CN112669332A - Method for judging sea and sky conditions and detecting infrared target based on bidirectional local maximum and peak local singularity - Google Patents

Abstract

The invention provides a method for judging sea and sky conditions and detecting an infrared target based on a bidirectional local maximum and a peak local singularity, which comprises the following steps: firstly, inputting a frame of marine infrared image, and performing wavelet denoising on the image based on a haar wavelet basis; then, refining the sea-sky area by adopting a bidirectional local maximum method, extracting a suspected sea-sky antenna, designing a 'false removing' strategy to remove wave fluctuation interference, and realizing accurate positioning of the sea-sky antenna; secondly, detecting small targets (pixels 2x 2-9 x9) in the sea and sky area by applying a peak local singularity method; and finally, a CEDoG filtering method is adopted to inhibit the background, the target significance is improved, and the accuracy and the integrity of the marine target detection are realized by searching the most significant region and a self-designed region growing rule by adopting a maximum between-class variance method.

Description

Method for judging sea and sky conditions and detecting infrared target based on bidirectional local maximum and peak local singularity

Technical Field

The invention relates to the technical field of image processing, in particular to a method for judging sea and sky conditions and detecting an infrared target based on bidirectional local maximum and peak local singularity

Background

With the development of economy and science and technology, the activities of the sea are continuously increased, and due to the complex and changeable marine environment, the detection, identification and tracking of targets at long distances on the sea are always difficult and bottleneck problems in the fields of modern military and civil use. The detection of small or multiple infrared targets of unknown position and velocity in complex environments is a significant problem in infrared search and tracking systems, a necessary application for the introduction of warnings from distant locations from targets at sea.

In recent years, maritime search and rescue equipment is mainly composed of a visible light camera and an infrared camera. Compared with a visible light camera, the infrared camera has the advantages of strong fog penetration, long shooting distance and capability of working day and night, so that an infrared search and tracking system becomes a main method for detecting a marine long-distance target. The marine environment with uncertain background clutter and sea wave noise is still the most complex situation that appears when detecting a remote target, and brings great difficulty and challenge to target detection. And has been studied and discussed by a wide range of scholars.

Disclosure of Invention

According to the technical problem, the invention provides a method for judging sea-sky conditions and detecting infrared targets based on two-way local maximum and peak local singularity, firstly, refining a sea-sky area by adopting the two-way local maximum method, extracting suspected sea-sky antennas, and designing a 'false-removing' strategy to obtain the precisely positioned sea-sky antennas; secondly, detecting small targets (pixels 2x 2-9 x9) in the sea and air area by using a local peak singularity method; and finally, a CEDoG filtering method is adopted to inhibit the background and improve the significance of the target, and the method adopts a maximum between-class variance method to search a most significant region and a self-designed region growth rule so as to ensure the accuracy and the integrity of the target detection in the ocean region.

The technical means adopted by the invention are as follows:

a method for judging sea and sky conditions and detecting infrared targets based on bidirectional local maximum and peak local singularity comprises the following steps:

s1, inputting a frame of infrared marine image, and performing wavelet denoising on the input infrared marine image based on a haar wavelet basis to obtain a low-frequency image;

s2, adopting [ 1/61/61/6; 000; -1/6-1/6-1/6 ] operator convolves the low-frequency image obtained in step S1, converts the spatial domain into a gradient domain, obtains all rough texture information under the low-frequency image, and determines a minimum cut-off value of the texture information by using a root mean square estimation threshold method;

s3, refining the texture information obtained in the step S2 through a bidirectional local maximum method to obtain sea-sky texture information, designing a 'false removing' strategy to remove wave fluctuation interference, realizing detection of sea-sky antennas, and obtaining accurately positioned sea-sky antennas;

s4, performing convolution on the area near the sea-sky-line obtained in the step S3 and a peak filter to detect a peak point, eliminating peak point interference on the sea-sky-line by adopting a peak local singularity method, and performing expansion operation on the peak point to detect a target near the sea-sky-line;

s5, processing the sea area below the sea antenna obtained in the step S3 by adopting a CEDoG filtering method, obtaining an adaptive threshold value through the mean value and the variance of a result graph, carrying out image segmentation on the sea area below the sea antenna after processing, segmenting the image by adopting a maximum inter-class variance method with a high threshold value, eliminating noise points and reconstructing the original area of a target area, and carrying out region growth on seed points to realize the detection of the sea target;

and S6, integrating the detection results obtained in the steps S4 and S5, and realizing the detection of the targets near the whole sea surface and sea antennas.

Further, the root mean square estimation threshold method adopted in step S2 has the following formula:

where scale represents the scale, h represents the height of the image, and w represents the width of the image.

Further, the bidirectional local maximum method in step S3 is specifically:

and if the reference point has the maximum gray value in any one of the two directions, the reference point is regarded as a suspected reference point constituting the sea-sky-line.

Further, the detection of the sea-sky-line in the step S3 specifically includes:

and calculating the difference value of the left end point and the right end point of each connected domain in the horizontal direction, and setting 0.78 × width as a final threshold value to accurately position the sea-sky-line, wherein the width represents the width of the image.

Further, the peak local singularity method adopted in step S4 specifically includes:

and finding out the position corresponding to the suspected target point, calculating the local singularity of each area, and reserving the area above the intermediate value as a detection result.

Further, the step S5, which uses the CEDoG filtering method specifically includes:

σ_fe＝2.7-S/1600

σ_bi＝0.5+S×0.15

wherein σ_feAnd σ_biRepresenting the parameters of foreground boosting and background suppression, respectively.

Compared with the prior art, the invention has the following advantages:

1. the method for judging the sea-sky condition and detecting the infrared target based on the two-way local maximum and the peak local singularity, provided by the invention, has the advantages that the sea-sky texture information is further refined by solving the two-way local maximum when the sea-sky antenna is detected, and compared with other sea-sky-antenna detection methods, the method provided by the invention is extremely favorable for accurately positioning the sea-sky antenna.

2. The invention provides a method for judging sea-sky conditions and detecting infrared targets based on bidirectional local maximum values and peak local singularities, which aims to realize accurate detection of targets near sea-sky antennas.

3. The invention provides a method for judging sea-sky conditions and detecting an infrared target based on a bidirectional local maximum and a peak local singularity, which aims to achieve the purposes of accurate detection and reservation of the original area size of the target.

4. Compared with other sea surface infrared image detection modes, the method has robustness for improving SCR and BSF values of images, has high accuracy, recall rate and short running time, and has excellent performance in the aspects of detection rate and false alarm rate.

For the above reasons, the present invention can be widely applied to the fields of image processing and the like.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to these drawings without creative efforts.

FIG. 1 is a schematic flow chart of the method of the present invention.

Fig. 2 is a test picture and a segmentation result provided in the embodiment of the present invention.

Detailed Description

In order to make the technical solutions of the present invention better understood, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

It should be noted that the terms "first," "second," and the like in the description and claims of the present invention and in the drawings described above are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprises," "comprising," and "having," and any variations thereof, are intended to cover a non-exclusive inclusion, such that a process, method, system, article, or apparatus that comprises a list of steps or elements is not necessarily limited to those steps or elements expressly listed, but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus.

Examples

As shown in fig. 1, the present invention provides a method for determining a sea-sky condition and detecting an infrared target based on a bidirectional local maximum and a peak local singularity, comprising the following steps:

s1, as shown in FIG. 2, inputting a frame of infrared marine image, and performing wavelet denoising on the input infrared marine image based on a haar wavelet basis to obtain a low-frequency image;

s2, adopting [ 1/61/61/6; 000; -1/6-1/6-1/6 ] operator convolves the low-frequency image obtained in step S1, converts the spatial domain into a gradient domain, obtains all rough texture information under the low-frequency image, and determines a minimum cut-off value of the texture information by using a root mean square estimation threshold method;

in a specific implementation, as a preferred embodiment of the present invention, the root mean square estimation threshold method used in step S2 has the following formula:

where scale represents the scale, h represents the height of the image, and w represents the width of the image.

S3, refining the texture information obtained in the step S2 through a bidirectional local maximum method to obtain sea-sky texture information, designing a 'false removing' strategy to eliminate wave fluctuation interference, realizing detection of sea antennas, obtaining accurately positioned sea antennas, removing sky areas, and splitting infrared pictures into sea-antenna areas and sea surface areas, as shown in figure 2.

In a specific implementation, as a preferred embodiment of the present invention, the bidirectional local maximum method in step S3 specifically includes: and if the reference point has the maximum gray value in any one of the two directions, the reference point is regarded as a suspected reference point constituting the sea-sky-line.

The detection of the sea-sky-line in the step S3 specifically includes: and calculating the difference value of the left end point and the right end point of each connected domain in the horizontal direction, and setting 0.78 × width as a final threshold value to accurately position the sea-sky-line, wherein the width represents the width of the image.

S4, performing convolution on the area near the sea-sky-line obtained in the step S3 and a peak filter to detect a peak point, as shown in figure 2, eliminating peak point interference on the sea-sky-line by adopting a peak local singularity method, as shown in figure 2, and performing expansion operation on the peak point to detect a target near the sea-sky-line;

in a specific implementation, as a preferred embodiment of the present invention, the method for local singularity of peak values adopted in step S4 specifically includes: and finding out the position corresponding to the suspected target point, calculating the local singularity of each area, and reserving the area above the intermediate value as a detection result.

S5, processing the sea area below the sea antenna obtained in the step S3 by adopting a CEDoG filtering method, obtaining an adaptive threshold value through the mean value and the variance of a result graph, carrying out image segmentation on the sea area below the sea antenna after processing, segmenting the image by adopting a maximum inter-class variance method with a high threshold value, eliminating noise points and reconstructing the original area of a target area, and carrying out region growth on seed points to realize the detection of the sea target;

in specific implementation, as a preferred embodiment of the present invention, the step S5 specifically includes:

σ_fe＝2.7-S/1600

σ_bi＝0.5+S×0.15

wherein σ_feAnd σ_biRepresenting the parameters of foreground boosting and background suppression, respectively. In the present embodiment, σ_fe＝2.66，σ_biThe lower sea area of the sea-sky-line after processing is shown in fig. 2, and the adaptive threshold obtained by the mean and variance of the result map is 1.58 ═ 10.10

S6, integrating the detection results obtained in the steps S4 and S5 to realize the detection of the targets near the whole sea surface and sea antennas, wherein the obtained detection results are shown in FIG. 2.

Finally, it should be noted that: the above embodiments are only used to illustrate the technical solution of the present invention, and not to limit the same; while the invention has been described in detail and with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some or all of the technical features may be equivalently replaced; and the modifications or the substitutions do not make the essence of the corresponding technical solutions depart from the scope of the technical solutions of the embodiments of the present invention.

Claims (6)

1. A method for judging sea and sky conditions and detecting infrared targets based on bidirectional local maximum and peak local singularity is characterized by comprising the following steps:

s1, inputting a frame of infrared marine image, and performing wavelet denoising on the input infrared marine image based on a haar wavelet basis to obtain a low-frequency image;

s2, adopting [ 1/61/61/6; 000; -1/6-1/6-1/6 ] operator convolves the low-frequency image obtained in step S1, converts the spatial domain into a gradient domain, obtains all rough texture information under the low-frequency image, and determines a minimum cut-off value of the texture information by using a root mean square estimation threshold method;

s3, refining the texture information obtained in the step S2 through a bidirectional local maximum method to obtain sea-sky texture information, designing a 'false removing' strategy to remove wave fluctuation interference, realizing detection of sea-sky antennas, and obtaining accurately positioned sea-sky antennas;

s4, performing convolution on the area near the sea-sky-line obtained in the step S3 and a peak filter to detect a peak point, eliminating peak point interference on the sea-sky-line by adopting a peak local singularity method, and performing expansion operation on the peak point to detect a target near the sea-sky-line;

s5, processing the sea area below the sea antenna obtained in the step S3 by adopting a CEDoG filtering method, obtaining an adaptive threshold value through the mean value and the variance of a result graph, carrying out image segmentation on the sea area below the sea antenna after processing, segmenting the image by adopting a maximum inter-class variance method with a high threshold value, eliminating noise points and reconstructing the original area of a target area, and carrying out region growth on seed points to realize the detection of the sea target;

and S6, integrating the detection results obtained in the steps S4 and S5, and realizing the detection of the targets near the whole sea surface and sea antennas.

2. The method for determining the conditions of the sea and detecting the infrared target based on the two-way local maxima and the peak local singularities of claim 1, wherein the root mean square estimation threshold method used in step S2 is as follows:

where scale represents the scale, h represents the height of the image, and w represents the width of the image.

3. The method for determining the sea-sky condition and detecting the infrared target based on the two-way local maxima and the peak local singularity according to claim 1, wherein the two-way local maxima method in step S3 is specifically:

and if the reference point has the maximum gray value in any one of the two directions, the reference point is regarded as a suspected reference point constituting the sea-sky-line.

4. The method for determining the sea-sky condition and detecting the infrared target based on the two-way local maximum and the peak local singularity according to claim 1, wherein the step S3 of detecting the sea-sky-line is specifically as follows:

and calculating the difference value of the left end point and the right end point of each connected domain in the horizontal direction, and setting 0.78 × width as a final threshold value to accurately position the sea-sky-line, wherein the width represents the width of the image.

5. The method for determining the sea-sky condition and detecting the infrared target based on the two-way local maximum and the peak local singularity according to claim 1, wherein the peak local singularity method adopted in the step S4 is specifically:

and finding out the position corresponding to the suspected target point, calculating the local singularity of each area, and reserving the area above the intermediate value as a detection result.

6. The method for determining the sea-sky condition and detecting the infrared target based on the two-way local maximum and the peak local singularity according to claim 1, wherein the step S5 of applying the CEDoG filtering method specifically comprises:

σ_fe＝2.7-S/1600

σ_bi＝0.5+S×0.15

wherein σ_feAnd σ_biRepresenting the parameters of foreground boosting and background suppression, respectively.

Images