CN113643315A - Infrared small target detection method based on self-adaptive peak gradient descent filter - Google Patents

Infrared small target detection method based on self-adaptive peak gradient descent filter Download PDF

Info

Publication number
CN113643315A
CN113643315A CN202110830356.8A CN202110830356A CN113643315A CN 113643315 A CN113643315 A CN 113643315A CN 202110830356 A CN202110830356 A CN 202110830356A CN 113643315 A CN113643315 A CN 113643315A
Authority
CN
China
Prior art keywords
gradient
peak
infrared
infrared image
small target
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Granted
Application number
CN202110830356.8A
Other languages
Chinese (zh)
Other versions
CN113643315B (en
Inventor
黄珺
马泳
樊凡
梅晓光
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Wuhan University WHU
Original Assignee
Wuhan University WHU
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Wuhan University WHU filed Critical Wuhan University WHU
Priority to CN202110830356.8A priority Critical patent/CN113643315B/en
Publication of CN113643315A publication Critical patent/CN113643315A/en
Application granted granted Critical
Publication of CN113643315B publication Critical patent/CN113643315B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/10—Segmentation; Edge detection
    • G06T7/136—Segmentation; Edge detection involving thresholding
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T5/00—Image enhancement or restoration
    • G06T5/70—Denoising; Smoothing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10048—Infrared image

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Photometry And Measurement Of Optical Pulse Characteristics (AREA)

Abstract

The invention discloses an infrared small target detection method based on a self-adaptive peak gradient descent filter, which solves the problems that irregular clutter cannot be inhibited and the detection performance of a small target is low due to the fact that a rectangular window is used for extracting local features of an infrared image in the existing infrared small target detection method. The method fully utilizes the characteristic that the gradient of the gray value of a local pixel in the infrared image is reduced and the characteristic that the local contrast difference exists between the infrared dim target and the background, can inhibit irregular clutter, does not need to set a rectangular window according to the size of the target, is beneficial to detecting the target with less than 9 multiplied by 9 pixels, improves the performance of detecting the infrared dim target and achieves the effect of early detection.

Description

Infrared small target detection method based on self-adaptive peak gradient descent filter
Technical Field
The invention belongs to the technical field of image target detection, and particularly relates to an infrared small target detection method based on a self-adaptive peak gradient descent filter.
Background
The infrared small target detection plays a key role in an infrared early warning system. The target appears as a point with little structural information and shape features in the infrared image, so that the target cannot be segmented and extracted with detailed features. In a real infrared image, small targets are easily swamped by noise and clutter due to the performance of the infrared focal plane and the environment in which the target is located. Therefore, how to detect weak infrared targets from infrared images full of noise and complex clutter without generating false alarms remains a challenging problem.
In the last decade, researchers have proposed single-frame detection methods and multi-frame detection methods that utilize prior information of infrared images. The multi-frame detection method mainly consumes a large amount of computing resources and has low efficiency; the single frame detection method cannot suppress various complex background clutter. Based on the characteristic of isotropy of the small target, the target of the infrared image is enhanced by methods such as local gradient strength and multi-scale flux density gradient direction diversity. This approach still has limited ability to suppress edge protrusions or partial clutter; the low rank and sparsity based approach is difficult to suppress small scale strong local interference and is very time consuming, requiring GPU acceleration to meet real-time requirements for surveillance and other scenarios.
The detection method based on the Human Visual System (HVS) has limited inhibition capability and detection capability on irregular clutter; HVS-based methods typically use rectangular windows of size 3 x 3 to extract local features of the infrared image. Since the rectangular window used in these methods may contain both irregular clutter and background, the irregular clutter cannot be effectively suppressed. When the window unit contains both clutter and objects smaller than the cell size, the detection performance of the weak and small objects is reduced.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides an infrared small target detection method based on a self-adaptive peak gradient descent filter. The method provided by the invention can clearly determine the target pixel and the background pixel, and breaks through the limitation of the rectangular window in clutter suppression and detection of the target smaller than 3 multiplied by 3. In order to achieve the purpose, the invention provides the technical scheme that: based on the gradient descent criterion of the infrared image, an infrared small target detection filter is designed by utilizing the gradual reduction of the gray difference value of the target and the background from inside to outside, so that irregular clutter is inhibited. The invention has the advantages that the rectangular window is not required to be set according to the size of the target, the target with less than 9 multiplied by 9 pixels can be detected, and the performance of detecting the infrared weak and small target is improved.
The technical scheme of the invention provides an infrared small target detection method based on a self-adaptive peak gradient descent filter, which comprises the following steps:
step 1: establishing a sliding window to traverse the infrared image to obtain a plurality of local gray peak points of the infrared image;
step 2: taking the local gray peak point as a center, and taking the gray value of the local gray peak point as a new gradient path if the gray value of the local gray peak point is greater than or equal to the gray value of the adjacent pixel point in each gradient direction; otherwise, the gradient path in that direction is ended. Performing multiple iterations on the gradient paths of all local gray peak points in each gradient direction to obtain a final gradient path;
and step 3: and establishing a peak gradient descent filter, extracting the small target to be determined through a self-adaptive threshold, wherein the extracted result is the real small target.
Further, step 1 comprises the following steps:
step 1.1: acquiring an original infrared image R to be processed, wherein the size of the original infrared image R is M multiplied by N, the gray value of each pixel of the image R is represented as R (x, y), x belongs to [1,2, …, M ], y belongs to [1,2, …, N ];
step 1.2: a sliding window W is established with a pixel size l x j, a window sliding step size s and a window sliding number q. And traversing the infrared image R through the sliding window W to obtain a plurality of local gray peak pixel points.
Further, step 2 comprises the following steps:
step 2.1: establishing a gradient filter Pi nN is iteration number, i is gradient direction, g is local gray peak pixel point, gi nThe pixel point after the nth iteration of the gradient direction i in the filter is represented mathematically as follows:
Figure BDA0003175299810000021
wherein the angle of each gradient direction is i x 45 °;
step 2.2: the next pixel point g along the gradient direction ii n+1Is less than or equal to gi nGray value of gi n +1A new path in the gradient direction i; when g isi n+1Is greater than gi nThe path in the gradient direction i terminates. And (3) iteratively calculating the final gradient path of the local gray peak pixel point in each gradient direction through n steps, wherein the mathematical expression is as follows:
Figure BDA0003175299810000022
resulting in a final gradient path of g in 8 gradient directions, denoted as P1,P2,P3,P4,P5,P6,P7,P8；
Step 2.3: calculating the final gradient path P of all local gray peak points in the infrared image R through the steps 2.1 and 2.2i。
Further, step 3 comprises the following steps:
step 3.1: designing a peak gradient descent filter on the basis of the existing gradient descent path: for final path P of each gradient direction of local gray peak pixel pointiCalculating local gray peak pixel points g and PiThe gray difference d between the minimum gray valuesiDifference in gray level diThe calculation formula of (2) is as follows:
di＝g-min(Pi),i＝1,2,3,4,5,6,7,8
d in diagonal directioniAnd di+4The values are multiplied to improve the saliency of the target and suppress clutter. The calculation formula is as follows:
Dj＝dj*dj+4,j＝1,2,3,4
get DjIs taken as a model of the gradient descent of the peak at point (x, y)And is assigned to R (x, y), which is calculated as follows:
R(x,y)＝min(Dj)
step 3.2: and (3) obtaining R (x, y) values of all local gray peak points through the step of 3.1, and setting the values of the rest pixel points in the infrared image to be 0.
Step 3.3: traversing all R (x, y) in the infrared image by adopting the adaptive threshold value T, comparing the R (x, y) with the R (x, y) value, and if the R (x, y) value is larger than or equal to the adaptive threshold value T, considering the R (x, y) as a real small target in the infrared image R. The adaptive threshold T is calculated as follows:
T＝λRmax+(1-λ)Rmean
wherein R ismaxIs the maximum value of R (x, y) in the infrared image R, RmeanIs the average value of R (x, y) in the infrared image R, and is lambada epsilon (0, 1).
Compared with the prior art, the invention has the following advantages and beneficial effects:
(1) the invention designs a peak gradient descent filter to divide the maximum value of local gray scale, and can effectively improve the detection rate of infrared dim targets.
(2) The invention designs a segmentation method based on the self-adaptive threshold, can effectively inhibit background clutter and noise, and reduces the false alarm rate of infrared weak and small target detection.
Drawings
FIG. 1 is a general flow diagram of the present invention;
FIG. 2 is a schematic diagram showing the gradient direction of the local gray peak in step 2 according to the present invention;
FIG. 3 is a schematic diagram of an adaptive peaking gradient filter according to the present invention.
Detailed Description
The technical solution of the present invention is further explained with reference to the drawings and the embodiments.
In order to more clearly illustrate the objects, technical solutions and advantages of the present invention, the following description is further provided with reference to the accompanying drawings and examples. It is to be understood that the invention is not to be limited by the disclosure of the embodiments, but is to be controlled by the scope of the appended claims.
The technical problem is as follows: aiming at the problems of large operation amount and insufficient effect of inhibiting irregular clutter in the existing method, the invention provides an infrared small target detection method based on a self-adaptive peak gradient descent filter, a rectangular window is not required to be arranged according to the size of a target, the gradient is gradually reduced from inside to outside by utilizing the gray difference value of the target and a background, the irregular clutter is inhibited, and the small target of an infrared image is extracted at a pixel level.
The technical means is as follows:
an infrared small target detection method based on a self-adaptive peak gradient descent filter comprises the following steps:
step 1: establishing a sliding window to traverse the infrared image to obtain a plurality of local gray peak points of the infrared image;
step 2: taking the local gray peak point as a center, and taking the gray value of the local gray peak point as a new gradient path if the gray value of the local gray peak point is greater than or equal to the gray value of the adjacent pixel point in each gradient direction; otherwise, the gradient path in that direction is ended. Performing multiple iterations on the gradient paths of all local gray peak points in each gradient direction to obtain a final gradient path;
and step 3: and establishing a peak gradient descent filter, extracting the small target to be determined through a self-adaptive threshold, wherein the extracted result is the real small target.
The step 1 comprises the following steps:
step 1.1: acquiring an original infrared image R to be processed, wherein the size of the original infrared image R is 640 multiplied by 480, the gray value of each pixel of the image R is represented as R (x, y), x belongs to [1,2, …,640], y belongs to [1,2, …,480 ];
step 1.2: a sliding window W is established with a pixel size of 80 x 60, a window sliding step of 8 and a window sliding number of times 64. And traversing the infrared image R through the sliding window W to obtain a plurality of local gray peak pixel points.
The step 2 comprises the following steps:
step 2.1: establishing a gradient filter Pi nN is iteration number, i is gradient direction, g is local gray peak pixel point, gi nThe pixel point after the nth iteration of the gradient direction i in the filter is represented mathematically as follows:
Figure BDA0003175299810000041
wherein the angle of each gradient direction is i x 45 °;
step 2.2: the next pixel point g along the gradient direction ii n+1Is less than or equal to gi nGray value of gi n +1A new path in the gradient direction i; when g isi n+1Is greater than gi nThe path in the gradient direction i terminates. And (3) iteratively calculating the final gradient path of the local gray peak pixel point in each gradient direction through n steps, wherein the mathematical expression is as follows:
Figure BDA0003175299810000042
resulting in a final gradient path of g in 8 gradient directions, denoted as P1,P2,P3,P4,P5,P6,P7,P8。
Where n is equal to or less than 5, i.e. the maximum number of iterations for each gradient direction is 5.
Step 2.3: calculating the final gradient path P of all local gray peak points in the infrared image R through the steps 2.1 and 2.2i。
The step 3 comprises the following steps:
step 3.1: based on the existing gradient descent path, a peak gradient descent filter is designed. For final path P of each gradient direction of local gray peak pixel pointiWe calculate local gray peak pixel points g and PiThe gray difference d between the minimum gray valuesiDifference in gray level diThe calculation formula of (2) is as follows:
di＝g-min(Pi),i＝1,2,3,4,5,6,7,8
d in diagonal directioniAnd di+4The values are multiplied to improve the saliency of the target and suppress clutter. The calculation formula is as follows:
Dj＝dj*dj+4,j＝1,2,3,4
get DjThe minimum value of (d) is taken as the result of the model of the decrease in peak gradient at point (x, y) and is assigned to R (x, y), which is calculated as follows:
R(x,y)＝min(Dj)
step 3.2: and (3) obtaining R (x, y) values of all local gray peak points through the step of 3.1, and setting the values of the rest pixel points in the infrared image to be 0.
Step 3.3: traversing all R (x, y) in the infrared image by adopting the adaptive threshold value T, comparing the R (x, y) with the R (x, y) value, and if the R (x, y) value is larger than or equal to the adaptive threshold value T, considering the R (x, y) as a real small target in the infrared image R. The adaptive threshold T is calculated as follows:
T＝λRmax+(1-λ)Rmean
wherein R ismaxIs the maximum value of R (x, y) in the infrared image R, RmeanIs the average value of R (x, y) in the infrared image R, and is lambada epsilon (0.6, 0.9).
The effect analysis was performed according to fig. 3: fig. 3 shows the response of the peak gradient descent filter to small targets and various types of clutter, with the infrared small targets in box 1, background clutter in box 2, highlight edges in box 3, and bright clutter in box 4 on the left. To the right are the gradient paths of the 4 local gray peak points of boxes 1-4. There is a large gray scale difference between the small object in box 1 and the neighborhood background, resulting in a large filter response. The gray scale difference between the peak value and the end point of the 8 paths is large, namely the D value is large, so that the value of R (x, y) reaches 19630; in block 2, the filter response is small, where the value of D in the up-down direction is the smallest, and the value of R (x, y) finally calculated is small, 98; in block 3 and block 4, although there is a large gray scale difference in the partial gradient descent path, there is a small gray scale difference in other directions, resulting in a weak filter response for this region, and the final R (x, y) values for the two regions are 135 and 54, respectively. Therefore, the R (x, y) values of various background clutters are far smaller than the R value of the target clutters, and the method provided by the invention can effectively enhance the target and inhibit the clutters.
The above description is only a preferred embodiment of the invention and should not be taken as limiting the invention, and any modification, equivalent replacement, or improvement made within the spirit and principle of the invention should be included in the protection scope of the invention.

Claims (6)

1. An infrared small target detection method based on an adaptive peak gradient descent filter is characterized by comprising the following steps:
step 1, traversing an infrared image R by adopting a sliding window, and constructing local gray peak pixel points in the infrared image R;
step 2, taking the local gray peak pixel point as a center, and taking the next pixel point as a new gradient path if the gray value of the local gray peak pixel point is greater than or equal to the gray value of the adjacent pixel point in each gradient direction; otherwise, ending the gradient path in the direction; performing multiple iterations on the gradient paths of all local gray peak pixel points in all gradient directions to obtain a final gradient path;
and step 3: and establishing a peak gradient descent filter, extracting the small target to be determined through a self-adaptive threshold, wherein the extracted result is the real small target.
2. The infrared small target detection method based on the adaptive peak gradient descent filter according to claim 1, characterized in that: the step 1 comprises the following steps:
step 1.1, obtaining an original infrared image R to be processed, which has a size of M × N, where a grayscale value of a pixel of the image R is represented as R (x, y), x is 1,2, …, M, y is 1,2, …, N, that is, R (x, y) represents a grayscale value of the original infrared image R at (x, y);
step 1.2, establishing a sliding window W, wherein the pixel size is l multiplied by j, the window sliding step length is s, the window sliding times q are obtained, and the sliding window W traverses the infrared image R to obtain a plurality of local gray peak pixel points.
3. The infrared small target detection method based on the adaptive peak gradient filter as claimed in claim 2, characterized in that: the step 2 comprises the following steps;
step 2.1, establish gradient filter Pi nN is iteration number, i is gradient direction, g is local gray peak pixel point, gi nThe pixel point after the nth iteration of the gradient direction i in the filter is represented mathematically as follows:
Figure FDA0003175299800000011
wherein the angle of each gradient direction is i x 45 °;
step 2.2: the next pixel point g along the gradient direction ii n+1Is less than or equal to gi nGray value of gi n+1A new path in the gradient direction i; when g isi n+1Is greater than gi nThe path in the gradient direction i terminates. And (3) iteratively calculating the final gradient path of the local gray peak pixel point in each gradient direction through n steps, wherein the mathematical expression is as follows:
Figure FDA0003175299800000012
resulting in a final gradient path of g in 8 gradient directions, denoted as P1,P2,P3,P4,P5,P6,P7,P8；
Step 2.3, calculating the final gradient path P of all local gray peak pixel points in the infrared image R through the steps 2.1 and 2.2i。
4. The infrared small target detection method based on the adaptive peak gradient descent filter according to claim 3, characterized in that: the step 3 comprises the following steps;
step 3.1, designing a peak gradient descent filter on the basis of the existing gradient descent path: for final path P of each gradient direction of local gray peak pixel pointiCalculating local gray peak pixel points g and PiThe gray difference d between the minimum gray valuesiDifference in gray level diThe calculation formula of (2) is as follows:
di＝g-min(Pi),i＝1,2,3,4,5,6,7,8
d in diagonal directioniAnd di+4Multiplying values to improve the significance of the target and suppress clutter, wherein the calculation formula is as follows:
Dj＝dj*dj+4,j＝1,2,3,4
get DjIs taken as the result of the peak gradient descent filter at point (x, y) and is assigned to R (x, y) by the following calculation:
R(x,y)＝min(Dj)
step 3.2, obtaining R (x, y) values of all local gray peak pixel points through the step 3.1, and setting the values of the rest pixel points in the infrared image to be 0;
step 3.3, traversing all R (x, y) in the infrared image by adopting the adaptive threshold T, comparing the values with the R (x, y), and if the value of R (x, y) is greater than or equal to the adaptive threshold T, considering the R (x, y) as a real small target in the infrared image R, wherein the calculation formula of the adaptive threshold T is as follows:
T＝λRmax+(1-λ)Rmean
wherein R ismaxIs the maximum value of R (x, y) in the infrared image R, RmeanIs the average value of R (x, y) in the infrared image R, and is lambada epsilon (0, 1).
5. The infrared small target detection method based on the adaptive peak gradient descent filter as claimed in claim 2, characterized in that: the pixel size of the sliding window W in step 1.2 is 80 × 60, the window sliding step size is 8, and the window sliding times are 64.
6. The infrared small target detection method based on the adaptive peak gradient descent filter as claimed in claim 2, characterized in that: in step 2.2, n is less than or equal to 5, i.e. the maximum number of iterations for each gradient direction is 5.
CN202110830356.8A 2021-07-22 2021-07-22 Infrared small target detection method based on self-adaptive peak gradient descent filter Active CN113643315B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202110830356.8A CN113643315B (en) 2021-07-22 2021-07-22 Infrared small target detection method based on self-adaptive peak gradient descent filter

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202110830356.8A CN113643315B (en) 2021-07-22 2021-07-22 Infrared small target detection method based on self-adaptive peak gradient descent filter

Publications (2)

Publication Number Publication Date
CN113643315A true CN113643315A (en) 2021-11-12
CN113643315B CN113643315B (en) 2023-08-15

Family

ID=78418064

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202110830356.8A Active CN113643315B (en) 2021-07-22 2021-07-22 Infrared small target detection method based on self-adaptive peak gradient descent filter

Country Status (1)

Country Link
CN (1) CN113643315B (en)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106251344A (en) * 2016-07-26 2016-12-21 北京理工大学 A kind of multiple dimensioned infrared target self-adapting detecting method of view-based access control model receptive field
CN107392885A (en) * 2017-06-08 2017-11-24 江苏科技大学 A kind of method for detecting infrared puniness target of view-based access control model contrast mechanism
CN108898573A (en) * 2018-04-23 2018-11-27 西安电子科技大学 Infrared small target rapid extracting method based on multi-direction annular gradient method
CN109272489A (en) * 2018-08-21 2019-01-25 西安电子科技大学 Inhibit the method for detecting infrared puniness target with multiple dimensioned local entropy based on background
CN109767439A (en) * 2019-01-10 2019-05-17 中国科学院上海技术物理研究所 A kind of multiple dimensioned difference of self-adapting window and the object detection method of bilateral filtering
CN111967466A (en) * 2020-07-08 2020-11-20 北京航空航天大学 Infrared weak and small target detection method utilizing multidirectional sparse low-rank decomposition

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106251344A (en) * 2016-07-26 2016-12-21 北京理工大学 A kind of multiple dimensioned infrared target self-adapting detecting method of view-based access control model receptive field
CN107392885A (en) * 2017-06-08 2017-11-24 江苏科技大学 A kind of method for detecting infrared puniness target of view-based access control model contrast mechanism
CN108898573A (en) * 2018-04-23 2018-11-27 西安电子科技大学 Infrared small target rapid extracting method based on multi-direction annular gradient method
CN109272489A (en) * 2018-08-21 2019-01-25 西安电子科技大学 Inhibit the method for detecting infrared puniness target with multiple dimensioned local entropy based on background
CN109767439A (en) * 2019-01-10 2019-05-17 中国科学院上海技术物理研究所 A kind of multiple dimensioned difference of self-adapting window and the object detection method of bilateral filtering
CN111967466A (en) * 2020-07-08 2020-11-20 北京航空航天大学 Infrared weak and small target detection method utilizing multidirectional sparse low-rank decomposition

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
危水根;王程伟;张聪炫;鄢慧斌: "多信息融合的红外弱小目标检测", 红外技术, vol. 41, no. 009 *

Also Published As

Publication number Publication date
CN113643315B (en) 2023-08-15

Similar Documents

Publication Publication Date Title
Xuan et al. An improved canny edge detection algorithm
KR101271092B1 (en) Method and apparatus of real-time segmentation for motion detection in surveillance camera system
CN109086724B (en) Accelerated human face detection method and storage medium
CN109214380B (en) License plate inclination correction method
CN105574855A (en) Method for detecting infrared small targets under cloud background based on temperate filtering and false alarm rejection
CN111046827A (en) Video smoke detection method based on convolutional neural network
CN104268877A (en) Infrared image sea-sky-line self adaption detection method
CN111047624A (en) Image dim target detection method, device, equipment and storage medium
CN112541486A (en) Infrared weak and small target detection algorithm based on improved Pixel segmentation
CN110827262A (en) Weak and small target detection method based on continuous limited frame infrared image
CN114693946A (en) Image anomaly detection method and device, computer equipment and storage medium
CN110913209B (en) Camera shielding detection method and device, electronic equipment and monitoring system
CN112329764A (en) Infrared dim target detection method based on TV-L1 model
CN115631191A (en) Coal blockage detection algorithm based on gray level features and edge detection
CN112115878A (en) Forest fire smoke root node detection method based on smoke area density
CN113205494B (en) Infrared small target detection method and system based on adaptive scale image block weighting difference measurement
Zhang et al. Research on the image segmentation based on improved threshold extractions
CN106780545A (en) A kind of Weak target energy accumulation Enhancement Method of combination time-space domain
CN112395944A (en) Multi-scale ratio difference combined contrast infrared small target detection method based on weighting
CN113643315B (en) Infrared small target detection method based on self-adaptive peak gradient descent filter
CN111881837A (en) Video SAR moving target detection method based on shadow extraction
CN108573236B (en) Method for detecting infrared weak and small target under cloud background based on discrete fraction Brown random field
Qin et al. A shadow removal algorithm for ViBe in HSV color space
WO2023042337A1 (en) Image processing system
CN108401563B (en) Infrared small target detection method based on Multiscale mean values filtering and conspicuousness detection

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant