CN102313886B - Coordinate conversion method in pulsed radar imaging - Google Patents

Abstract

The invention discloses a coordinate conversion method in pulsed radar imaging. According to the invention, a novel look-up table is established; a dead point situation is indicated by additional information; when a coordinate is output, additional point processing is carried out on the dead point according to the additional information to improve a radar imaging quality; the established look-up table only records conversion from polar coordinates from 0 degrees to 45 degrees to a rectangular coordinate; in a circumference, conversion from other polar coordinates to the rectangular coordinate are completed by a mapping formula, a table look up method, and a conversion formula, so that a storage space needed by the look-up table is reduced; and a storage space in chip is saved for a processor by utilizing an offchip memory. The method provided in the invention can be applied to a pulsed radar image display based on FPGA hardware platform and pulsed radar image display based on PC; besides, the method provided in the invention can also be applied to various polar coordinate software and hardware display system.

Description

Coordinate transformation method in a kind of pulsed radar imaging

Technical field

The present invention relates to the disposal route that a kind of pulsed radar image shows, especially relate to the coordinate transformation method in a kind of pulsed radar imaging.

Background technology

Pulsed radar is having a wide range of applications in fields such as civil navigation, investigations.Display is as the direct reaction of radar information, and display effect often directly affects radar effect and ship safety.The scan mode of display is lining by line scan or staggered scanning from left to right, the mode that belongs to rectangular coordinate, and being the circular radial scan mode, the information that radar obtains belongs to polar coordinate mode, at first the polar coordinates information of radar to be converted to rectangular coordinate information and deposit video memory in so realize the demonstration of radar data, from video memory, read and show in the mode of frame scan again.

What at present polar coordinates mainly adopted to the conversion of rectangular coordinate in the radar imagery is look-up table, this method is saved in look-up table with polar coordinates by the rectangular coordinate that trigonometric function calculates, then with look-up table stores in storage space, after in processor, inputting the radar scanning line, angle value addressable storage space according to the radar scanning line obtains one group of rectangular coordinate corresponding to all analyzing spots on this radar scanning line, realized the quick conversion of polar coordinates to rectangular coordinate, but on the rectangular coordinate in rectangular coordinate system that this method calculates by trigonometric function the polar coordinates of analyzing spot and the radar display screen rectangular coordinate of pixel in rectangular coordinate system one to one problem do not process, " dead point " can appear when video picture, some pixel did not have pixel to show when " dead point " was the display screen video picture, caused imaging unintelligible; And the look-up table record is rectangular coordinate corresponding to all polar coordinates in the circumference range, and used storage space is larger.

Summary of the invention

The present invention is directed to above technical matters by the coordinate transformation method in the pulsed radar imaging that a kind of imaging clearly is provided, solved well " dead point " problem, only use a small amount of storage space simultaneously.

Technical scheme of the present invention is the coordinate transformation method in a kind of pulsed radar imaging, comprises following treatment step:

S1. the sweep trace in one week of radar scanning adds up to N, and analyzing spot M is arranged on the every radar scanning line, and n is n bar sweep trace, in 0 °～45 ° scopes of radar scanning circumference, the polar coordinates (r, θ) of the analyzing spot of all radar scanning lines is passed through formula $\left\{\begin{array}{c}{x}_i=r\cos \left(\mathrm{\θ}\right)\\ y_i=r\sin \left(\mathrm{\θ}\right)\end{array}\right.$ Calculate corresponding rectangular coordinate (x _i, y _i), form array { (x _i, y _i), wherein, N, M, n, i, r are integer, and n ∈ [0, N/8], i ∈ [0, MN/8), r ∈ [0, M),

S2. will be positioned at 0 °～45 ° scopes of rectangular coordinate system first quartile and satisfy condition $\left\{\begin{array}{c}y\≤x\\ x^2+y^2\≤M^2\end{array}\right.$ All rectangular coordinate points (x, y) find out, form array { (x, y) }, again with each the rectangular coordinate point in the array { (x, y) } respectively with array { (x _i, y _i) in each rectangular coordinate point relatively, if satisfy condition $\left\{\begin{array}{c}y=y_i\\ x=x_i\end{array}\right.,$ Then with the deletion from the array { (x, y) } of this coordinate points (x, y), garbled array { (x, y) } is redefined be array { (x ', y ') } at last;

S3. select array { (x _i, y _i) in i rectangular coordinate point, i=0,1,2 ..., MN/8-1 travels through array { (x ', y ') } again, finds out to satisfy condition $\left\{\begin{array}{c}{y}^{\′}\≤y_i\\ x^{\′}\≤x_i\end{array}\right.$ All rectangular coordinates form new array { (x _Ij', y _Ij') and from array { (x ', y ') } deletion, wherein (x _Ij', y _Ij') expression array { (x _i, y _i) in satisfy condition around i coordinate points $\left\{\begin{array}{c}{y}^{\′}\≤y_i\\ x^{\′}\≤x_i\end{array}\right.$ J coordinate points, j is that [0, J), J represents array { (x for integer and j ∈ _Ij', y _Ij') in contained element number;

With rectangular coordinate (x _i, y _i) horizontal ordinate x _iDeduct respectively array { (x _Ij', y _Ij') in the horizontal ordinate x of each rectangular coordinate _Ij', rectangular coordinate (x _i, y _i) ordinate y _iDeduct respectively array { (x _Ij', y _Ij') in the ordinate y of each rectangular coordinate _Ij' obtain (x _i-x _I1', y _i-y _I1'), (x _i-x _I2', y _i-y _I2') ... (x _i-x _Ij', y _i-y _Ij'), the binary number add that is 2J*k bit with a width again represents (x _i-x _I1', y _i-y _I1'), (x _i-x _I2', y _i-y _I2') ... (x _i-x _Ij', y _i-y _Ij'), method for expressing is: 0～(k-1) bit of binary number add represents numerical value x _i-x _I1', binary number add k～(2k-1) bit represents numerical value y _i-y _I1', 2k～(3k-1) bit represents numerical value x _i-x _I2', binary number add 3k～(4k-1) bit represents numerical value y _i-y _I2' ..., (2 (j-1) * k)～((2j-1) * k-1) bit of binary number add represents numerical value x _i-x _Ij', ((2j-1) * k)～((2j * k)-1) bit of binary number add represents numerical value y _i-y _Ij', wherein k is all x _i-x _Ij' and y _i-y _IjMaximization in the ' numerical value is the bit wide behind the binary number;

With rectangular coordinate (x _i, y _i) and corresponding add value save as (x, y, add) _i, (x, y, add) _iRepresented rectangular coordinate (x _i, y _i) and array { (x _Ij', y _Ij');

S4. repeat s3, i value 0,1,2 successively wherein ..., MN/8-1 is again with all (x, y, add) _iBe saved in look-up table;

S5. processor n in receiving radar scanning circumference range ₁Behind bar sweep trace and the corresponding half-tone information thereof, with n ₁The bar sweep trace is transformed into respective scan line n in 0 °～45 ° scopes by the mapping formula;

Wherein, n ₁[0, N), the mapping formula is ∈ $\left\{\begin{array}{cc}n=n_1,& 0\&le;n_1\&le;N/8\\ n=N/4-n_1,& N/8<n_1\&le;N/4\\ n=n_1-N/4,& N/4<n_1\&le;3N/8\\ n=N/2-n_1,& 3N/8<n_1\&le;N/2\\ n=n_1-N/2,& N/2<n_1\&le;5N/8\\ n=3N/4-n_1,& 5N/8<n_1\&le;3N/4\\ n=n_1-3N/4,& 3N/4<n_1\&le;7N/8\\ n=N-n_1,& 7N/8<n_1<N\end{array}\right.;$

Obtain M corresponding to a sweep trace n number of scan points certificate by the addressing look-up table, with this M number of scan points according to forming array

n ₁∈ [0, N);

S6. the array that s5 is obtained

Be saved in the first memory of processor, with n ₁The gray-scale value that each analyzing spot is corresponding on the bar sweep trace is saved in the second memory of processor;

S7. from first memory, take out n ₁R number of scan points certificate on the bar sweep trace

R ∈ [0, M), and from second memory read with

Corresponding gray-scale value; According to

In rectangular coordinate

And the add value obtains array

With rectangular coordinate

And array

In each coordinate element substitution conversion formula obtain rectangular coordinate With

Wherein, conversion formula is $\left\{\begin{array}{cc}{x}_1=x,y_1=y,& 0\&le;n_1\&le;N/8\\ x_1=y,y_1=x,& N/8<n_1\&le;N/4\\ x_1=-y,y_1=x,& N/4<n_1\&le;3N/8\\ x_1=-x,y_1=y,& 3N/8<n_1\&le;N/2\\ x_1=-x,y_1=-y,& N/2<n_1\&le;5N/8\\ x_1=-y,y_1=-x,& 5N/8<n_1\&le;3N/4\\ x_1=y,y_1=-x,& 3N/4<n_1\&le;7N/8\\ x_1=x,y_1=-y,& 7N/8<n_1<N\end{array}\right.;$

With rectangular coordinate

With

Corresponding gray-scale value outputs to video memory together;

With array In each rectangular coordinate respectively with

Corresponding gray-scale value outputs to video memory together.

Described step s4 saves as readable documents with look-up table and stores on the external memorizer of processor, when processor starts, look-up table is copied to the internal memory from external memorizer; First memory among described step s6, the described step s7 and second memory are the on-chip memory of processor.

Described processor adopting FPGA.

Described external memorizer adopts NANDFLASH, and described internal memory adopts SDRAM, and described on-chip memory adopts the FIFO storer.

Compared with prior art, the invention has the advantages that and set up the novel look-up table that comprises rectangular coordinate and additional information, indicate " dead point " situation by additional information, when rectangular coordinate is exported, by additional information a processing is mended at the dead point, improved the radar imagery quality; The look-up table of setting up has only recorded 0 °～45 ° polar coordinates to the conversion of rectangular coordinate, and other polar coordinates are finished by mapping, look-up table and conversion process to the conversion of rectangular coordinate in the circumference, reduced the storage area of look-up table needs; Saved storage space in the sheet by using chip external memory as processor.

Description of drawings

Fig. 1 is that look-up table of the present invention generates schematic diagram;

Fig. 2 is pulsed radar coordinate conversion realization flow schematic diagram of the present invention;

Fig. 3 is the design sketch that actual measurement radar image of the present invention shows.

Embodiment

Below in conjunction with accompanying drawing, provide specific embodiments of the invention.

Instance parameter is: the processor adopting FPGA of radar imagery, the external memorizer of FPGA adopts NANDFLASH, the internal memory of FPGA adopts SDRAM, the on-chip memory of FPGA adopts the FIFO storer, and one week of radar scanning is that 384 polar coordinates, LCD size are arranged is that 15 cun, radar image viewing area are 768 * 768 for 2048 sweep traces, every sweep trace.Specifically be unfolded as follows.

S1. the polar coordinates (r, θ) with the analyzing spot on all radar scanning lines in 0 °～45 ° scopes pass through formula $\left\{\begin{array}{c}x=r\cos \left(\mathrm{\&theta;}\right)\\ y=r\sin \left(\mathrm{\&theta;}\right)\end{array}\right.$ Calculate corresponding rectangular coordinate (x, y), form array { (x, y) }, n is n bar sweep trace, and n, r are integer and n ∈ [0,256], r ∈ [0,384),

S2. will be positioned at 0 °～45 ° scopes of rectangular coordinate system first quartile and satisfy condition $\left\{\begin{array}{c}y\&le;x\\ x^2+y^2\&le;{383}^2\end{array}\right.$ All rectangular coordinate points (x, y) find out, form array { (x, y) }, again with each the rectangular coordinate point in the array { (x, y) } respectively with array { (x _i, y _i) in each rectangular coordinate point relatively, if satisfy condition $\left\{\begin{array}{c}y=y_i\\ x=x_i\end{array}\right.$ Then with the deletion from the array { (x, y) } of this coordinate points (x, y), garbled array { (x, y) } is redefined be array { (x ', y ') } at last;

S3. select array { (x _i, y _i) in i rectangular coordinate point, i=0,1,2 ..., 98303, travel through again array { (x ', y ') }, find out and satisfy condition $\left\{\begin{array}{c}{y}^{\&prime;}\&le;y_i\\ x^{\&prime;}\&le;x_i\end{array}\right.$ All rectangular coordinates form new array { (x _Ij', y _Ij'), again with array { (x _Ij', y _Ij') deletion from array { (x ', y ') } of the rectangular coordinate point that contains, avoid these coordinate points to be repeated to find out, wherein (x _Ij', y _Ij') expression array { (x _i, y _i) in satisfy condition around i coordinate points $\left\{\begin{array}{c}{y}^{\&prime;}\&le;y_i\\ x^{\&prime;}\&le;x_i\end{array}\right.$ J coordinate points, j is that [0, J), J represents array { (x for integer and j ∈ _Ij', y _Ij') in contained element number;

With rectangular coordinate (x _i, y _i) horizontal ordinate x _iDeduct respectively array { (x _Ij', y _Ij') in the horizontal ordinate x of each rectangular coordinate _Ij', rectangular coordinate (x _i, y _i) ordinate y _iDeduct respectively array { (x _Ij', y _Ij') in the ordinate y of each rectangular coordinate _Ij' obtain (x _i-x _I1', y _i-y _I1'), (x _i-x _I2', y _i-y _I2') ... (x _i-x _Ij', y _i-y _Ij'), the binary number add that is 2J*k bit with a width again represents (x _i-x _I1', y _i-y _I1'), (x _i-x _I2', y _i-y _I2') ... (x _i-x _Ij', y _i-y _Ij'), method for expressing is: 0～(k-1) bit of binary number add represents numerical value x _i-x _I1', binary number add k～(2k-1) bit represents numerical value y _i-y _I1', 2k～(3k-1) bit represents numerical value x _i-x _I2', binary number add 3k～(4k-1) bit represents numerical value y _i-y _I2' ..., (2 (j-1) * k)～((2j-1) * k-1) bit of binary number add represents numerical value x _i-x _Ij', ((2j-1) * k)～((2j*k)-1) bit of binary number add represents numerical value y _i-y _Ij', wherein k is all x _i-x _Ij' and y _i-y _IjMaximization in the ' numerical value is the bit wide behind the binary number;

With rectangular coordinate (x _i, y _i) and corresponding add value save as (x, y, add) _i, (x, y, add) _iRepresented rectangular coordinate (x _i, y _i) and array { (x _Ij', y _Ij');

S4. repeat s3, i value 0,1,2 successively wherein ..., 98303, again with all (x, y, add) _iBe saved in look-up table; Look-up table is sent to processor by serial ports, and processor deposits look-up table in external memorizer; When processor starts, the look-up table in the external memorizer is deposited in the internal memory;

S5. input n ₁During bar radar scanning line data, polar coordinates (r, n that each analyzing spot in the radar scanning line is corresponding ₁) being converted to polar coordinates (r, n) by the mapping formula, polar coordinates (r, n) obtain corresponding data by the addressing look-up table and are designated as

Wherein, n ₁, n is integer and n ₁∈ [0,2047], n ∈ [0,256],

The mapping formula is: $\left\{\begin{array}{cc}n=n_1,& 0\&le;n_1\&le;N/8\\ n=N/4-n_1,& N/8<n_1\&le;N/4\\ n=n_1-N/4,& N/4<n_1\&le;3N/8\\ n=N/2-n_1,& 3N/8<n_1\&le;N/2\\ n=n_1-N/2,& N/2<n_1\&le;5N/8\\ n=3N/4-n_1,& 5N/8<n_1\&le;3N/4\\ n=n_1-3N/4,& 3N/4<n_1\&le;7N/8\\ n=N-n_1,& 7N/8<n_1<N\end{array}\right.;$

S6. will

Deposit in the first memory; Be (r, n with polar coordinates ₁) the half-tone information of analyzing spot deposit in the second memory, first memory and second memory are the on-chip memory of processor in the present embodiment;

S7. from first memory, take out n ₁R number of scan points certificate on the bar sweep trace

R be integer and r ∈ [0,384), and from second memory read with

Corresponding gray-scale value; According to

In rectangular coordinate

And the add value obtains array

With rectangular coordinate And array

In each coordinate element substitution conversion formula obtain rectangular coordinate

With

Conversion formula is $\left\{\begin{array}{cc}{x}_1=x,y_1=y,& 0\&le;n_1\&le;N/8\\ x_1=y,y_1=x,& N/8<n_1\&le;N/4\\ x_1=-y,y_1=x,& N/4<n_1\&le;3N/8\\ x_1=-x,y_1=y,& 3N/8<n_1\&le;N/2\\ x_1=-x,y_1=-y,& N/2<n_1\&le;5N/8\\ x_1=-y,y_1=-x,& 5N/8<n_1\&le;3N/4\\ x_1=y,y_1=-x,& 3N/4<n_1\&le;7N/8\\ x_1=x,y_1=-y,& 7N/8<n_1<N\end{array}\right.;$

With rectangular coordinate

With

Corresponding gray-scale value outputs to video memory together, with array

In each rectangular coordinate respectively with Corresponding gray-scale value outputs to video memory together.

Claims (4)

1. the coordinate transformation method in the pulsed radar imaging is characterized in that comprising following treatment step:

S1. the sweep trace in one week of radar scanning adds up to N, and analyzing spot M is arranged on the every radar scanning line, and n is n bar sweep trace, in 0 °～45 ° scopes of radar scanning circumference, the polar coordinates (r, θ) of the analyzing spot of all radar scanning lines is passed through formula $\left\{\begin{array}{c}{x}_i=r\cos \left(\mathrm{\&theta;}\right)\\ y_i=r\sin \left(\mathrm{\&theta;}\right)\end{array}\right.$ Calculate corresponding rectangular coordinate (x _i, y _i), form array { (x _i, y _i), wherein, N, M, n, i, r are integer, and n ∈ [0, N/8], i ∈ [0, MN/8), r ∈ [0, M),

S2. will be positioned at 0 °～45 ° scopes of rectangular coordinate system first quartile and satisfy condition $\left\{\begin{array}{c}y\&le;x\\ x^2+y^2\&le;M^2\end{array}\right.$ All rectangular coordinate points (x, y) find out, form array { (x, y) }, again with each the rectangular coordinate point in the array { (x, y) } respectively with array { (x _i, y _i) in each rectangular coordinate point relatively, if satisfy condition $\left\{\begin{array}{c}y=y_i\\ x=x_i\end{array}\right.,$ Then with the deletion from the array { (x, y) } of this coordinate points (x, y), garbled array { (x, y) } is redefined be array { (x ', y ') } at last;

S3. select array { (x _i, y _i) in i rectangular coordinate point, i=0,1,2 ..., MN/8-1 travels through array { (x ', y ') } again, finds out to satisfy condition $\left\{\begin{array}{c}{y}^{\&prime;}\&le;y_i\\ x^{\&prime;}\&le;x_i\end{array}\right.$ All rectangular coordinates form new array { (x _Ij', y _Ij') and from array { (x ', y ') } deletion, wherein (x _Ij', y _Ij') expression array { (x _i, y _j) in satisfy condition around i coordinate points $\left\{\begin{array}{c}{y}^{\&prime;}\&le;y_i\\ x^{\&prime;}\&le;x_i\end{array}\right.$ J coordinate points, j is that [0, J), J represents array { (x for integer and j ∈ _Ij', y _Ij') in contained element number;

With rectangular coordinate (x _i, y _i) horizontal ordinate x _iDeduct respectively array { (x _Ij', y _Ij') in the horizontal ordinate x of each rectangular coordinate _Ij', rectangular coordinate (x _i, y _i) ordinate y _iDeduct respectively array { (x _Ij', y _Ij') in the ordinate y of each rectangular coordinate _Ij' obtain (x _i-x _I1', y _i-y _I1'), (x _i-x _I2', y _i-y _I2') ... (x _i-x _Ij', y _i-y _Ij'), the binary number add that is 2J*k bit with a width again represents (x _i-x _I1', y _i-y _I1'), (x _i-x _I2', y _i-y _I2') ... (x _i-x _Ij', y _i-y _Ij'), method for expressing is: 0～(k-1) bit of binary number add represents numerical value x _i-x _I1', binary number add k～(2k-1) bit represents numerical value y _i-y _I1', 2k～(3k-1) bit represents numerical value x _i-x _I2', binary number add 3k～(4k-1) bit represents numerical value y _i-y _I2' ..., (2 (j-1) * k)～((2j-1) * k-1) bit of binary number add represents numerical value x _i-x _Ij', ((2j-1) * k)～((2j*k)-1) bit of binary number add represents numerical value y _i-y _Ij', wherein k is all x _i-x _Ij' and y _i-y _IjMaximization in the ' numerical value is the bit wide behind the binary number;

With rectangular coordinate (x _i, y _i) and corresponding add value save as (x, y, add) _i, (x, y, add) _iRepresented rectangular coordinate (x _i, y _i) and array { (x _Ij', y _Ij');

S4. repeat s3, i value 0,1,2 successively wherein ..., MN/8-1 is again with all (x, y, add) _iBe saved in look-up table;

S5. processor n in receiving radar scanning circumference range ₁Behind bar sweep trace and the corresponding half-tone information thereof, with n ₁The bar sweep trace is transformed into respective scan line n in 0 °～45 ° scopes by the mapping formula;

Wherein, n ₁[0, N), the mapping formula is ∈ $\left\{\begin{array}{cc}n=n_1,& 0\&le;n_1\&le;N/8\\ n=N/4-n_1,& N/8<n_1\&le;N/4\\ n=n_1-N/4,& N/4<n_1\&le;3N/8\\ n=N/2-n_1,& 3N/8<n_1\&le;N/2\\ n=n_1-N/2,& N/2<n_1\&le;5N/8\\ n=3N/4-n_1,& 5N/8<n_1\&le;3N/4\\ n=n_1-3N/4,& 3N/4<n_1\&le;7N/8\\ n=N-n_1,& 7N/8<n_1<N\end{array}\right.;$

Obtain M corresponding to a sweep trace n number of scan points certificate by the addressing look-up table, with this M number of scan points according to forming array

n ₁∈ [0, N);

S6. the array that s5 is obtained

Be saved in the first memory of processor, with n ₁The gray-scale value that each analyzing spot is corresponding on the bar sweep trace is saved in the second memory of processor;

S7. from first memory, take out n ₁R number of scan points certificate on the bar sweep trace

R ∈ [0, M), and from second memory read with

Corresponding gray-scale value; According to

In rectangular coordinate

And the add value obtains array

With rectangular coordinate

And array

In each coordinate element substitution conversion formula obtain rectangular coordinate With

Wherein, conversion formula is $\left\{\begin{array}{cc}{x}_1=x,y_1=y,& 0\&le;n_1\&le;N/8\\ x_1=y,y_1=x,& N/8<n_1\&le;N/4\\ x_1=-y,y_1=x,& N/4<n_1\&le;3N/8\\ x_1=-x,y_1=y,& 3N/8<n_1\&le;N/2\\ x_1=-x,y_1=-y,& N/2<n_1\&le;5N/8\\ x_1=-y,y_1=-x,& 5N/8<n_1\&le;3N/4\\ x_1=y,y_1=-x,& 3N/4<n_1\&le;7N/8\\ x_1=x,y_1=-y,& 7N/8<n_1<N\end{array}\right.;$

With rectangular coordinate

With

Corresponding gray-scale value outputs to video memory together;

With array

In each rectangular coordinate respectively with

Corresponding gray-scale value outputs to video memory together.

2. the coordinate transformation method in a kind of pulsed radar imaging according to claim 1, it is characterized in that described step s4 is kept at look-up table in the external memorizer of processor, when processor starts, look-up table is copied to the internal memory from external memorizer; First memory among described step s6, the described step s7 and second memory are the on-chip memory of processor.

3. the coordinate transformation method in a kind of pulsed radar imaging according to claim 2 is characterized in that described processor adopting FPGA.

4. the coordinate transformation method in a kind of pulsed radar imaging according to claim 3 is characterized in that described external memorizer adopts NANDFLASH, and described internal memory adopts SDRAM, and described on-chip memory adopts the FIFO storer.

Images