CN102288945B - Image enhancing method for ship radar - Google Patents

Abstract

The invention discloses an image enhancing method for a ship radar. Due to the noise interference in radar image data, certain isolated points, isolated regions, small gaps and holes exist apart from a needed target, which influence the post-processing accuracy. The method is disclosed specific to the defect and requirement. By performing quick time accumulation, CFAR (Constant False Alarm Rate) processing and slow time accumulation, signal attenuation is avoided as much as possible while noise is removed, an appropriate morph operator and an appropriate filtering window are selected, isolated noise generated after binaryzation is suppressed, fine gaps inside an object are eliminated, radar image display is more clear and definite, noise can be suppressed effectively, a displayed object is more mellow and full, source data with higher quality are provided for post-processing, and the post-processing performance is finally enhanced. The method comprises the following specific steps of: parameter initializing, quick time accuracy, CFAR processing, slow time accumulation, binaryzation, corroding and expanding.

Description

A kind of marine radar image enchancing method

Technical field

The invention belongs to the marine radar technical field, particularly wherein image enchancing method.

Background technology

Marine radar is one of indispensable navigator of boats and ships; But the detection of marine radar and tracking performance usually can receive wave surface scattering electromagnetic wave (being extra large clutter) and various The noise, in interference of noise, in the image of cutting apart except the target of needs; Also there are some isolated point, isolated zonule, little gap and holes; Need give prominence to marine object when marine radar shows and exist, prompting is controlled personnel and is dodged, and there is interference with the zone in these points to the judgement of target location from now on; Cause next following the tracks of and lose or trail-and-error, therefore need strengthen the marine radar view data.At present, reasonable method does not also appear addressing these problems.

Summary of the invention

The objective of the invention is to handle the problem that occurs, proposed a kind of Enhancement Method of marine radar image in order to solve existing marine radar technical field view data.

To achieve these goals, technical scheme of the present invention is: a kind of Enhancement Method of marine radar image comprises the steps:

S1. the initialization setting of key control parameter is accomplished in initialization, specifically comprises following parameter:

The window size of fast time filtering: wf, the window size that CFAR handles: wc, the adjustability coefficients that CFAR handles: α, CFAR threshold value: Threshold, the window size of slow time filtering: ws, binary-state threshold: T _b, echo growth intensity: s; Sweep trace counter: i=0;

S2. import i bar radar scanning line data, be designated as { T _i(k), k=1,2 ..., Num}, simultaneously the sweep trace counter adds 1, i.e. i++, wherein Num is the number of data on the sweep trace, i is the sweep trace counter;

S3. fast time integral is to scan-line data T _i(k) carry out fast time integral, specific as follows:

$\{{\mathrm{TF}}_i\left(k\right)=\frac{1}{\mathrm{wf}}\underset{p=-[\mathrm{wf}/2]}{\overset{[\mathrm{wf}/2]}{\mathrm{\Σ}}}{T}_i(k+p),k=[\mathrm{wf}/2]+1,[\mathrm{wf}/2]+2,...,\mathrm{Num}-[\mathrm{wf}/2\left]\right\}$

Wherein [] is rounding operation symbol, TF _i(k) be fast time result;

S4.CFAR handles, to fast time result TF _i(k) carry out CFAR and handle, guarantee that promptly the data point of sweep trace and the difference of its each [wc/2] data points of front and back are not more than CFAR threshold value Threshold, specific as follows:

$\{{\mathrm{TR}}_i\left(k\right)={\mathrm{TF}}_i\left(k\right)-\mathrm{\α}\underset{p=-[\mathrm{wc}/2]}{\overset{[\mathrm{wc}/2]}{\mathrm{\Σ}}}{\mathrm{TF}}_i(k+p),k=[\mathrm{wc}/2]+1,[\mathrm{wc}/2]+2,...,\mathrm{Num}-[\mathrm{wc}/2\left]\right\}$

{TFR _i(k)＝max(TR _i(k)，Threshold)，k＝1，2，...，Num}

Wherein [] is rounding operation, TFR _i(k) be the CFAR result;

S5. slow time integral is to CFAR result TFR _i(k) carry out slow time integral, specific as follows:

$\left\{\begin{array}{cc}{\mathrm{TFRS}}_i\left(k\right)=\frac{1}{\mathrm{ws}}\underset{p=i-\mathrm{ws}+1}{\overset{i}{\mathrm{\Σ}}}{\mathrm{TFR}}_p\left(k\right),k=\mathrm{1,2},...,\mathrm{Num}& \mathrm{if\; i}>\mathrm{ws}\\ {\mathrm{TFRS}}_i\left(k\right)={\mathrm{TFR}}_i\left(k\right)& \mathrm{otherwise}\end{array}\right.$

TFRS wherein _i(k) be slow time result;

S6. binary conversion treatment is to slow time result TFRS _i(k) carry out binary conversion treatment:

$D_i\left(k\right)=\left\{\begin{array}{cc}1& {\mathrm{TFRS}}_i\left(k\right)>T_b,k=\mathrm{1,2},...,\mathrm{Num}\\ 0& \mathrm{else}\end{array}\right.$

D wherein _i(k) be the binary conversion treatment result;

S7. echo expansion, specific as follows:

S71. to the data D of binaryzation _i(k) carry out the erosion operation that intensity is s, obtain De as a result _i(k);

S72. to De _i(k) carry out the dilation operation that intensity is s;

S9. judge whether marine radar range or parameter preset change, if changed, execution in step S1 reinitializes parameter, otherwise, execution in step S2.

Behind step S7, also comprise step S8, show the data that current scan line is handled.

Beneficial effect of the present invention: the Enhancement Method of the marine radar image that the present invention proposes, handle and slow time integral through fast time integral, CFAR, choose the suitable morphological operator and the parameters such as size of spectral window, the good restraining noise spot; Eliminate the small spaces of interior of articles, make target full mellow and full, made things convenient for operating personnel's use; For the marine radar subsequent treatment provides the image source data of better quality, improved the performance of marine radar subsequent treatment.

Description of drawings

Fig. 1 is the schematic flow sheet of marine radar image enchancing method of the present invention.

Fig. 2 is the measured data synoptic diagram of the marine radar of the embodiment of the invention.

Fig. 3 be the embodiment of the invention measured data is directly carried out the synoptic diagram as a result after the binaryzation.

Fig. 4 is the synoptic diagram as a result to measured data employing marine radar image enchancing method of the embodiment of the invention.

Embodiment

Below in conjunction with accompanying drawing, provide specific embodiment of the present invention.Need to prove: the parameter among the embodiment does not influence generality of the present invention.

The enhancing of marine radar data mainly is to suppress background and noise from the angle analysis of Flame Image Process, highlights target as far as possible, thereby improves signal to noise ratio (S/N ratio).The performance of the enhancement algorithms of marine radar image is most important for Signal and Information Processing algorithms such as design object detection, estimation, tracking and identifications.The schematic flow sheet of the embodiment of the invention is as shown in Figure 1, the time 2.5s of the marine radar run-down of this instance, and all 3600 scan-datas, every sweep trace 1000 data points, the concrete expansion as follows:

S1. the initialization setting of parameter is accomplished in initialization.Can specifically choose parameter according to actual conditions; Being provided with as follows of parameter in the present embodiment: the window size of fast time filtering: wf=5; The window size that CFAR handles: wc=32, the adjustable threshold that CFAR handles: α=1, Threshold=32; The window size of slow time filtering: ws=5, the CFAR P of target detection _f=0.01, echo growth intensity: s=5; Sweep trace counter: i=0;

S2. read current radar scanning line data, be designated as { T _i(k), k=1,2 ..., Num}, Num=1000, the sweep trace counter adds 1 simultaneously, i.e. i++;

S3. fast time integral carries out fast time integral to current scan line data Ti (k), as shown in the formula

$\{{\mathrm{TF}}_i\left(k\right)=\frac{1}{5}\underset{p=-2}{\overset{2}{\mathrm{\Σ}}}{T}_i(k+p),k=\mathrm{3,4},...,998\}$

TF wherein _i(k) be fast time result.

S4. CFAR detection (CFAR, Constant False Alarm Rate) is handled, to scan-line data TF _i(k) carry out CFAR and handle, guarantee that promptly the data point of sweep trace and the difference of its each [wc/2]=16 data points of front and back are not more than CFAR threshold value Threshold=32, as shown in the formula:

$\{{\mathrm{TR}}_i\left(k\right)={\mathrm{TF}}_i\left(k\right)-\underset{p=-16}{\overset{16}{\mathrm{\Σ}}}{\mathrm{TF}}_i(k+p),k=\mathrm{17,18},...,986\}$

{TFR _i(k)＝max(TR _i(k)，32)，k＝1，2，...，1000}

TFR wherein _i(k) be the CFAR result.

S5. slow time integral is to CFAR result TFR _i(k) carry out slow time integral, specific as follows:

$\left\{\begin{array}{cc}{\mathrm{TFRS}}_i\left(k\right)=\frac{1}{5}\underset{p=i-4}{\overset{i}{\mathrm{\Σ}}}{\mathrm{TFR}}_p\left(k\right),k=\mathrm{1,2},...,1000& \mathrm{if\; i}>15\\ {\mathrm{TFRS}}_i\left(k\right)={\mathrm{TFR}}_i\left(k\right)& \mathrm{otherwise}\end{array}\right.$

TFRS wherein _i(k) be slow time result.

S6. binary conversion treatment is to slow time result TFR _i(k) carry out binary conversion treatment, as shown in the formula:

$D_i\left(k\right)=\left\{\begin{array}{cc}1& {\mathrm{TFRS}}_i\left(k\right)>\mathrm{Thresh},k=\mathrm{1,2},...,1000\\ 0& \mathrm{else}\end{array}\right.$

D wherein _i(k) be the binary conversion treatment result.

S7. echo expansion, specific as follows:

S71: to the data D of binaryzation _i(k) carrying out intensity is 5 erosion operation, obtains De as a result _i(k);

S72: to De _i(k) carrying out intensity is 5 dilation operation; Promptly accomplished the enhancing of the view data of current scan line.

S8: show the data that current scan line is handled;

The time 2.5s of the marine radar run-down of this instance, all 3600 scan-datas, the time of display delay is 0.00348s, visible algorithm of the present invention has the high characteristics of real-time.

S9. judge whether radar range or parameter preset change, if changed, execution in step S1, from new initialization, otherwise execution in step S2.

To the employing of this example is a frame radar image data.Fig. 2 is the measured data synoptic diagram of the marine radar of the embodiment of the invention.

Fig. 3 be the embodiment of the invention measured data is directly carried out the synoptic diagram as a result after the binaryzation; Fig. 4 is the synoptic diagram as a result to measured data employing marine radar image enchancing method of the embodiment of the invention.Can find out from result's contrast of Fig. 3 and Fig. 4,, eliminate a large amount of assorted points, thereby made target more clear and mellow and full, be convenient to operating personnel's identification, dodge timely through the enforcement of this method.

Can find out from present embodiment, the marine radar image enchancing method that the present invention proposes, the user can choose the suitable morphological operator and the parameters such as size of spectral window according to different sea conditions, thus fine inhibition noise spot; Eliminate the small spaces of interior of articles, make target full mellow and full, the handled easily personnel use; For the marine radar subsequent treatment provides the image source data of better quality, improved marine radar subsequent treatment performance.

Those of ordinary skill in the art will appreciate that embodiment described here is in order to help reader understanding's principle of the present invention, should to be understood that the protection domain of inventing is not limited to such special statement and embodiment.Every making according to foregoing description variously possible be equal to replacement or change, and all is considered to belong to the protection domain of claim of the present invention.

Claims (2)

1. the method for marine radar figure image intensifying is characterized in that, comprises the steps:

S1. the initialization setting of key control parameter is accomplished in initialization, and concrete parameter is provided with as follows:

The window size of fast time filtering: wf, the window size that CFAR handles: wc, the adjustability coefficients that CFAR handles: α, CFAR threshold value: Threshold, the window size of slow time filtering: ws, binary-state threshold: Tb, echo growth intensity: s; Sweep trace counter: i=0;

S2. import i bar radar scanning line data, be designated as { T _i(k), k=1,2 ..., Num}, simultaneously the sweep trace counter adds 1, i.e. i++, wherein Num is the number of data on the sweep trace, i is the sweep trace counter;

S3. fast time integral is to scan-line data T _i(k) carry out fast time integral, specific as follows:

$\{{\mathrm{TF}}_i\left(k\right)=\frac{1}{\mathrm{wf}}\underset{p=-[\mathrm{wf}/2]}{\overset{[\mathrm{wf}/2]}{\mathrm{\Σ}}}{T}_i(k+p),k=[\mathrm{wf}/2]+1,[\mathrm{wf}/2]+2,...,\mathrm{Num}-[\mathrm{wf}/2\left]\right\}$

Wherein [] is rounding operation symbol, TF _i(k) be fast time result;

S4.CFAR handles, to fast time result TF _i(k) carry out CFAR and handle, guarantee that promptly the data point of sweep trace and the difference of its each [wc/2] data points of front and back are not more than CFAR threshold value Threshold, specific as follows:

$\{{\mathrm{TR}}_i\left(k\right)={\mathrm{TF}}_i\left(k\right)-\mathrm{\α}\underset{p=-[\mathrm{wc}/2]}{\overset{[\mathrm{wc}/2]}{\mathrm{\Σ}}}{\mathrm{TF}}_i(k+p),k=[\mathrm{wc}/2]+1,[\mathrm{wc}/2]+2,...,\mathrm{Num}-[\mathrm{wc}/2\left]\right\}$

{TFR _i(k)=max(TR _i(k),Threshold)，k=1,2,...,Num}

Wherein [] is rounding operation, TFR _i(k) be the CFAR result;

S5. slow time integral is to CFAR result TFR _i(k) carry out slow time integral, specific as follows:

$\left\{\begin{array}{cc}{\mathrm{TFRS}}_i\left(k\right)=\frac{1}{\mathrm{ws}}\underset{p=i-\mathrm{ws}+1}{\overset{i}{\mathrm{\Σ}}}{\mathrm{TFR}}_p\left(k\right),k=\mathrm{1,2},...,\mathrm{Num}& \mathrm{if\; i}>\mathrm{ws}\\ {\mathrm{TFRS}}_i\left(k\right)={\mathrm{TFR}}_i\left(k\right)& \mathrm{otherwise}\end{array}\right.$

TFRS wherein _i(k) be slow time result;

S6. binary conversion treatment is to slow time result TFRS _i(k) carry out binary conversion treatment:

$D_i\left(k\right)=\left\{\begin{array}{cc}1& {\mathrm{TFRS}}_i\left(k\right)>T_b,k=\mathrm{1,2},...,\mathrm{Num}\\ 0& \mathrm{else}\end{array}\right.$

D wherein _i(k) be the binary conversion treatment result;

S7. echo expansion, specific as follows:

S71. to the data D of binaryzation _i(k) carry out the erosion operation that intensity is s, obtain De as a result _i(k);

S72. to De _i(k) carry out the dilation operation that intensity is s;

S9. judge whether marine radar range or parameter preset change, if changed, execution in step S1 reinitializes parameter, otherwise, execution in step S2.

2. the method for marine radar figure according to claim 1 image intensifying is characterized in that, behind step S7, also comprises step S8, shows the data that current scan line is handled.

Images