CN111401195A - Sea surface target detection method based on multiband infrared image - Google Patents
Sea surface target detection method based on multiband infrared image Download PDFInfo
- Publication number
- CN111401195A CN111401195A CN202010162630.4A CN202010162630A CN111401195A CN 111401195 A CN111401195 A CN 111401195A CN 202010162630 A CN202010162630 A CN 202010162630A CN 111401195 A CN111401195 A CN 111401195A
- Authority
- CN
- China
- Prior art keywords
- multiband
- target
- vector
- background
- infrared
- 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
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 32
- 239000013598 vector Substances 0.000 claims abstract description 84
- 238000000034 method Methods 0.000 claims abstract description 48
- 230000001629 suppression Effects 0.000 claims abstract description 13
- 230000011218 segmentation Effects 0.000 claims description 12
- 230000015572 biosynthetic process Effects 0.000 claims description 7
- 238000003384 imaging method Methods 0.000 claims description 7
- 238000003786 synthesis reaction Methods 0.000 claims description 7
- 238000001914 filtration Methods 0.000 claims description 5
- 230000003044 adaptive effect Effects 0.000 claims description 3
- 230000002194 synthesizing effect Effects 0.000 claims description 3
- 230000005855 radiation Effects 0.000 abstract description 8
- 238000003331 infrared imaging Methods 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000003595 spectral effect Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000001308 synthesis method Methods 0.000 description 2
- 238000010521 absorption reaction Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 239000002131 composite material Substances 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000002401 inhibitory effect Effects 0.000 description 1
- 230000005764 inhibitory process Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000010187 selection method Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/10—Terrestrial scenes
- G06V20/13—Satellite images
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/30—Noise filtering
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Astronomy & Astrophysics (AREA)
- Remote Sensing (AREA)
- Photometry And Measurement Of Optical Pulse Characteristics (AREA)
- Image Analysis (AREA)
Abstract
A sea surface target detection method based on multiband infrared images belongs to the field of infrared target detection. The method aims at the acquired infrared radiation images of the same scene in four different wave band ranges to complete target detection based on infrared multiband image information. Firstly, background estimation based on multiband vector characteristics is carried out; then based on the multiband vector characteristic difference between the target and the background, calculating the vector distance and the vector included angle between each pixel in the target scene and the background, increasing the difference between the target and the background, and further finishing background suppression; and finally, segmenting the target by adopting a self-adaptive threshold method. The method greatly improves the signal-to-noise ratio of the target image by utilizing the multiband vector characteristic difference between the target and the background, has the performance independent of the size, the shape or the motion information of the target, has strong adaptability, is easy to realize, and can be applied to target detection and tracking under the complex sea surface background.
Description
Technical Field
The invention relates to a sea surface target detection method based on a multiband infrared image, namely a complex sea clutter multiband target detection method based on background suppression, and belongs to the field of infrared target detection.
Background
The infrared target detection under the complex background is the core technology and key problem of an infrared imaging guidance system, and is one of the hot spots of the research of scholars at home and abroad in recent years. For the sea surface infrared target image, as the target and the surrounding environment have heat exchange, and the air has scattering and absorption effects on heat radiation, the target signal is greatly interfered by the sea wave background, so that the contrast between the target and the background in the infrared image is poor, and the target edge is fuzzy, which brings great difficulty to the detection of the infrared target.
Aiming at the detection of infrared weak and small targets in a complex environment, an infrared imaging detection mode of a single waveband faces a plurality of bottlenecks. Therefore, the infrared imaging detection is gradually developed to the direction of multi-dimensional imaging such as dual-band, multi/hyperspectral and the like, and the infrared imaging detection can provide richer multi-band or multispectral information besides the common two-dimensional intensity image information, thereby creating better information input conditions for the detection and identification of targets. However, how to effectively utilize and extract target information quickly and accurately is a key for improving the performance of the weapon system for the acquired multi-dimensional image information.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: the method aims at the acquired infrared radiation images of the same scene in four different wave band ranges, and completes target detection based on infrared multiband image information. Firstly, background estimation based on multiband vector characteristics is carried out; then based on the multiband vector characteristic difference between the target and the background, calculating the vector distance and the vector included angle between each pixel in the target scene and the background, increasing the difference between the target and the background, and further finishing background suppression; and finally, segmenting the target by adopting a self-adaptive threshold method. The method greatly improves the signal-to-noise ratio of the target image by utilizing the multiband vector characteristic difference between the target and the background, has the performance independent of the size, the shape or the motion information of the target, has strong adaptability, is easy to realize, and can be applied to target detection and tracking under the complex sea surface background.
The purpose of the invention is realized by the following technical scheme:
a sea surface target detection method based on multiband infrared images comprises the following steps:
acquiring infrared images of multiple wave bands of the same target scene;
secondly, performing background estimation on the acquired infrared images of multiple wave bands by utilizing the multi-wave band vector characteristic to obtain a background estimation vector;
calculating the distance and the included angle between the multiband characteristic vector of each pixel point in the target scene and the background estimation vector;
step four, synthesizing and calculating the distance and the included angle in the step three to obtain an image after background suppression;
and step five, segmenting the image obtained in the step four by adopting a minimum error self-adaptive threshold method, and detecting the target position.
Preferably, the sea surface target detection method based on the multiband infrared image adopts an infrared multiband imaging system based on rotary filtering to acquire a target scene.
According to the sea surface target detection method based on the multiband infrared images, preferably, the infrared images of four wave bands of the same target scene are collected; the wavelengths of the four bands are respectively: 3.7-4.8 μm, 3.0-3.3 μm, 3.4-4.1 μm, 4.4-4.9 μm.
In the method for detecting the sea surface target based on the multiband infrared image, preferably, the frame frequency of each band of the infrared multiband imaging system based on the rotary filtering is not lower than 80 Hz.
Preferably, the method for detecting a sea surface target based on a multiband infrared image includes: firstly, calculating the mean value of the infrared image of each wave band, and then forming a vector by all the mean values to be used as a background estimation vector.
Preferably, the method for detecting the sea surface target based on the multiband infrared image is to calculate the distance and the included angle between the multiband characteristic vector of each pixel point in the target scene and the background estimation vector, and comprises the following steps: firstly, calculating the gray level of the infrared image of each wave band, and then forming a vector by the gray level of the infrared image of each wave band to be used as a multi-band feature vector; and finally, calculating the distance and included angle between the multiband characteristic vector and the background estimation vector.
Preferably, the synthesis of the distance and the included angle is calculated as a product of the distance and the included angle of the multiband characteristic vector and the background estimation vector.
Preferably, the method for detecting the sea surface target based on the multiband infrared image adopts a minimum error adaptive threshold method to segment the image obtained in the fourth step, and then calculates the centroid coordinates of the obtained target segmentation region, namely, obtains the required target detection result.
Compared with the prior art, the invention has the following beneficial effects:
(1) compared with the prior art that target detection is carried out based on the difference of infrared radiation intensity, the method can effectively distinguish the target from the background by utilizing the multiband vector characteristic difference between the target and the background, and achieves the purposes of background inhibition and image signal-to-noise ratio improvement;
(2) the existing method mostly adopts a method of fusing or pseudo-color reproduction of a plurality of images in the utilization of multiband image information, increases the information content of the images to a certain extent, and cannot achieve the purpose of inhibiting the background noise of the images; according to the method, based on a plurality of different waveband images, the multiband vector distance and the included angle of each pixel point and the background estimation are calculated, and the difference between the target and the background is maximized through a specific synthesis method, so that the signal-to-noise ratio of the image is greatly improved, and the detection and the identification of the infrared target are facilitated;
(3) the method carries out target detection based on the multiband vector characteristic difference of the artificial target and the natural background, the multiband vector characteristic difference of the target and the background is determined by respective spectral radiation curves, the detection effect does not depend on the size, the shape, the radiation intensity and the motion information of the target, and the method has strong adaptability;
(4) when background suppression is carried out, through directly extracting multi-band vector distance and included angle information of a target and a background and simple synthesis operation, complex operations such as space domain window scanning or frequency domain transformation are not needed, the method is easy to implement and good in real-time performance.
Drawings
FIG. 1 is a block flow diagram of the method of the present invention;
FIG. 2(a) is an original input 3.7 μm-4.8 μm infrared image of an embodiment of the present invention;
FIG. 2(b) is an original input 3.0 μm-3.3 μm infrared image of an embodiment of the present invention;
FIG. 2(c) is an original input 3.4 μm-4.1 μm infrared image of an embodiment of the present invention;
FIG. 2(d) is an original input 4.4 μm-4.9 μm infrared image of an embodiment of the present invention;
fig. 3(a) is a distance image of the multiband feature vectors and the background estimation vectors of the pixel points of the target scene in fig. 2;
FIG. 3(b) is an image of an included angle between a multiband feature vector and a background estimation vector of each pixel point of the target scene in FIG. 2;
FIG. 4 is a background-suppressed image after the information of FIGS. 3(a) and 3(b) is synthesized;
FIG. 5 is the target image of FIG. 4 after threshold segmentation;
FIG. 6 is a schematic diagram of results obtained by the method of the present embodiment under different sea surface scenes; wherein:
a1, a2, and A3: sequentially obtaining original infrared intensity images of different scenes;
b1, B2 and B3: multiband feature vector distance images corresponding to a1, a2 and A3 in this order;
c1, C2 and C3: sequentially forming multiband characteristic vector angle images corresponding to A1, A2 and A3;
d1, D2 and D3: the information-synthesized background suppressed images corresponding to a1, a2, and A3 were in order.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention will be described in detail with reference to the accompanying drawings.
A sea surface target detection method based on multiband infrared images comprises the following steps:
acquiring infrared images of multiple wave bands of the same target scene by adopting an infrared multi-band imaging system based on rotary filtering; in the example, infrared images of four wave bands of the same target scene are collected; the wavelengths of the four bands are respectively: 3.7-4.8 μm, 3.0-3.3 μm, 3.4-4.1 μm, 4.4-4.9 μm; the frame frequency of each wave band is not lower than 80Hz, and the time difference of collecting images in different wave bands is ignored.
Secondly, performing background estimation on the acquired infrared images of multiple wave bands by utilizing the multi-wave band vector characteristic to obtain a background estimation vector; the background estimation vector is obtained by the following method: firstly, calculating the mean value of the infrared image of each wave band, and then forming a vector by all the mean values to be used as a background estimation vector.
Calculating the distance and the included angle between the multiband characteristic vector of each pixel point in the target scene and the background estimation vector; the specific method comprises the following steps: firstly, calculating the gray level of the infrared image of each wave band, and then forming a vector by the gray level of the infrared image of each wave band to be used as a multi-band feature vector; and finally, calculating the distance and included angle between the multiband characteristic vector and the background estimation vector.
Step four, synthesizing and calculating the distance and the included angle in the step three to obtain an image after background suppression; the synthesis calculation method is the product of the distance and the included angle of the multiband characteristic vector and the background estimation vector.
And step five, segmenting the image obtained in the step four by adopting a minimum error adaptive threshold method, and then calculating the centroid coordinates of the obtained target segmentation region, namely detecting the target position.
Example (b):
FIG. 1 is a schematic block diagram of a process flow for an embodiment of the present invention. The method mainly comprises the following steps: the method comprises four steps of multiband infrared image input, background estimation based on multiband vector characteristics, background suppression based on vector distance and included angle and threshold segmentation.
The method specifically comprises the following steps:
step 1, an infrared multiband imaging system based on rotary filtering installs a rotary filter in the light path of the traditional infrared imaging system, the light transmission wave bands of the filter are respectively 3.7-4.8 μm, 3.0-3.3 μm, 3.4-4.1 μm and 4.4-4.9 μm, the system sequentially obtains different wave band images of a target scene, the frame frequency of each wave band can reach 80Hz, and the time errors of the different wave band images in the method are approximately ignored. Namely, the acquired 3.7-4.8 μm, 3.0-3.3 μm, 3.4-4.1 μm and 4.4-4.9 μm band infrared images of the same scene are input, as shown in fig. 2(a), fig. 2(b), fig. 2(c) and fig. 2(d), respectively.
And 2, carrying out background estimation based on the multiband vector characteristics aiming at the four collected multiband infrared images.
Calculating the mean values of the infrared images of the wave bands of 3.7-4.8 μm, 3.0-3.3 μm, 3.4-4.1 μm and 4.4-4.9 μm, namely the mean values of the infrared images of the image bands of FIG. 2(a), FIG. 2(b), FIG. 2(c) and FIG. 2(d), which are recorded as u1、u2、u3、u4Then the background estimation based on the multiband behaviour is taken as vector a, a ═ u1，u2，u3，u4)。
For small target images, most areas in the images are background regions, so the vector A represents multiband characteristics of the background in the target scene.
And 3, calculating the distance and the included angle between the multiband characteristic vector of each pixel point in the target scene and the background estimation vector.
Let four bands (i.e. 3.7) in FIG. 2(a), FIG. 2(b), FIG. 2(c) and FIG. 2(d)Mum-4.8 μm, 3.0 μm-3.3 μm, 3.4 μm-4.1 μm, 4.4 μm-4.9 μm) infrared image gray scale is f1(i,j)、f2(i,j)、f3(i,j)、f4(i, j), i, j are pixel coordinates in the image. The multiband feature vector for each pixel location in the target scene is denoted as B (i, j), where B (i, j) is (f)1(i,j)，f2(i,j)，f3(i,j)，f4(i,j))。
The distance d (i, j) and the included angle α (i, j) between the multiband feature vector of each pixel point in the target scene and the background estimation vector are calculated according to the formula (1) and the formula (2), and the resulting images are shown in fig. 3(a) and fig. 3 (b).
The multiband characteristic vector difference between the artificial target and the sea surface background is determined by the spectral radiation curve difference between the artificial target and the sea surface background, so that the vector distance d (i, j) and the included angle α (i, j) between the artificial target and the sea surface background represent the difference between the artificial target and the background from the angle of the spectral radiation difference, and can be used as the basis for subsequent background suppression.
And 4, carrying out specific synthesis calculation on the vector distance and the vector included angle based on the multiband characteristics to obtain an image with suppressed background, and calculating the following steps
g(i,j)＝α(i,j)×d(i,j) (3)
In the formula, g (i, j) is an image after background suppression, α (i, j) is a multiband vector angle image of each pixel point and the background in a target scene, d (i, j) is a multiband vector distance image, d (i, j) and α (i, j) reflect multiband characteristic difference of the target and the background, compared with an infrared intensity image, the signal-to-noise ratio of the sea background is improved to a certain extent, but background noise components in the image are still more, and by adopting a synthesis method of the formula (3), nonlinear stretching based on a vector angle is carried out on the characteristic vector distance of the target and the background, so that the background suppression effect is further improved, as shown in fig. 4.
And 5, performing target segmentation on the image with the suppressed background by using a minimum error method. The minimum error threshold method is derived based on the Bayes minimum error classification criterion on the premise that the gray distribution of an ideal target and a background is assumed to follow a mixed normal distribution. The method is less influenced by the relative sizes of the target and the background, and is a good threshold value selection method for small target images. The segmentation threshold is selected as follows.
η(t)＝1-ωo(t)lnωo(t)-ωb(t)lnωb(t)+ωo(t)lnσo(t)+ωb(t)lnσb(t)
In the formula, ωo(t) and ωb(t) respectively representing prior probabilities of the object and background at a threshold t of the original grey scale image,
and
when η (t) takes the minimum value, the optimal segmentation threshold t is obtained*：
L, the gray scale of the image, the obtained segmentation result is shown in figure 5 by using the optimal segmentation threshold value, and on the basis, the centroid coordinates of the obtained target segmentation area are calculated, so that the required target detection result can be obtained.
A schematic diagram of the target detection results in different sea-sky scenes by using the method of the present embodiment is shown in fig. 6, in FIG. 6, the original infrared intensity image (3.7 μm-4.8 μm), the vector distance image, the vector angle image, and the background suppression image after information synthesis are sequentially arranged from left to right, it can be seen that, compared with the infrared intensity image, the background suppression information composite image obtained by the method of the present embodiment, the signal-to-clutter ratio is greatly improved, and the target image with the signal-to-clutter ratio greatly improved compared with the infrared intensity image is obtained after the multi-band characteristic vector synthesis processing based on the acquired multi-band infrared image, so that the subsequent target segmentation and extraction are facilitated, the performance of the method does not depend on the size, shape or motion information of the target, the method is strong in adaptability and easy to implement, and the method can be applied to target detection and tracking under the complex sea surface background.
Those skilled in the art will appreciate that those matters not described in detail in the present specification are well known in the art.
Although the present invention has been described with reference to the preferred embodiments, it is not intended to limit the present invention, and those skilled in the art can make variations and modifications of the present invention without departing from the spirit and scope of the present invention by using the methods and technical contents disclosed above.
Claims (8)
1. A sea surface target detection method based on multiband infrared images is characterized by comprising the following steps:
acquiring infrared images of multiple wave bands of the same target scene;
secondly, performing background estimation on the acquired infrared images of multiple wave bands by utilizing the multi-wave band vector characteristic to obtain a background estimation vector;
calculating the distance and the included angle between the multiband characteristic vector of each pixel point in the target scene and the background estimation vector;
step four, synthesizing and calculating the distance and the included angle in the step three to obtain an image after background suppression;
and step five, segmenting the image obtained in the step four by adopting a minimum error self-adaptive threshold method, and detecting the target position.
2. The method of claim 1, wherein the target scene is collected by an infrared multiband imaging system based on rotating filtering.
3. The method for detecting the sea surface target based on the multiband infrared image as claimed in claim 1, wherein the infrared images of four wave bands of the same target scene are collected; the wavelengths of the four bands are respectively: 3.7-4.8 μm, 3.0-3.3 μm, 3.4-4.1 μm, 4.4-4.9 μm.
4. The method of claim 2, wherein the frame frequency of each band of the infrared multiband imaging system based on the rotating filter is not lower than 80 Hz.
5. The method for detecting the sea surface target based on the multiband infrared image according to one of claims 1 to 4, wherein the background estimation vector is obtained by: firstly, calculating the mean value of the infrared image of each wave band, and then forming a vector by all the mean values to be used as a background estimation vector.
6. The method for detecting the sea surface target based on the multiband infrared image as claimed in one of claims 1 to 4, wherein the method for calculating the distance and the included angle between the multiband characteristic vector of each pixel point in the target scene and the background estimation vector comprises the following steps: firstly, calculating the gray level of the infrared image of each wave band, and then forming a vector by the gray level of the infrared image of each wave band to be used as a multi-band feature vector; and finally, calculating the distance and included angle between the multiband characteristic vector and the background estimation vector.
7. The method for detecting the sea surface target based on the multiband infrared image according to one of claims 1 to 4, wherein the synthesis of the distance and the included angle is calculated as a product of the distance and the included angle of the multiband feature vector and the background estimation vector.
8. The method for detecting the sea surface target based on the multiband infrared image according to one of claims 1 to 4, wherein the image obtained in the fourth step is segmented by a minimum error adaptive threshold method, and then the centroid coordinates of the obtained target segmentation region are calculated, so as to obtain the required target detection result.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010162630.4A CN111401195B (en) | 2020-03-10 | 2020-03-10 | Sea surface target detection method based on multiband infrared image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010162630.4A CN111401195B (en) | 2020-03-10 | 2020-03-10 | Sea surface target detection method based on multiband infrared image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111401195A true CN111401195A (en) | 2020-07-10 |
| CN111401195B CN111401195B (en) | 2023-11-14 |
Family
ID=71432316
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010162630.4A Active CN111401195B (en) | 2020-03-10 | 2020-03-10 | Sea surface target detection method based on multiband infrared image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111401195B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113589311A (en) * | 2021-07-15 | 2021-11-02 | 中国科学院上海技术物理研究所 | Infrared differential detection method for dark and weak targets |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101676744A (en) * | 2007-10-31 | 2010-03-24 | 北京航空航天大学 | Method for tracking small target with high precision under complex background and low signal-to-noise ratio |
| CN108682004A (en) * | 2018-04-16 | 2018-10-19 | 华中科技大学 | A kind of infrared small target in complex background detection method based on local message |
| CN109102003A (en) * | 2018-07-18 | 2018-12-28 | 华中科技大学 | A kind of small target detecting method and system based on Infrared Physics Fusion Features |
| CN109492543A (en) * | 2018-10-18 | 2019-03-19 | 广州市海林电子科技发展有限公司 | The small target detecting method and system of infrared image |
| CN109697431A (en) * | 2018-12-29 | 2019-04-30 | 哈尔滨工业大学 | A kind of detection method of small target based on high spectrum image |
-
2020
- 2020-03-10 CN CN202010162630.4A patent/CN111401195B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101676744A (en) * | 2007-10-31 | 2010-03-24 | 北京航空航天大学 | Method for tracking small target with high precision under complex background and low signal-to-noise ratio |
| CN108682004A (en) * | 2018-04-16 | 2018-10-19 | 华中科技大学 | A kind of infrared small target in complex background detection method based on local message |
| CN109102003A (en) * | 2018-07-18 | 2018-12-28 | 华中科技大学 | A kind of small target detecting method and system based on Infrared Physics Fusion Features |
| CN109492543A (en) * | 2018-10-18 | 2019-03-19 | 广州市海林电子科技发展有限公司 | The small target detecting method and system of infrared image |
| CN109697431A (en) * | 2018-12-29 | 2019-04-30 | 哈尔滨工业大学 | A kind of detection method of small target based on high spectrum image |
Non-Patent Citations (2)
| Title |
|---|
| 刘翔等: "复杂背景中红外多光谱目标检测算法研究" * |
| 杨磊: "复杂背景条件下的红外小目标检测与跟踪算法研究" * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113589311A (en) * | 2021-07-15 | 2021-11-02 | 中国科学院上海技术物理研究所 | Infrared differential detection method for dark and weak targets |
| CN113589311B (en) * | 2021-07-15 | 2023-11-28 | 中国科学院上海技术物理研究所 | Infrared differential detection method for dim and weak targets |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111401195B (en) | 2023-11-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108389188B (en) | Sparse hyperspectral abnormal target detection method | |
| CN106778551B (en) | Method for identifying highway section and urban road lane line | |
| Yang et al. | Fusion of infrared polarization and intensity images using support value transform and fuzzy combination rules | |
| Wang et al. | Window zooming–based localization algorithm of fruit and vegetable for harvesting robot | |
| Zhang et al. | Vehicle recognition algorithm based on Haar-like features and improved Adaboost classifier | |
| CN109919960B (en) | Image continuous edge detection method based on multi-scale Gabor filter | |
| CN110782477A (en) | Moving target rapid detection method based on sequence image and computer vision system | |
| CN102222322A (en) | Multiscale non-local mean-based method for inhibiting infrared image backgrounds | |
| CN104036523A (en) | Improved mean shift target tracking method based on surf features | |
| CN110660065B (en) | Infrared fault detection and identification algorithm | |
| CN113298147B (en) | Image fusion method and device based on regional energy and intuitionistic fuzzy set | |
| CN114758219B (en) | Trace identification method based on spectral data and infrared temperature data fusion | |
| CN110580705B (en) | Method for detecting building edge points based on double-domain image signal filtering | |
| CN112150386B (en) | SAR image speckle non-local average inhibition method based on contrast mean value | |
| Yang et al. | Adversarial reconstruction CNN for illumination-robust frontal face image recovery and recognition | |
| CN111401195A (en) | Sea surface target detection method based on multiband infrared image | |
| CN110276782B (en) | Hyperspectral target tracking method combining spatial spectral features and related filtering | |
| CN110047041A (en) | A kind of empty-frequency-domain combined Traffic Surveillance Video rain removing method | |
| CN112163606B (en) | Infrared small target detection method based on block contrast weighting | |
| Alsulami et al. | Detection and tracking of dim objects in infrared (IR) images using Support Vector Machine | |
| CN106355576A (en) | SAR image registration method based on MRF image segmentation algorithm | |
| CN116342653A (en) | Target tracking method, system, equipment and medium based on correlation filter | |
| CN112883796B (en) | SAR image multi-type building height estimation method based on overlapping and masking information | |
| CN112329796B (en) | Infrared imaging cloud detection method and device based on visual saliency | |
| CN112183633B (en) | Hyperspectral image salient target detection method based on improved FT algorithm |
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 |