CN104103073A - Infrared polarization image edge detection method - Google Patents
Infrared polarization image edge detection method Download PDFInfo
- Publication number
- CN104103073A CN104103073A CN201410333224.4A CN201410333224A CN104103073A CN 104103073 A CN104103073 A CN 104103073A CN 201410333224 A CN201410333224 A CN 201410333224A CN 104103073 A CN104103073 A CN 104103073A
- Authority
- CN
- China
- Prior art keywords
- image
- polarization
- infrared
- polarization image
- pixel
- 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.)
- Pending
Links
Landscapes
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention belongs to the field of infrared image processing, and provides an infrared polarization image edge detection method. The infrared polarization image edge detection method comprises the following steps: 1) colleting and pre-processing infrared polarization image information; 2) carrying out image shift processing, and separately carrying out a shift operation to the pre-processed polarization images of three channels; 3) carrying out Stokes polarization state resolution; 4) finishing polarization component decomposition; 5) carrying out superposition fusion and threshold value segmentation; and 6) carrying out morphological processing, smoothening a target outline through open operation, weakening a narrow part, removing a thin extruded part, and extracting the edge of a single pixel by morphological refined operation. Infrared polarization characteristic differences are utilized to enhance target edge outline information with transition nature and decompose intensity information ineffective to target identification, and edge information can be effectively highlighted to realize edge detection.
Description
Technical field
The invention belongs to infrared image processing field, be specifically related to a kind of edge detection method of infrared polarization image information.
Background technology
Infrared imagery technique due to have passive work, target recognition capability strong, can all weather operations etc. advantage, be widely used at aspects such as military surveillance, supervision and guidances.But being subject to the impact of atmospheric attenuation and complex background, infrared imaging quality is unsatisfactory; Along with the development of Optoelectronic Countermeasure Technology, the particularly use of some novel stealthy measures of infrared camouflage, the infrared emittance of target is reduced greatly, and the infrared signature of target is distorted, cause in infrared imaging target-background contrasts little, the contour structure characteristic of target is not outstanding, and this has brought serious challenge to the detection and identify ability of infrared imaging detection system.
For infrared imaging system, in the problem facing aspect target detection identification, existing solution can be divided into three kinds: first method is active probe technology, is the additional active IR imaging instrument of infrared eye or millimetre-wave radar; Second method is multispectral technology, utilizes multiple wave bands to survey, and realizes the complementation of information; The third method is Infrared Polarization Imaging Technology, not only obtains Infrared Targets strength information, also gets polarization information.
Infrared Polarization Imaging Technology is the novel imaging detection technology developing rapidly in the recent decade, compared with traditional infrared intensity imaging, infrared polarization is imaged on made Target detection and identify aspect and has clear superiority, can be used for resisting the hidden measure of infrared camouflage, strengthens antijamming capability.Be embodied in:
(1) polarization characteristic of target is determined by self properties such as its material properties, texture structure, surfacenesses, and polarization information has reflected target self information.
(2), compared with infrared intensity imaging, owing to having increased polarization information, can distinguish the things that traditional infrared intensity imaging cannot be distinguished.
(3) made Target has obvious polarizability with its surface smoothness, and the discreteness of natural scene (dewatering outer) polarization is larger, and overall polarizability is less.Between made Target and natural background, the larger difference of polarizability can be used for distinguishing of made Target in natural background.
(4) metal and insulator, because the difference of electric conductivity own causes the difference of the two polarizability, can be used for distinguishing of the two.
Edge, is defined as gray scale in image zone boundary jumpy occurs, and is the essential characteristic of image, has contained the most information of image.Rim detection plays an important role in the fields such as target identification, computer vision.For the made Target under some complex environment, and in pretending the target of hidden state, its infrared imaging signal noise ratio (snr) of image is low, target background contrast is little, objective contour is fuzzy, and the method effect that single order is led in conventional utilizing, second order is led extraction edge is unsatisfactory.
In sum, infrared polarization imaging detection technology is the effective means of the detection and identify of made Target or infrared camouflage Stealthy Target under reply complex environment, by to the abundant excavation of infrared polarization image information and rationally utilization, just can effectively detect object edge.From current disclosed technical literature, less for the edge detection method of infrared polarization image.Thus, by infrared polarization imaging mechanism and infrared polarization image information analysis and research, a kind of new infrared polarization method for detecting image edge has been proposed.
Summary of the invention
The problem to be solved in the present invention is the rim detection of made Target under complex background condition, by hyperchannel infrared polarization image information is processed, has proposed a kind of new infrared polarization method for detecting image edge.
Concrete technical scheme is as follows, and a kind of infrared polarization method for detecting image edge, comprises the following steps:
The first step: the collection of infrared polarization image information and pre-service, utilize the infrared polarization imaging detector infrared polarization image that collection is no less than three passages to target scene, the image gathering is carried out to pre-service;
Second step: image shift processing, by the operation of getting off respectively of pretreated triple channel polarization image in first step,
(1) keep first passage polarization image constant, by second channel polarization image, toward N pixel of lower left translation, third channel polarization image, toward N pixel of upper right side translation, obtains the triple channel polarization image after displacement, and N is natural number;
(2) keep first passage polarization image constant, by second channel polarization image, toward N pixel of lower right translation, third channel polarization image, toward N pixel of upper left side translation, obtains the triple channel polarization image after displacement;
The 3rd step: Stokes polarization state is resolved, carries out respectively object scene polarization state according to two width triple channel polarization images after displacement in second step and resolves;
The 4th step: polarized component decomposes completely, and the polarization state calculation result of object scene in third step is decomposed into complete polarized component and natural light composition sum;
The 5th step: additive fusion and Threshold segmentation, two complete polarized components that decomposition is obtained carry out additive fusion, and edge information is carried out complementation, and it is carried out to Threshold segmentation;
The 6th step: morphology processing, by the level and smooth objective contour of opening operation, weaken narrow, remove thin outshot, and operate by morphologic thinning, extract the edge of single pixel.
Preferably, in second step, in described step (1), by N pixel of clockwise second channel polarization image 135 ° of direction translations, third channel polarization image, toward N pixel of counterclockwise 45 ° of direction translations, obtains the triple channel polarization image after displacement; In described step (2) by second channel polarization image toward N pixel of clockwise 45 ° of direction translations, third channel polarization image is toward N pixel of counterclockwise 135 ° of direction translations, obtains the triple channel polarization image after displacement.
Preferably, described N value is 2.
Ultimate principle of the present invention is:
The infrared intensity information of object scene is due to the impact of environmental baseline and other jamming countermeasures, do not possess stability and reliability, and the polarization information being determined by object self property is more reliable and more stable, bond strength information and polarization information can obtain better detection and identify performance.Infrared polarization characteristic between target and background or between different target there are differences, and makes object edge form the transition of polarization information.Utilize this infrared polarization property difference, the object edge profile information with transition nature is strengthened, and decomposite the strength information invalid to target identification, just projecting edge information effectively, realizes rim detection.
Image shift handling principle: the hyperchannel polarization image that target scene is gathered, the polarization strength information between its target and background has larger contrast, in the slow transition of several pixels place intensity at object edge place.Even if in this target of simulation and stealthy target and the minimum situation of background infrared intensity contrast, due to the existence of target polarizability and with the difference of polarization of background, in the polarization image that hyperchannel obtains, the polarization intensity contrast of target and background still exists.
Generally speaking, the polarizability of made Target is greater than the polarizability of natural background, and the polarizability of object edge is between between the two, totally weak (being less than 10%) of the polarizability of target and background, directly utilizes polarization information to be difficult to strengthen object edge information.Here adopt simply effective image shift disposal route, at object edge place, target information, marginal information and background information mixing have been carried out to polarization state and resolve, just can greatly improve the polarizability of edge; And information in objective plane and background surfaces is after displacement by a small margin, bring marked change can not to its polarizability.
Polarized component decomposes completely: the infrared radiation of object scene is all partial poolarized light conventionally, can be considered by natural light composition and complete polarized component and is formed by stacking.Conventionally because the polarizability of object is relatively little, natural light composition is much larger than complete polarized component, under certain situation, thisly occupies heavy natural light composition and but do not possess ability to see things in their true light; And due to the difference of different objects polarizability, the complete polarized component that component is less has possessed ability to see things in their true light on the contrary, just can strengthen target detection recognition capability by decompositing complete polarized component.
Adopt the beneficial effect of acquisition of the present invention:
The LONG WAVE INFRARED polarization image data that utilization measures, adopt method of the present invention, have realized the detection of object edge profile.The present invention is based on the ultimate principle of infrared polarization imaging, utilized the infrared polarization property difference of made Target and natural background, by strengthening the polarization information of object edge, and extract the complete polarized component of image, finally realized the rim detection of target.The present invention also has following features:
(1) entire image is processed, algorithm flow is relatively easy, and applicability is very strong.
(2) good to the rejection of background clutter, in the complete polarized component information of extraction, by being as for followed successively by by force information in object edge information, target face, background clutter information.
(3) wider than conventional edge detection algorithm for its applicability of polarization image, can not only be applicable to the rim detection of made Target in natural background, can also be applied to the rim detection of infrared camouflage Stealthy Target.
Brief description of the drawings
Fig. 1 is schematic flow sheet of the present invention;
Fig. 2 is schematic diagram of the present invention;
Fig. 3 is corresponding 0 ° of two targets obtaining in the embodiment of the present invention, 60 °, and the polarization image of 120 °;
Fig. 4 is three parameter I of Stokes vector corresponding to two targets in the present invention
1, Q
1, U
1image;
Fig. 5 is three parameter I of Stokes vector corresponding to two targets in the present invention
2, Q
2, U
2image;
Fig. 6 is two complete polarized component I that target is corresponding in the present invention
p1with natural light composition I
n1image;
Fig. 7 is two complete polarized component I that target is corresponding in the present invention
p2with natural light composition I
n2image;
Fig. 8 is additive fusion result images in the present invention;
Fig. 9 is Threshold segmentation result images in the present invention;
Figure 10 is edge detection results image in the present invention.
Embodiment
Below in conjunction with Figure of description, to embodiment, the present invention will be described.
The problem to be solved in the present invention is the rim detection of made Target under complex background condition, by hyperchannel infrared polarization image information is processed, has proposed a kind of infrared polarization method for detecting image edge.
The first step: the collection of infrared polarization image information and pre-service
Utilize the infrared polarization imaging detector infrared polarization image that collection is no less than three passages (different analyzing directions) to target scene, just can carry out Stokes polarization state and resolve.Image pre-service comprises image rectification, filtering, registration, cutting etc., proofreaies and correct for the heterogeneity of offset detector response the imperfection of compensate for optical components and parts.Filtering is for reducing picture noise and remove bad point.The micrometric displacement deviation that registration exists for eliminating the each polarization image of Same Scene, finally crops the frame after image registration.In the present embodiment, three analyzing directions corresponding to passage are respectively that a passage analyzing direction is 0 °, and two passage analyzing directions are 60 °, and triple channel analyzing direction is 120 °; Adopting infrared polarization method for detecting image edge of the present invention to process two different targets, as shown in Figure 3, is two target image pre-service result figure that gather, and wherein, Fig. 3-a is that two targets are the polarization image I of 0 ° in analyzing direction
0 °, Fig. 3-b is that two targets are the polarization image I of 60 ° in analyzing direction
60 °, Fig. 3-c is that two targets are the polarization image I of 120 ° in analyzing direction
120 °.
Second step: image shift processing
For three pretreated polarization image (I of passage
0 °, I
60 °, I
120 °) carry out image shift processing.One passage polarization image remains unchanged, and two passage polarization images are (clockwise 135 ° of directions) translation toward lower left, and triple channel polarization image is (counterclockwise 45 ° of directions) translation toward upper right side, gives over to subsequent treatment; Then, a passage polarization image is still constant, and two passage polarization images are (clockwise 45 ° of directions) translation toward lower right, and triple channel polarization image is (counterclockwise 135 ° of directions) translation toward upper left side, again carries out subsequent treatment.Complete respectively shifting function and carry out subsequent treatment two mutually perpendicular directions, the displacement size of displacement is 2 pixels in the present embodiment.Image shift is processed computation process as shown in formula one, and wherein, i, j are image pixel coordinate, and the value of i, j is integer, the total pixel of width that i scope is image, the total pixel of length that j scope is image.
(formula one)
The 3rd step: Stokes polarization state is resolved
Utilize the triple channel polarization image after displacement
and
carry out respectively object scene polarization state and resolve, obtain respectively three parameter I of Stokes vector
1, Q
1, U
1and I
2, Q
2, U
2, as formula two (Tang Kun, Zou Jiwei, Jiang Tao, Yu great Bin. the infrared polarization characteristic research [J] of target and background. infrared and laser engineering, 2007,36 (5): 611-614.).
(formula two)
Wherein, n value is 1,2.
In the image result obtaining in the present embodiment, as shown in Figure 4, Fig. 4-a is the I in Stokes vector corresponding to two targets
1, the Q in Stokes vector corresponding to two targets of Fig. 4-b
1, the U in Stokes vector corresponding to two targets of Fig. 4-c
1; As shown in Figure 5, Fig. 5-a is the I in Stokes vector corresponding to two targets
2, the Q in Stokes vector corresponding to two targets of Fig. 5-b
2, the U in Stokes vector corresponding to two targets of Fig. 5-c
2.
The 4th step: polarized component decomposes completely
The polarization state of object scene partial polarization often, its Stokes vector has
a partial poolarized light can be decomposed into complete polarized component and natural light composition sum, see formula three.Wherein complete polarized component intensity
natural light composition intensity
by I
1, Q
1, U
1and I
2, Q
2, U
2substitution formula can obtain two complete polarized component I respectively
p1with I
p2.
In the present embodiment, as shown in Figure 6, Fig. 6-a is complete polarized component I
p1, Fig. 6-b is natural light composition intensity I
1'; As shown in Figure 7, Fig. 7-a is complete polarized component I
p2, Fig. 7-b is natural light composition intensity I
2'.
The 5th step: additive fusion and Threshold segmentation
Two complete polarized component I that decomposition is obtained
p1with I
p2carry out additive fusion, edge information is carried out complementation, and at this moment the marginal information of fused images significantly strengthens, and it is carried out to Threshold segmentation and just can Preliminary detection go out object edge, and formula four is shown in operating process.
I
PT＝I
P1+I
P2
Wherein, I
p1with I
p2be respectively the complete polarized component decomposing under two kinds of shifting processing, I
pTfor the result after additive fusion, T is the threshold value constant setting in advance, and W is Threshold segmentation result.
As shown in Figure 8, two targets result images after additive fusion separately; Two targets result images after Threshold segmentation separately as shown in Figure 9.
The 6th step: morphology processing
Opening operation.Due to reasons such as noises, there are some spuious point and burrs in Threshold segmentation result, by smoothly objective contour of opening operation, weakens narrow, removes thin outstanding.Opening operation is defined as
(formula five)
Wherein K is structural element, and W is exactly that result after W is corroded by K is expanded by K again by K opening operation.
Refinement.Due to reasons such as the fuzzy of the edge of image own and image shift processing, the border width splitting has occupied several pixels, operates by morphologic thinning, just can extract the edge of single pixel.
Wherein structural element sequence is { B}={B
1, B
2..., B
k, k is natural number, represents structural element sequence number, B
kb
k-1postrotational form.Whole process repeats until not further variation occurs, and net result as shown in figure 10.(Ruan Qiuqi. Digital Image Processing [M]. Electronic Industry Press, 2001.).
The foregoing is only the better enforcement example of this name, not in order to limit the present invention, all any amendments of doing within the spirit and principles in the present invention, be equal to and replace and improvement etc., within all should being included in protection scope of the present invention.
Claims (3)
1. an infrared polarization method for detecting image edge, is characterized in that, comprises the following steps:
The first step: the collection of infrared polarization image information and pre-service, utilize the infrared polarization imaging detector infrared polarization image that collection is no less than three passages to target scene, the image gathering is carried out to pre-service;
Second step: image shift processing, by the operation of getting off respectively of the polarization image of pretreated three passages in first step,
(1) keep first passage polarization image constant, by second channel polarization image, toward N pixel of lower left translation, third channel polarization image, toward N pixel of upper right side translation, obtains the polarization image of three passages after displacement, and N is natural number;
(2) keep first passage polarization image constant, by second channel polarization image, toward N pixel of lower right translation, third channel polarization image, toward N pixel of upper left side translation, obtains the polarization image of three passages after displacement;
The 3rd step: Stokes polarization state is resolved, carries out respectively object scene polarization state according to the polarization image obtaining after displacement in second step and resolves;
The 4th step: polarized component decomposes completely, and the polarization state calculation result of object scene in third step is decomposed into complete polarized component and natural light composition sum;
The 5th step: additive fusion and Threshold segmentation, two complete polarized components that decomposition is obtained carry out additive fusion, and the image after merging is carried out to Threshold segmentation;
The 6th step: morphology processing, by the level and smooth objective contour of opening operation, weaken narrow, remove thin outshot, and operate by morphologic thinning, extract the edge of single pixel.
2. a kind of infrared polarization method for detecting image edge according to claim 1, it is characterized in that: in second step, in described step (1) by second channel polarization image toward N pixel of clockwise 135 ° of direction translations, third channel polarization image, toward N pixel of counterclockwise 45 ° of direction translations, obtains the polarization image of three passages after displacement; In described step (2) by second channel polarization image toward N pixel of clockwise 45 ° of direction translations, third channel polarization image is toward N pixel of counterclockwise 135 ° of direction translations, obtains the polarization image of three passages after displacement.
3. a kind of infrared polarization method for detecting image edge according to claim 1, is characterized in that: described N value is 2.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410333224.4A CN104103073A (en) | 2014-07-14 | 2014-07-14 | Infrared polarization image edge detection method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410333224.4A CN104103073A (en) | 2014-07-14 | 2014-07-14 | Infrared polarization image edge detection method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN104103073A true CN104103073A (en) | 2014-10-15 |
Family
ID=51671193
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410333224.4A Pending CN104103073A (en) | 2014-07-14 | 2014-07-14 | Infrared polarization image edge detection method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104103073A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104793343A (en) * | 2015-04-13 | 2015-07-22 | 西安电子科技大学 | Three-channel and single-Wollaston prism polarization imaging device and polarization information detecting method |
| CN104835113A (en) * | 2015-04-30 | 2015-08-12 | 北京环境特性研究所 | Polarization image fusion method based on super-resolution image reconstruction |
| CN105890777A (en) * | 2016-03-31 | 2016-08-24 | 北京理工大学 | Infrared controllable partially-polarized radiation source |
| CN105959514A (en) * | 2016-04-20 | 2016-09-21 | 河海大学 | Weak target imaging detection device and method |
| CN106289542A (en) * | 2016-07-26 | 2017-01-04 | 北京理工大学 | A kind of radiation correction method of preposition polaroid type infrared polarization imaging system |
| CN107064905A (en) * | 2017-05-26 | 2017-08-18 | 北京环境特性研究所 | A kind of detection method of target |
| CN108492274A (en) * | 2018-04-03 | 2018-09-04 | 中国人民解放军国防科技大学 | Long-wave infrared polarization feature extraction and fusion image enhancement method |
| CN109934822A (en) * | 2019-03-26 | 2019-06-25 | 北京计算机技术及应用研究所 | A kind of infrared point target detecting method decomposed based on optimization polarization |
| CN110232694A (en) * | 2019-06-12 | 2019-09-13 | 安徽建筑大学 | A kind of infrared polarization thermal imagery threshold segmentation method |
| CN111667519A (en) * | 2020-06-05 | 2020-09-15 | 北京环境特性研究所 | Registration method and device for polarized images with different fields of view |
| CN112508852A (en) * | 2020-11-12 | 2021-03-16 | 深圳科瑞技术股份有限公司 | Method and system for detecting liquid level in liquid storage bottle |
| CN112581485A (en) * | 2020-11-18 | 2021-03-30 | 北京理工大学 | Imaging-free edge information detection method and system |
| CN113421206A (en) * | 2021-07-16 | 2021-09-21 | 合肥工业大学 | Image enhancement method based on infrared polarization imaging |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103530876A (en) * | 2013-10-10 | 2014-01-22 | 中国科学院上海技术物理研究所 | Infrared image analysis method based on Fourier transformation |
-
2014
- 2014-07-14 CN CN201410333224.4A patent/CN104103073A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103530876A (en) * | 2013-10-10 | 2014-01-22 | 中国科学院上海技术物理研究所 | Infrared image analysis method based on Fourier transformation |
Non-Patent Citations (4)
| Title |
|---|
| 唐坤等: "目标与背景的红外偏振特性研究", 《红外与激光工程》 * |
| 李艳斌等: "用于图象运动模糊检测及轮廓提取的空间域相关法研究", 《无线电工程》 * |
| 邱跳文等: "长波红外偏振图像信息的提取与增强", 《红外》 * |
| 陈伟力等: "采用中波红外偏振成像的目标探测实验", 《红外与激光工程》 * |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104793343A (en) * | 2015-04-13 | 2015-07-22 | 西安电子科技大学 | Three-channel and single-Wollaston prism polarization imaging device and polarization information detecting method |
| CN104835113A (en) * | 2015-04-30 | 2015-08-12 | 北京环境特性研究所 | Polarization image fusion method based on super-resolution image reconstruction |
| CN105890777A (en) * | 2016-03-31 | 2016-08-24 | 北京理工大学 | Infrared controllable partially-polarized radiation source |
| CN105890777B (en) * | 2016-03-31 | 2018-11-06 | 北京理工大学 | Infrared controllable part polarised radiation source |
| CN105959514A (en) * | 2016-04-20 | 2016-09-21 | 河海大学 | Weak target imaging detection device and method |
| CN105959514B (en) * | 2016-04-20 | 2018-09-21 | 河海大学 | A kind of weak signal target imaging detection device |
| CN106289542A (en) * | 2016-07-26 | 2017-01-04 | 北京理工大学 | A kind of radiation correction method of preposition polaroid type infrared polarization imaging system |
| CN106289542B (en) * | 2016-07-26 | 2018-10-30 | 北京理工大学 | A kind of radiation correction method of preposition polaroid type infrared polarization imaging system |
| CN107064905A (en) * | 2017-05-26 | 2017-08-18 | 北京环境特性研究所 | A kind of detection method of target |
| CN108492274B (en) * | 2018-04-03 | 2020-08-07 | 中国人民解放军国防科技大学 | Long-wave infrared polarization feature extraction and fusion image enhancement method |
| CN108492274A (en) * | 2018-04-03 | 2018-09-04 | 中国人民解放军国防科技大学 | Long-wave infrared polarization feature extraction and fusion image enhancement method |
| CN109934822A (en) * | 2019-03-26 | 2019-06-25 | 北京计算机技术及应用研究所 | A kind of infrared point target detecting method decomposed based on optimization polarization |
| CN110232694A (en) * | 2019-06-12 | 2019-09-13 | 安徽建筑大学 | A kind of infrared polarization thermal imagery threshold segmentation method |
| CN110232694B (en) * | 2019-06-12 | 2021-09-07 | 安徽建筑大学 | Infrared polarization thermal image threshold segmentation method |
| CN111667519A (en) * | 2020-06-05 | 2020-09-15 | 北京环境特性研究所 | Registration method and device for polarized images with different fields of view |
| CN111667519B (en) * | 2020-06-05 | 2023-06-20 | 北京环境特性研究所 | Registration method and device for polarized images with different fields of view |
| CN112508852A (en) * | 2020-11-12 | 2021-03-16 | 深圳科瑞技术股份有限公司 | Method and system for detecting liquid level in liquid storage bottle |
| CN112581485A (en) * | 2020-11-18 | 2021-03-30 | 北京理工大学 | Imaging-free edge information detection method and system |
| CN113421206A (en) * | 2021-07-16 | 2021-09-21 | 合肥工业大学 | Image enhancement method based on infrared polarization imaging |
| CN113421206B (en) * | 2021-07-16 | 2022-11-15 | 合肥工业大学 | Image enhancement method based on infrared polarization imaging |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN104103073A (en) | Infrared polarization image edge detection method | |
| Magri et al. | T-linkage: A continuous relaxation of j-linkage for multi-model fitting | |
| CN102184550B (en) | Mobile platform ground movement object detection method | |
| CN109754384B (en) | Infrared polarization image fusion method of uncooled infrared focal plane array | |
| Abdollahi et al. | Extraction of road features from UAV images using a novel level set segmentation approach | |
| Kaur et al. | Comparative study of different edge detection techniques | |
| CN107705313A (en) | A kind of remote sensing images Ship Target dividing method | |
| Hsieh et al. | Fast and robust infrared image small target detection based on the convolution of layered gradient kernel | |
| Zhang et al. | Multiple Saliency Features Based Automatic Road Extraction from High‐Resolution Multispectral Satellite Images | |
| Zhu et al. | Saliency‐Based Diver Target Detection and Localization Method | |
| Li et al. | Automatic infrared ship target segmentation based on structure tensor and maximum histogram entropy | |
| Xu et al. | COCO-Net: A dual-supervised network with unified ROI-loss for low-resolution ship detection from optical satellite image sequences | |
| Miller et al. | Person tracking in UAV video | |
| Ren et al. | Effective SAR sea ice image segmentation and touch floe separation using a combined multi-stage approach | |
| Maqueda et al. | Fast millimeter wave threat detection algorithm | |
| Ellmauthaler et al. | Infrared-visible image fusion using the undecimated wavelet transform with spectral factorization and target extraction | |
| Jaiswal et al. | A survey of image segmentation based on artificial intelligence and evolutionary approach | |
| Sivapriya et al. | ViT-DexiNet: a vision transformer-based edge detection operator for small object detection in SAR images | |
| Slamani et al. | Image processing tools for the enhancement of concealed weapon detection | |
| Weihua et al. | A real-time detection algorithm for unmanned aerial vehicle target in infrared search system | |
| Wang et al. | Hyperspectral target detection via global spatial-spectral attention network and background suppression | |
| Ren et al. | Target detection of maritime search and rescue: Saliency accumulation method | |
| Kaldane et al. | Land Cover Change Detection for Fully Polarimetric SAR Images | |
| Wu et al. | Moving-object Detection Based on Shadow Removal and Prospect Reconstruction | |
| Jabarullah et al. | Segmentation Using Saliency–Color Mapping Technique |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20141015 |