd with the pixel column at the initial column position place of the capable pixel of described i, by pixel, move the capable pixel of i+1.

In like manner, also the pixel of the column position minimum of the capable pixel of described i+1 can be alignd with the pixel column of the position, end column of the capable pixel of described i;

The capable pixel of described i+1 is moved by pixel to described initial column position direction, until the pixel column at the pixel of the column position maximum of the capable pixel of described i+1 and the initial column position place of the capable pixel of described i aligns.

In above-described embodiment, preferred, in order to reduce calculated amount, the initial column position that described definite capable pixel of described i+1 moves with respect to the capable pixel of described i and the another kind of implementation of position, end column can be:

Determine that the secondary series position that user inputs is the position, end column that the capable pixel of described i+1 moves with respect to the capable pixel of described i.

Wherein, described first row position and secondary series position are rule of thumb determined by expert of the art, described first row position is different from the pixel column position of the column position minimum of the capable pixel of described i, described secondary series position is different from the pixel column position of the column position maximum of the capable pixel of described i, and described first row position is different from described secondary series position.

In above-described embodiment, preferred, in order to reduce calculated amount, the initial column position that described definite capable pixel of described i+1 moves with respect to the capable pixel of described i and another implementation of position, end column are:

Wherein, n is the number of pixels of the capable pixel of i; R is the degree of overlapping of described predetermined adjacent lines, 0<r<1.

represent downward rounding operation;

The degree of overlapping of adjacent lines is that high-spectrum remote sensing does not occur in the situation of geometric distortion, can atural object between adjacent lines continuously corresponding number of pixels divided by the width (being the number of pixel in one-row pixels) of image.In the embodiment of the present application, the degree of overlapping of described predetermined adjacent lines by can atural object between the adjacent lines of pre-estimating continuously corresponding number of pixels divided by the width of high-spectrum remote sensing, obtain.

Illustrate the specific implementation process that the capable pixel of i+1 is moved by pixel with respect to the capable pixel of i below, please refer to Fig. 4, the schematic diagram that the capable pixel of i+1 that Fig. 4 provides for the embodiment of the present application moves by pixel with respect to the capable pixel of i;

The number of pixels of supposing every one-row pixels is 8, and the degree of overlapping of adjacent lines is 0.4, so, can determine that initial column position is position, end column is 8-3+1=6, therefore, embodiment illustrated in fig. 4 in, initial column position is the column position at R3 place in the capable pixel of i, position, end column is the column position at R6 place in the capable pixel of i;

When the capable pixel of mobile i+1, the reference position of the capable pixel of i+1 is: the C8 pixel in the capable pixel of i+1 and the R3 column alignment in the capable pixel of i, then, by pixel, the column position to R6 place in the capable pixel of i moves the capable pixel of i+1, until R6 column alignment in the R1 in the capable pixel of i+1 and the capable pixel of i, the afterwards no longer mobile capable pixel of i+1.

As seen from Figure 4, the number of times that the capable pixel of i+1 moves is 11 times, and the number of times of the corresponding error of calculation is also 11 times;

And if using R1 column position in the capable pixel of i as initial column position, using R8 column position as position, end column, the number of times that the capable pixel of i+1 need to move is 8*2=16 time, the number of times that corresponding error is calculated is also 16 times.

It should be noted that, in Fig. 4 example shown, i+1 is capable only has a line, and this example is for after illustrating and moving, and the position relationship between the capable pixel of i+1 and the capable pixel of i, shows the capable pixel of a plurality of i+1.

In order further to optimize above-described embodiment, the another kind of realization flow figure of the high-spectrum remote sensing bearing calibration that the embodiment of the present application provides as shown in Figure 5, after the high-spectrum remote sensing after obtaining the first correction, can also comprise:

Step S51: the distortion contour curve that operates in render target atural object in the reference wave band of the described first high-spectrum remote sensing after proofreading and correct according to first user;

Step S52: operate in the correct contour curve of drawing described Target scalar in the reference wave band of the described first high-spectrum remote sensing after proofreading and correct according to the second user;

In the embodiment of the present application, by user, in reference to wave band, sketched out the atural object profile after proofreading and correct, and correct atural object profile.Wherein, correct atural object profile can be checked and know according to high-resolution satellite image comparison by domain expert, or can be determined by the photo gathering on the spot.And when actual atural object profile more complicated, can lay in advance straight line indicia band, can in remote sensing images, delineate like this profile of generation distortion of linear mark band and correct profile.

Step S53: determine described the 3rd column position with reference to the capable pixel of i of wave band and the intersection point of described distortion contour curve, and determine described the 4th column position with reference to the capable pixel of i of wave band and the intersection point of described correct contour curve;

Step S54: obtain the second absolute offset values of the described capable pixel of i with reference to wave band, the second absolute offset values of the described capable pixel of i with reference to wave band is the difference of described the 3rd column position and described the 4th column position;

The absolute value of the second absolute offset values of the capable pixel of i is that the pixel that i is capable departs from the pixel count of tram.

Step S55: according to the second absolute offset values of the capable pixel of described i, the capable pixel of i of the high-spectrum remote sensing after proofreading and correct described first by wave band moves to the direction of described correct contour curve, obtains the high-spectrum remote sensing after the second correction; The capable pixel of described i moves to the direction of described correct contour curve the absolute value that amount of movement is described the second absolute offset values.

Corresponding with embodiment of the method, the embodiment of the present application also provides a kind of high-spectrum remote sensing means for correcting, and a kind of structural representation of the high-spectrum remote sensing means for correcting that the embodiment of the present application provides as shown in Figure 6, can comprise:

The first determination module 61, the first computing module 62, the second determination module 63, the second computing modules 64 and the first correction modules 65; Wherein,

The first determination module 61 is for determining with reference to wave band, and described is to meet pre-conditioned wave band in all wave bands of high-spectrum remote sensing to be corrected with reference to wave band;

The first computing module 62 moves with respect to the described capable pixel of i with reference to wave band the described capable pixel of i+1 with reference to wave band for the movement rule according to default by pixel, location of pixels of every movement, the error amount that calculates the described capable pixel of i+1 and the described capable pixel of i with reference to wave band with reference to wave band according to the first formula, described the first formula is:

E＝[∑(R _p-C _q) ²]/m ²

The second determination module 63 is for by error amount hour, the capable pixel of described i+1 is defined as the relative displacement of the capable pixel of described i+1 with respect to the side-play amount of the capable pixel of described i, wherein, the capable pixel of described i+1 is positive side-play amount or negative side-play amount with respect to the side-play amount of the capable pixel of described i, the just skew orientation determination that the capable pixel of described i+1 is default with respect to the positive and negative foundation of the value of the side-play amount of the capable pixel of described i;

The second computing module 64, for calculating the first absolute offset values of described each row pixel with reference to wave band, comprising: calculate the relative displacement sum of all row between the capable pixel of i and predetermined target line pixel, obtain first and value; The first absolute offset values of the capable pixel of described i is described first and the relative displacement sum of value and the capable pixel of described i;

The first correction module 65, for according to the first absolute offset values of described each row pixel with reference to wave band, moves row corresponding in high-spectrum remote sensing to be corrected by wave band, obtains the high-spectrum remote sensing after the first correction; Wherein, the pixel count that the capable pixel of i moves is the absolute value of the first absolute offset values of the described capable pixel of i with reference to wave band, and the direction that the capable pixel of described each wave band i moves is identical with the first corresponding offset direction of absolute offset values of the described capable pixel of i with reference to wave band; Wherein, i is more than or equal to 1 positive integer.

A kind of high-spectrum remote sensing means for correcting that the embodiment of the present application provides, from high-spectrum remote sensing to be corrected, determine that a wave band is with reference to wave band, by the least error of adjacent lines, determine the relative displacement of each row, and by the relative displacement of each row, determine the absolute offset values (being that each row is with respect to the side-play amount of the first row) of each row, thereby by wave band, move row corresponding in high-spectrum remote sensing to be corrected, the high-spectrum remote sensing after being proofreaied and correct according to the absolute offset values of each row.Hence one can see that, the high-spectrum remote sensing means for correcting that the embodiment of the present application provides, from the feature of high-spectrum remote sensing own, the row of statistics high-spectrum remote sensing inside and the related data between row, by line correlation error minimum and line by line the method for position translation proofread and correct, do not rely on external attitude data, realized the correction for high-spectrum remote sensing geometric distortion.

In above-described embodiment, preferred, a kind of structural representation of described the first computing module 62 as shown in Figure 7, can comprise:

Mover module 71 and calculating sub module 72; Wherein,

Described mover module 71 moves with respect to the described capable pixel of i with reference to wave band the described capable pixel of i+1 with reference to wave band for the movement rule according to default by pixel; Specifically can comprise:

The first determining unit 711, for initial column position and the position, end column of determining that the capable pixel of described i+1 moves with respect to the capable pixel of described i;

Mobile unit 712, for the pixel of the column position maximum of the capable pixel of described i+1 is alignd with the pixel column at the initial column position place of the capable pixel of described i, the capable pixel of described i+1 is moved by pixel to described end column locality, until the pixel of the column position minimum of the capable pixel of described i+1 is alignd with the pixel column of the position, end column of the capable pixel of described i;

Described calculating sub module 72, at described location of pixels of the every movement of the capable pixel of i+1 with reference to wave band, is calculated the error amount of the described capable pixel of i+1 and the described capable pixel of i with reference to wave band with reference to wave band according to the first formula, described the first formula is:

E＝[∑(R _p-C _q) ²]/m ²

Above-described embodiment, preferred, a kind of structural representation of the first determining unit 711 that the embodiment of the present application provides as shown in Figure 8, can comprise:

First determines subelement 81, for determining that the pixel column position of the column position minimum of the capable pixel of described i is the initial column position that the capable pixel of described i+1 moves with respect to the capable pixel of described i;

Second determines subelement 82, for determining that the pixel column position of the column position maximum of the capable pixel of described i is the position, end column that the capable pixel of described i+1 moves with respect to the capable pixel of described i.

Above-described embodiment, preferred, the another kind of structural representation of the first determining unit 711 that the embodiment of the present application provides as shown in Figure 9, also can comprise:

The 3rd determines subelement 91, for determining that the first row position of user's input is the initial column position that the capable pixel of described i+1 moves with respect to the capable pixel of described i;

The 4th determines subelement 92, for determining that the secondary series position of user's input is the position, end column that the capable pixel of described i+1 moves with respect to the capable pixel of described i, described first row position is different from the pixel column position of the column position minimum of the capable pixel of described i, described secondary series position is different from the pixel column position of the column position maximum of the capable pixel of described i, and described first row position is different from described secondary series position.

Above-described embodiment, preferred, another structural representation of the first determining unit 711 that the embodiment of the present application provides as shown in figure 10, can comprise:

The 5th determines subelement 101, for the degree of overlapping according to predetermined adjacent lines, determines initial column position and the position, end column that the capable pixel of described i+1 moves with respect to the capable pixel of described i, wherein,

Wherein, n is the number of pixels of the capable pixel of i; R is the degree of overlapping of described adjacent lines, 0<r<1.

Above-described embodiment, preferred, the another kind of structural representation of the high-spectrum remote sensing means for correcting that the embodiment of the present application provides is logical as shown in figure 11, can also comprise:

The first drafting module 111, the second drafting module 112, the three determination modules 113, acquisition module 114 and the second correction module 115; Wherein,

The first drafting module 111 is for operating in the distortion contour curve of the reference wave band render target atural object of the high-spectrum remote sensing after described the first correction according to first user;

The second drafting module 112 is drawn the correct contour curve of described Target scalar for operate in the reference wave band of the high-spectrum remote sensing after described the first correction according to the second user;

The 3rd determination module 113 is for determining described the 3rd column position with reference to the capable pixel of i of wave band and the intersection point of described distortion contour curve, and definite described the 4th column position with reference to the capable pixel of i of wave band and the intersection point of described correct contour curve;

Acquisition module 114 is for obtaining the second absolute offset values of the described capable pixel of i with reference to wave band, and the second absolute offset values of the described capable pixel of i with reference to wave band is the difference of described the 3rd column position and described the 4th column position;

The second correction module 115 is for the second absolute offset values according to the capable pixel of described i, the capable pixel of i of the high-spectrum remote sensing after proofreading and correct described first by wave band moves to the direction of described correct contour curve, obtains the high-spectrum remote sensing after the second correction; The capable pixel of described i moves to the direction of described correct contour curve the absolute value that amount of movement is described the second absolute offset values.

It should be noted that, high-spectrum remote sensing bearing calibration and device that the embodiment of the present application provides, except applying the geometry correction of carrying out high-spectrum remote sensing on unmanned plane, can also be applied to the geometry correction of the upper high-spectrum remote sensing of other aerial platform (if any man-machine, dirigible, balloon etc.).

In addition, the high-spectrum remote sensing after proofreading and correct by the embodiment of the present application can also utilize the data of attitude recording unit record further to proofread and correct, further to improve calibration result.

High-spectrum remote sensing shown in Fig. 1 take below as example, the calibration result of the embodiment of the present application is described, as shown in figure 12, Figure 12 is the design sketch after the first correction obtaining after high-spectrum remote sensing shown in Fig. 1 being proofreaied and correct by the embodiment of the present application.

The distortion contour curve that Figure 13 sketches in high-spectrum remote sensing shown in Figure 12 for user (as curve thicker in figure, i.e. L1) and correct contour curve (as straight line thinner in figure, i.e. L2);

Above-mentioned explanation to the disclosed embodiments, makes professional and technical personnel in the field can realize or use the present invention.To the multiple modification of these embodiment, will be apparent for those skilled in the art, General Principle as defined herein can, in the situation that not departing from the spirit or scope of the present invention, realize in other embodiments.Therefore, the present invention will can not be restricted to these embodiment shown in this article, but will meet the widest scope consistent with principle disclosed herein and features of novelty.

Claims

1. a high-spectrum remote sensing bearing calibration, is characterized in that, comprising:

E＝[∑(R _p-C _q) ²]/m ²

2. method according to claim 1, is characterized in that, describedly according to default movement rule, the described capable pixel of i+1 with reference to wave band is moved and is comprised by pixel with respect to the described capable pixel of i with reference to wave band:

3. method according to claim 2, is characterized in that, described initial column position and the position, end column of determining that the capable pixel of described i+1 moves with respect to the capable pixel of described i comprises:

4. method according to claim 2, is characterized in that, described initial column position and the position, end column of determining that the capable pixel of described i+1 moves with respect to the capable pixel of described i comprises:

5. method according to claim 2, is characterized in that, described initial column position and the position, end column of determining that the capable pixel of described i+1 moves with respect to the capable pixel of described i comprises:

6. according to the method described in claim 1-5 any one, it is characterized in that, after the high-spectrum remote sensing after obtaining the first correction, also comprise:

7. a high-spectrum remote sensing means for correcting, is characterized in that, comprising:

E＝[∑(R _p-C _q) ²]/m ²

8. device according to claim 7, is characterized in that, described the first computing module comprises: mover module and calculating sub module; Wherein,

E＝[∑(R _p-C _q) ²]/m ²

9. device according to claim 8, is characterized in that, described the first determining unit comprises:

10. device according to claim 8, is characterized in that, described the first determining unit comprises:

11. devices according to claim 8, is characterized in that, described the first determining unit comprises:

12. according to the device described in claim 7-11 any one, it is characterized in that, also comprises: