Ship target segmentation method and system based on improved Harris eagle optimization algorithm

Abstract

The invention provides a ship target segmentation method and system based on an improved Harris eagle optimization algorithm, and relates to the technical field of image processing and mode recognition. According to the technical key points, the method comprises the following steps: processing an airborne grayscale image including a sky domain, a sea area, a marine ship and a coastline, and obtaining a region-of-interest image; obtaining an optimal segmentation threshold value by using an improved Harris eagle optimization algorithm, and segmenting the airborne grayscale image by using the optimal segmentation threshold value to obtain a segmented image based on the optimal segmentation threshold value; and fusing the region-of-interest image and the segmented image based on the optimal segmentation threshold to obtain a final segmented image. According to the method, the high response area of the ship target is accurately locked through the SIFT features, and adaptive threshold optimization is realized in the target gray band in combination with the improved Harris eagle optimization algorithm, so that the ship target segmentation precision and robustness under the complex sea condition are remarkably improved, and high-reliability technical support is provided for maritime affair monitoring and other scenes.

Description

Technical Field

[0001] The present invention relates to the technical field of image processing and pattern recognition, and in particular to a ship target segmentation method and system based on an improved Harris Hawk optimization algorithm. Background Art

[0002] With the in-depth application of infrared imaging technology in the field of ocean monitoring, airborne infrared imagery has become a key means of ship target identification due to its all-weather perception capability. However, in the complex sea and sky background, there are core challenges such as weak radiation difference between target and background, strong interference from sea surface clutter, and significant attenuation from mist, which leads to bottlenecks such as poor adaptability and insufficient accuracy of traditional segmentation methods. In recent years, the Scale-Invariant Feature Transform (SIFT) has demonstrated unique advantages in low-contrast scenes due to its stable representation of ship structural features. At the same time, the innovative application of meta-heuristic optimization algorithms has provided a new solution path for adaptive threshold technology. It is worth noting that traditional optimization algorithms are prone to falling into local optimality in the dynamic interference environment of the sea surface, and there is an urgent need to combine feature space constraint mechanisms to improve robustness. Summary of the Invention

[0003] In view of the above problems, the present invention proposes a ship target segmentation method and system based on an improved Harris Hawk optimization algorithm.

[0004] According to one aspect of the present invention, a ship target segmentation method based on an improved Harris Hawk optimization algorithm is proposed, the method comprising:

[0005] Process the airborne grayscale image containing the sky, sea, ships and coastline to obtain the image of the area of ​​interest;

[0006] Using an improved Harris Hawk optimization algorithm to obtain an optimal segmentation threshold, and using the optimal segmentation threshold to segment the airborne grayscale image to obtain a segmented image based on the optimal segmentation threshold;

[0007] The region of interest image and the segmentation image based on the optimal segmentation threshold are fused to obtain a final segmentation image.

[0008] Furthermore, the processing of the airborne grayscale image including the sky area, the sea area, the ships at sea and the coastline to obtain the image of the area of ​​interest includes:

[0009] Extracting multiple key points from the airborne grayscale image using a SIFT algorithm, including: constructing a Gaussian difference pyramid and detecting local extreme points; filtering the multiple local extreme points based on contrast and edge response to obtain multiple key points;

[0010] Perform ellipse fitting on multiple key points to obtain the minimum circumscribed ellipse containing all key points;

[0011] Assign the image area corresponding to the minimum circumscribed ellipse to white, and the rest of the image area to black to generate a preliminary mask feature map of the region of interest;

[0012] Performing edge detection and land-sea segmentation line extraction on the airborne grayscale image; assigning the sea area below the land-sea segmentation line to white and the area above the land-sea segmentation line to black, thereby generating a land-sea segmentation mask map;

[0013] The preliminary region of interest mask feature map and the land and sea segmentation mask map are multiplied pixel by pixel to obtain the region of interest image.

[0014] Furthermore, the method of obtaining the optimal segmentation threshold using the improved Harris Hawk optimization algorithm includes:

[0015] In the exploration phase, the position update formula of the population individuals is:

[0016] ;

[0017] Where, represents the position of the i-th individual in the t+1-th iteration; Represents the position of a randomly selected individual in the population; r and q are uniform random numbers between [0,1]; is a uniformly distributed random number in [0,1]; lb and ub are the lower and upper bounds of the search space respectively;

[0018] In the transition phase from exploration to exploitation, the hawks catch prey, and the energy loss model of the prey consuming energy in the process of escaping is:

[0019] ;

[0020] Where, represents the nonlinear attenuation exponent; represents the initial escape energy of the prey; represents the decay curve index; represents the disturbance intensity coefficient; is the standard normal distribution noise; t represents the current number of iterations; T represents the maximum number of iterations;

[0021] In the development phase, the position update formula of the population individuals is:

[0022] ;

[0023] Where, represents the optimal prey position in the tth iteration; It indicates the energy of prey escape; Indicates the jump amplitude; represents the improved Levy flight term, , is the step size coefficient, is a standard normal random variable, is the Levy index, is the scaling factor, and λ is the iterative decay exponent.

[0024] Furthermore, the method of obtaining the optimal segmentation threshold using the improved Harris Hawk optimization algorithm further includes: after each iteration of each stage, truncating the out-of-bounds solution using the following formula:

[0025] .

[0026] According to another aspect of the present invention, a ship target segmentation system based on an improved Harris Hawk optimization algorithm is proposed, the system comprising:

[0027] An area of ​​interest acquisition module is configured to process an airborne grayscale image including a sky area, a sea area, a ship at sea, and a coastline to acquire an area of ​​interest image;

[0028] an optimal segmentation image acquisition module configured to obtain an optimal segmentation threshold using an improved Harris Hawk optimization algorithm, and segment the airborne grayscale image using the optimal segmentation threshold to obtain a segmented image based on the optimal segmentation threshold;

[0029] The ship target segmentation module is configured to fuse the region of interest image with the segmentation image based on the optimal segmentation threshold to obtain a final segmentation image.

[0030] Furthermore, the processing of the airborne grayscale image containing the sky area, the sea area, the ships at sea and the coastline in the region of interest acquisition module to obtain the region of interest image includes:

[0031] Extracting multiple key points of the airborne grayscale image using a SIFT algorithm; including: constructing a Gaussian difference pyramid and detecting local extreme points; filtering the multiple local extreme points based on contrast and edge response to obtain multiple key points;

[0032] Perform ellipse fitting on multiple key points to obtain the minimum circumscribed ellipse containing all key points;

[0033] Assign the image area corresponding to the minimum circumscribed ellipse to white, and the rest of the image area to black to generate a preliminary mask feature map of the region of interest;

[0034] Performing edge detection and land-sea segmentation line extraction on the airborne grayscale image; assigning the sea area below the land-sea segmentation line to white and the area above the land-sea segmentation line to black, thereby generating a land-sea segmentation mask map;

[0035] The preliminary region of interest mask feature map and the land and sea segmentation mask map are multiplied pixel by pixel to obtain the region of interest image.

[0036] Furthermore, the optimal segmentation image acquisition module uses the improved Harris Hawk optimization algorithm to obtain the optimal segmentation threshold, including:

[0037] In the exploration phase, the position update formula of the population individuals is:

[0038] ;

[0039] Where, represents the position of the i-th individual in the t+1-th iteration; Represents the position of a randomly selected individual in the population; r and q are uniform random numbers between [0,1]; is a uniformly distributed random number in [0,1]; lb and ub are the lower and upper bounds of the search space respectively;

[0040] In the transition phase from exploration to exploitation, the hawks catch prey, and the energy loss model of the prey consuming energy in the process of escaping is:

[0041] ;

[0042] Where, represents the nonlinear attenuation exponent; represents the initial escape energy of the prey; represents the decay curve index; represents the disturbance intensity coefficient; is the standard normal distribution noise; t represents the current number of iterations; T represents the maximum number of iterations;

[0043] In the development phase, the position update formula of the population individuals is:

[0044] ;

[0045] Where, represents the optimal prey position in the tth iteration; It indicates the energy of prey escape; Indicates the jump amplitude; represents the improved Levy flight term, , is the step size coefficient, is a standard normal random variable, is the Levy index, is the scaling factor, and λ is the iterative decay exponent.

[0046] Furthermore, the method of obtaining the optimal segmentation threshold using the improved Harris Hawk optimization algorithm in the optimal segmentation image acquisition module further includes: after each iteration of each stage, truncating the out-of-bounds solution using the following formula:

[0047] .

[0048] The beneficial technical effects of the present invention are:

[0049] The present invention proposes a method and system for ship target segmentation in airborne infrared images based on an improved Harris Hawk optimization algorithm. The high-response area of ​​the ship target is accurately locked through SIFT features, and the improved Harris Hawk optimization algorithm is combined to realize adaptive threshold optimization within the target grayscale band, thereby realizing directional optimization segmentation and effectively solving the convergence stability problem in complex radiation environments. In the improved Harris Hawk optimization algorithm, by expanding the global search space, the search adaptability under strong background interference is improved; and the exploration and development strategy is optimized, the response capability to weak targets is enhanced, and the accuracy of local detail segmentation details is improved; the addition of an improved Levy flight mechanism improves the ability to jump out of the local optimum and ensures the convergence effect of the algorithm under complex images; the boundary processing mechanism optimizes the continuity of the individual search path and maintains the integrity of the image grayscale structure. The present invention significantly suppresses the negative effects of wave reflection and mist interference; significantly improves the segmentation accuracy and robustness under complex sea conditions; breaks through the difficult problem of balancing segmentation accuracy and real-time performance under dynamic sea conditions, and provides highly reliable technical support for scenarios such as maritime monitoring and oil spill emergency response. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] The above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily apparent by reading the following detailed description with reference to the accompanying drawings, in which several embodiments of the present invention are shown by way of example and not limitation, in which:

[0051] Figure 1 Flowchart of a ship target segmentation method based on an improved Harris Hawk optimization algorithm according to an embodiment of the present invention;

[0052] Figure 2 is an example diagram of an original grayscale image in an embodiment of the present invention;

[0053] Figure 3 This is an example of an image retaining SIFT feature points in an embodiment of the present invention;

[0054] Figure 4 is an example of a minimum fitting ellipse image containing all SIFT feature points in an embodiment of the present invention;

[0055] Figure 5 This is an example of a preliminary ROI mask feature map in an embodiment of the present invention;

[0056] Figure 6 is an example of a Canny edge detection graph in an embodiment of the present invention;

[0057] Figure 7 This is an example of extracting a coastline image in an embodiment of the present invention;

[0058] Figure 8 This is an example of a land-sea segmentation mask map in an embodiment of the present invention;

[0059] Figure 9 is an example of an image of a region of interest in an embodiment of the present invention;

[0060] Figure 10 This is an example of image segmentation based on an optimal threshold in an embodiment of the present invention;

[0061] Figure 11 This is an example of the final ship segmentation image in the embodiment of the present invention. DETAILED DESCRIPTION

[0062] The principles and spirit of the present invention will be described below with reference to several exemplary embodiments. It should be understood that these embodiments are provided solely to enable those skilled in the art to better understand and implement the present invention, and are not intended to limit the scope of the present invention in any way. Rather, these embodiments are provided to make this disclosure more thorough and complete, and to fully convey the scope of the present disclosure to those skilled in the art.

[0063] Those skilled in the art will appreciate that embodiments of the present invention may be implemented as a system, apparatus, device, method, or computer program product. Therefore, the present disclosure may be implemented in the following forms: entirely in hardware, entirely in software (including firmware, resident software, microcode, etc.), or in a combination of hardware and software. It should be understood that any number of elements in the figures is for illustrative purposes only and not limiting, and any nomenclature is for distinction only and does not have any limiting meaning.

[0064] This paper combines SIFT features with an improved Harris Hawk optimization algorithm to propose a ship target segmentation method and system based on these features. By extracting SIFT features from airborne infrared imagery, the method effectively characterizes the local differences between the ship target and the sea-sky background. Furthermore, the improved Harris Hawk optimization algorithm is used to dynamically adjust the segmentation threshold to accommodate target segmentation requirements in varying sea conditions.

[0065] The embodiment of the present invention proposes a ship target segmentation method based on the improved Harris Hawk optimization algorithm. Figure 1 As shown, the method includes:

[0066] S1. Processing the airborne grayscale image containing the sky, sea, ships at sea and coastline to obtain an image of the area of ​​interest;

[0067] S2. Using an improved Harris Hawk optimization algorithm to obtain an optimal segmentation threshold, and using the optimal segmentation threshold to segment the airborne grayscale image to obtain a segmented image based on the optimal segmentation threshold;

[0068] S3. Fusing the region of interest image and the segmented image based on the optimal segmentation threshold to obtain a final segmented image.

[0069] The method starts from S1, in which an airborne grayscale image including a sky area, a sea area, ships at sea and a coastline is processed to obtain an image of a region of interest.

[0070] According to an embodiment of the present invention, the grayscale image of the airborne infrared image is as follows: Figure 2 As shown. The specific steps of S1 include: S11, using the SIFT algorithm to extract multiple key points of the airborne grayscale image; including: constructing a Gaussian difference pyramid and detecting local extreme points; filtering multiple local extreme points based on contrast and edge response to obtain multiple key points; S12, performing ellipse fitting on multiple key points to obtain the minimum circumscribed ellipse containing all key points; assigning the image area corresponding to the minimum circumscribed ellipse to white and the remaining image areas to black to generate a preliminary interest region mask feature map; S13, performing edge detection and sea-land dividing line extraction on the airborne grayscale image; assigning the sea area below the sea-land dividing line to white and the upper area to black to generate a sea-land segmentation mask map; S14, multiplying the preliminary interest region mask feature map and the sea-land segmentation mask map pixel by pixel to obtain an interest region image.

[0071] First, in S11, the SIFT feature extraction method is used to extract key points from the original image, specifically including:

[0072] 1) Scale space extreme value detection: Detect potential key points through the Difference of Gaussians (DoG) pyramid:

[0073]

[0074] in, is the Gaussian kernel, is the Gaussian scale parameter, k is the scale factor, usually ;I(m, n) is the image. After calculation, the difference between the blurred images at two different scales is output In each level of the DoG pyramid, local extrema are detected by comparing each pixel with its 26 neighboring pixels (8 neighbors at the same scale and 9×2 neighbors at the upper and lower scales).

[0075] 2) Precise key point localization:

[0076] 21) Taylor expansion fitting: At the candidate key point ( ), locally fit the DoG function D(x, y, σ) using a two-dimensional Taylor expansion:

[0077]

[0078] where , is a continuous variable in the scale space, representing any position and scale.

[0079]

[0080] is the offset of the key point (obtained by taking the derivative and setting it to zero).

[0081] 22) Eliminate low-contrast points: Substitute into the Taylor expansion formula to calculate the response (contrast) value of the key point:

[0082]

[0083] If |D( )| < tc, it is considered that the contrast of this key point is too low and is eliminated. In the embodiment of the present invention, tc is 0.01.

[0084] D( ) reflects the response strength near the key point and is the core area for judging whether this point is a valid key point. If D( ) is large, it indicates that this point is a significant extreme value in the image (such as a strong response at a corner or an edge); if D( ) is close to zero, it indicates that this point may be a weak response in noise or a flat area and has no distinguishability.

[0085] 23) Edge response filtering: Filter points with strong edge responses through the eigenvalue ratio of the Hessian matrix. The Hessian matrix is:

[0086]

[0087] where , , are the second-order derivatives of DoG. The eigenvalues α and β of the Hessian matrix (assuming α ≥ β) respectively represent the curvatures of the image in the two main directions at this point. The eigenvalue ratio γ:

[0088]

[0089] Among them, if , it is determined to be an edge corresponding point and is removed. In the embodiment of the present invention, ET represents the edge threshold, and ET is, for example, 10. The image after extracting SIFT feature points is as follows Figure 3 shown.

[0090] Then, in S12, ellipse fitting is performed on multiple key points to construct a minimum circumscribed ellipse. This ellipse is centered on the data average value, and its size and tilt angle accurately reflect the overall distribution of the data and can cover all the feature points therein, such as Figure 4 As shown. Assign the image area corresponding to the minimum circumscribed ellipse to white, and the rest of the image area to black to generate a preliminary mask feature map of the region of interest, as shown Figure 5 shown.

[0091] Then, in S13, edge detection and land-sea segmentation line extraction are performed on the airborne grayscale image; the sea area below the land-sea segmentation line is assigned white, and the upper area is assigned black to generate a land-sea segmentation mask map; wherein, the Canny edge detection algorithm is used to extract edge features of the original image, such as Figure 6 As shown; select the longest edge as the sea-land dividing line, as Figure 7 As shown; the generated sea and land segmentation mask is as follows Figure 8 shown.

[0092] Then, in S14, the preliminary interest region mask feature map and the land and sea segmentation mask map are multiplied pixel by pixel to obtain the interest region image, such as Figure 9 shown.

[0093] Then, S2 is executed. In S2, an optimal segmentation threshold is obtained by using an improved Harris Hawk optimization algorithm, and the airborne grayscale image is segmented by using the optimal segmentation threshold to obtain a segmented image based on the optimal segmentation threshold.

[0094] According to an embodiment of the present invention, the process of the existing Harris Hawks Optimization (HHO) algorithm includes:

[0095] 1) Population initialization: In the initialization phase, N initial solutions are randomly generated in the search space. (The position of the eagle), each solution represents a possible threshold candidate. This uniform random initialization ensures population diversity, and the calculation formula is:

[0096]

[0097] Among them, lb and ub are the lower bound and upper bound of the search space respectively; is a uniformly distributed random number in [0,1].

[0098] 2) Exploration Phase: When prey is energetic, the hawk flock adopts two random exploration strategies: collaborative surveillance, based on random individual position updates (q ≥ 0.5); and random exploration, based on the population average position and random vector updates (q < 0.5). This phase encourages global search to avoid premature convergence. Individual position updates use a formula based on the positions of randomly selected individuals in the population. The selected individuals are screened based on image grayscale features to ensure that the position update process is indirectly guided by grayscale information, thereby facilitating the search direction and step size to areas with better grayscale feature performance. The calculation formula is:

[0099]

[0100] in, is the position of the i-th individual in the eagle group at the t+1th generation, is the position of the eagle individual randomly selected in the tth generation; is the current optimal solution (prey position); is the average position of the population; r1, r2, r3, r4, q are random numbers in [0, 1].

[0101] 3) Transition from exploration to exploitation: During this period, while the eagles are catching prey, the prey will consume energy in the process of escaping. In the energy model, energy , ∈[-1,1] simulates the escape ability of the prey (optimal solution). |E| ≥ 1: the prey is energetic and the hawks are in the exploration phase (global search); |E| < 1: the prey is exhausted and the hawks enter the exploitation phase (local fine search). Energy decreases linearly with iteration, ensuring that the algorithm focuses on exploration in the early stages and exploitation in the later stages. The energy model is:

[0102]

[0103] Where, is the initial energy, ; t is the current iteration number; T is the maximum iteration number; rand() is a uniform random number on [0,1].

[0104] 4) Development phase: Based on the prey escape energy E and the random number r (r∈[0,1]), the Harris Hawk optimization algorithm adopts four hunting strategies:

[0105] 41) Strategy 1 - Soft Siege: When 0.5≤|E|<1 and r≥0.5, the prey has enough energy to escape by jumping, and the hawks consume its energy through soft siege:

[0106]

[0107] in, is the distance difference between the individual and the prey (optimal solution), h = 2(1-r5) (r5∈[0,1] is a random number), which is used to control the jumping intensity of the prey behavior.

[0108] 42) Strategy 2 - Hard Siege: When |E| < 0.5 and r ≥ 0.5, the prey is exhausted and the hawks quickly pounce:

[0109]

[0110] 43) Strategy 3 - Soft Encirclement with Asymptotically Fast Dives: When 0.5≤|E|<1 and r<0.5, the hawks perform random walks using Lévy flight to expand the search range:

[0111]

[0112] in, , ; Jump amplitude J = 1.8(1-rand()); S is a D-dimensional random vector (element ∈ [0,1]), is the Levy flight function; is the fitness of the current individual i in the tth generation (i.e., the objective function value of the current solution).

[0113] In the embodiment of the present invention, the objective function, i.e., the fitness function, is an improvement of the weighted inter-class variance calculated based on the image grayscale histogram, and the expression is:

[0114]

[0115] Where, is the fitness function value; is the weight of class 0 (the ratio of pixels below the threshold), is the weight of class 1 (the proportion of pixels above the threshold); is the class 0 average degree, ; reward represents the threshold reward coefficient, which is used to adjust the fitness weight according to the threshold size; T represents the control segmentation threshold; B represents the grayscale range; p(i) represents the probability distribution of grayscale i in the image.

[0116] The position update is guided by the currently determined objective function value, strengthening the grayscale direction of the local search and achieving rapid convergence of the population to the threshold solution with the best grayscale segmentation effect.

[0117] 44) Strategy 4 - Hard Encirclement with Gradual Rapid Dive: When |E| < 0.5 and r < 0.5, the hawks build a hard siege to capture prey:

[0118]

[0119] in, .

[0120] 5) Boundary processing: truncate out-of-bounds solutions to ensure that all solutions are within the feasible region; the calculation formula is:

[0121]

[0122] The embodiment of the present invention makes the following improvements to the traditional Harris Hawk optimization algorithm:

[0123] a) In the exploration phase, in order to expand the exploration range, simplify calculations, focus on the effective area, and adjust parameters, the population individual position update formula is modified to:

[0124]

[0125] Among them, X k is the position of a random individual selected from the population; r,q~U(0,1) are uniform random numbers, r is used to control the disturbance amplitude during the search process; q is used to determine whether to execute a specific strategy.

[0126] b) In the transition phase from exploration to development, to reserve more exploration space and ensure rapid entry into the development phase when t approaches T, the escape energy model is modified as follows:

[0127]

[0128] in, is the nonlinear decay exponent; is the decay curve index; is the disturbance intensity coefficient; is the standard normally distributed noise.

[0129] c) In the development phase, to strengthen local search and reduce invalid perturbations, the calculation formula is selected based on development efficiency instead of choosing from four strategies. The position update formula in the four strategy selections in the original algorithm is improved to:

[0130]

[0131] The above formula increases development efficiency from 50% to 85%; For improved Levi flight items.

[0132] Lévy flight generates random steps alternating between long leaps and short wandering steps, simulating the swooping pattern of a flock of hawks. This heavy-tailed distribution enhances the algorithm's ability to escape local optima. To avoid late-stage oscillation and optimize step-length variance, the present invention further optimizes the Lévy flight formula. The improved Lévy flight term is:

[0133]

[0134] Wherein, k is the step size coefficient (0.004 in the embodiment of the present invention); is the Levy index (1.48 in the embodiment of the present invention); λ is the iterative decay exponent (0.68 in the embodiment of the present invention); is a standard normal random variable, is the scaling factor.

[0135] d) In boundary processing, in order to preserve gradient information and improve convergence speed, elastic rebound is used instead of innovative interception. The improved calculation formula is:

[0136]

[0137] The improved Harris Eagle optimization algorithm is used to obtain the optimal segmentation threshold, and the airborne grayscale image is segmented using the obtained optimal segmentation threshold to obtain a segmented image based on the optimal segmentation threshold, such as Figure 10 shown.

[0138] Then, step S3 is executed, in which the image of the region of interest and the segmented image based on the optimal segmentation threshold are fused to obtain a final segmented image.

[0139] According to an embodiment of the present invention, Figure 10 and Figure 9 Perform pixel-by-pixel multiplication and remove spots to obtain the final ship segmentation result, as shown in the following example: Figure 11 shown.

[0140] The embodiment of the present invention further proposes a ship target segmentation system based on an improved Harris Hawk optimization algorithm, the system comprising:

[0141] An area of ​​interest acquisition module is configured to process an airborne grayscale image including a sky area, a sea area, a ship at sea, and a coastline to acquire an area of ​​interest image;

[0142] an optimal segmentation image acquisition module configured to obtain an optimal segmentation threshold using an improved Harris Hawk optimization algorithm, and segment the airborne grayscale image using the optimal segmentation threshold to obtain a segmented image based on the optimal segmentation threshold;

[0143] The ship target segmentation module is configured to fuse the region of interest image with the segmentation image based on the optimal segmentation threshold to obtain a final segmentation image.

[0144] In this embodiment, preferably, the processing of the airborne grayscale image including the sky area, the sea area, the ships at sea, and the coastline in the region of interest acquisition module to acquire the region of interest image includes:

[0145] Extracting multiple key points of the airborne grayscale image using a SIFT algorithm; including: constructing a Gaussian difference pyramid and detecting local extreme points; filtering the multiple local extreme points based on contrast and edge response to obtain multiple key points;

[0146] Perform ellipse fitting on multiple key points to obtain the minimum circumscribed ellipse containing all key points;

[0147] Assign the image area corresponding to the minimum circumscribed ellipse to white, and the rest of the image area to black to generate a preliminary mask feature map of the region of interest;

[0148] Performing edge detection and land-sea segmentation line extraction on the airborne grayscale image; assigning the sea area below the land-sea segmentation line to white and the area above the land-sea segmentation line to black, thereby generating a land-sea segmentation mask map;

[0149] The preliminary region of interest mask feature map and the land and sea segmentation mask map are multiplied pixel by pixel to obtain the region of interest image.

[0150] In this embodiment, preferably, the step of obtaining the optimal threshold value by using the improved Harris Hawk optimization algorithm in the optimal segmentation image acquisition module includes:

[0151] In the exploration phase, the position update formula of the population individuals is:

[0152] ;

[0153] Where, represents the position of the i-th individual in the t+1-th iteration; Represents the position of a randomly selected individual in the population; r and q are uniform random numbers between [0,1]; is a uniformly distributed random number in [0,1]; lb and ub are the lower and upper bounds of the search space respectively;

[0154] In the transition phase from exploration to exploitation, the hawks catch prey, and the energy loss model of the prey consuming energy in the process of escaping is:

[0155] ;

[0156] Where, represents the nonlinear attenuation exponent; represents the initial escape energy of the prey; is the decay curve index; is the disturbance intensity coefficient; is the standard normal distribution noise;

[0157] In the development phase, the position update formula of the population individuals is:

[0158] ;

[0159] Where, represents the optimal prey position in the tth iteration; It indicates the energy of prey escape; Indicates the jump amplitude; represents the improved Levy flight term, , is the step size coefficient, is a standard normal random variable, is the Levy index, is the scaling factor, and λ is the iterative decay exponent.

[0160] In this embodiment, preferably, the step of obtaining the optimal threshold using the improved Harris Hawk optimization algorithm in the optimal segmentation image acquisition module further includes: after each iteration of each stage, truncating the out-of-bounds solution using the following formula:

[0161] .

[0162] The functions of the ship target segmentation system based on the improved Harris Hawk optimization algorithm described in the embodiment of the present invention can be explained by the aforementioned ship target segmentation method based on the improved Harris Hawk optimization algorithm. Therefore, for the parts not described in detail in the system embodiment, please refer to the above method embodiment and will not be repeated here.

[0163] It should be noted that although several units, modules, or submodules are mentioned in the detailed description above, such division is merely exemplary and not mandatory. In fact, according to embodiments of the present invention, the features and functions of two or more modules described above may be embodied in one module. Conversely, the features and functions of one module described above may be further divided and embodied by multiple modules.

[0164] Furthermore, although the operations of the method of the present invention are described in a particular order in the accompanying drawings, this does not require or imply that these operations must be performed in this particular order, or that all illustrated operations must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps.

[0165] Although the spirit and principles of the present invention have been described with reference to several specific embodiments, it should be understood that the present invention is not limited to the specific embodiments disclosed, and the division into various aspects does not mean that the features of these aspects cannot be combined to benefit. Such division is only for the convenience of expression. The present invention is intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.

Claims

1. A ship target segmentation method based on the improved Harris Hawk optimization algorithm is characterized by: include: Process the airborne grayscale image containing the sky, sea, ships and coastline to obtain the image of the area of ​​interest; Using an improved Harris Hawk optimization algorithm to obtain an optimal segmentation threshold, and using the optimal segmentation threshold to segment the airborne grayscale image to obtain a segmented image based on the optimal segmentation threshold; The region of interest image and the segmentation image based on the optimal segmentation threshold are fused to obtain a final segmentation image.

2. The ship target segmentation method based on the improved Harris Hawk optimization algorithm according to claim 1 is characterized in that: The processing of the airborne grayscale image including the sky area, the sea area, the ships at sea and the coastline to obtain the image of the area of ​​interest includes: Extracting multiple key points from the airborne grayscale image using a SIFT algorithm, including: constructing a Gaussian difference pyramid and detecting local extreme points; filtering the multiple local extreme points based on contrast and edge response to obtain multiple key points; Perform ellipse fitting on multiple key points to obtain the minimum circumscribed ellipse containing all key points; Assign the image area corresponding to the minimum circumscribed ellipse to white, and the rest of the image area to black to generate a preliminary mask feature map of the region of interest; Performing edge detection and land-sea segmentation line extraction on the airborne grayscale image; assigning the sea area below the land-sea segmentation line to white and the area above the land-sea segmentation line to black, thereby generating a land-sea segmentation mask map; The preliminary region of interest mask feature map and the land and sea segmentation mask map are multiplied pixel by pixel to obtain the region of interest image.

3. The ship target segmentation method based on the improved Harris Hawk optimization algorithm according to claim 2 is characterized in that: The method of obtaining the optimal segmentation threshold by using the improved Harris Hawk optimization algorithm includes: In the exploration phase, the position update formula of the population individuals is: ； Where, represents the position of the i-th individual in the t+1-th iteration; Represents the position of a randomly selected individual in the population; r and q are uniform random numbers between [0,1]; is a uniformly distributed random number in [0,1]; lb and ub are the lower and upper bounds of the search space respectively; In the transition phase from exploration to exploitation, the hawks catch prey, and the energy loss model of the prey consuming energy in the process of escaping is: ； Where, represents the nonlinear attenuation exponent; represents the initial escape energy of the prey; represents the decay curve index; represents the disturbance intensity coefficient; is the standard normal distribution noise; t represents the current number of iterations; T represents the maximum number of iterations; In the development phase, the position update formula of the population individuals is: ； Where, represents the optimal prey position in the tth iteration; It indicates the energy of prey escape; Indicates the jump amplitude; represents the improved Levy flight term, , is the step size coefficient, is a standard normal random variable, is the Levy index, is the scaling factor, and λ is the iterative decay exponent.

4. The ship target segmentation method based on the improved Harris Hawk optimization algorithm according to claim 3 is characterized in that: The method of obtaining the optimal segmentation threshold using the improved Harris Hawk optimization algorithm further includes: after each iteration of each stage, truncating the out-of-bounds solution using the following formula: 。 5. The ship target segmentation system based on the improved Harris Hawk optimization algorithm is characterized by: include: An area of ​​interest acquisition module is configured to process an airborne grayscale image including a sky area, a sea area, a ship at sea, and a coastline to acquire an area of ​​interest image; an optimal segmentation image acquisition module configured to obtain an optimal segmentation threshold using an improved Harris Hawk optimization algorithm, and segment the airborne grayscale image using the optimal segmentation threshold to obtain a segmented image based on the optimal segmentation threshold; The ship target segmentation module is configured to fuse the region of interest image with the segmentation image based on the optimal segmentation threshold to obtain a final segmentation image.

6. The ship target segmentation system based on the improved Harris Hawk optimization algorithm according to claim 5 is characterized in that: The region of interest acquisition module processes the airborne grayscale image including the sky area, the sea area, the ships at sea and the coastline to obtain the region of interest image, including: Extracting multiple key points of the airborne grayscale image using a SIFT algorithm; including: constructing a Gaussian difference pyramid and detecting local extreme points; filtering the multiple local extreme points based on contrast and edge response to obtain multiple key points; Perform ellipse fitting on multiple key points to obtain the minimum circumscribed ellipse containing all key points; Assign the image area corresponding to the minimum circumscribed ellipse to white, and the rest of the image area to black to generate a preliminary mask feature map of the region of interest; Performing edge detection and land-sea segmentation line extraction on the airborne grayscale image; assigning the sea area below the land-sea segmentation line to white and the area above the land-sea segmentation line to black, thereby generating a land-sea segmentation mask map; The preliminary region of interest mask feature map and the land and sea segmentation mask map are multiplied pixel by pixel to obtain the region of interest image.

7. The ship target segmentation system based on the improved Harris Hawk optimization algorithm according to claim 6 is characterized in that: The optimal segmentation image acquisition module uses the improved Harris Hawk optimization algorithm to obtain the optimal segmentation threshold, which includes: In the exploration phase, the position update formula of the population individuals is: ； Where, represents the position of the i-th individual in the t+1-th iteration; Represents the position of a randomly selected individual in the population; r and q are uniform random numbers between [0,1]; is a uniformly distributed random number in [0,1]; lb and ub are the lower and upper bounds of the search space respectively; In the transition phase from exploration to exploitation, the hawks catch prey, and the energy loss model of the prey consuming energy in the process of escaping is: ； Where, represents the nonlinear attenuation exponent; represents the initial escape energy of the prey; represents the decay curve index; represents the disturbance intensity coefficient; is the standard normal distribution noise; t represents the current number of iterations; T represents the maximum number of iterations; In the development phase, the position update formula of the population individuals is: ； Where, represents the optimal prey position in the tth iteration; It indicates the energy of prey escape; Indicates the jump amplitude; represents the improved Levy flight term, , is the step size coefficient, is a standard normal random variable, is the Levy index, is the scaling factor, and λ is the iterative decay exponent.

8. The ship target segmentation system based on the improved Harris Hawk optimization algorithm according to claim 7 is characterized in that: The method of obtaining the optimal segmentation threshold using the improved Harris Hawk optimization algorithm in the optimal segmentation image acquisition module further includes: after each iteration of each stage, truncating the out-of-bounds solution using the following formula: 。

Images