CN115345845A - Feature fusion smoke screen interference efficiency evaluation and processing method based on direction gradient histogram and electronic equipment - Google Patents
Feature fusion smoke screen interference efficiency evaluation and processing method based on direction gradient histogram and electronic equipment Download PDFInfo
- Publication number
- CN115345845A CN115345845A CN202210956882.3A CN202210956882A CN115345845A CN 115345845 A CN115345845 A CN 115345845A CN 202210956882 A CN202210956882 A CN 202210956882A CN 115345845 A CN115345845 A CN 115345845A
- Authority
- CN
- China
- Prior art keywords
- image
- smoke screen
- histogram
- eemssj
- hog
- 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
- 239000000779 smoke Substances 0.000 title claims abstract description 61
- 238000011156 evaluation Methods 0.000 title claims abstract description 21
- 230000004927 fusion Effects 0.000 title claims abstract description 16
- 238000003672 processing method Methods 0.000 title claims description 6
- 239000011159 matrix material Substances 0.000 claims abstract description 36
- 230000009466 transformation Effects 0.000 claims abstract description 9
- 230000003321 amplification Effects 0.000 claims abstract description 4
- 238000003199 nucleic acid amplification method Methods 0.000 claims abstract description 4
- 238000000034 method Methods 0.000 claims description 20
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000013441 quality evaluation Methods 0.000 claims description 5
- 239000000126 substance Substances 0.000 claims description 3
- 238000003491 array Methods 0.000 claims description 2
- 230000006870 function Effects 0.000 claims description 2
- 230000002708 enhancing effect Effects 0.000 claims 1
- 238000003384 imaging method Methods 0.000 abstract description 3
- 230000000694 effects Effects 0.000 description 7
- 238000001514 detection method Methods 0.000 description 4
- 238000011158 quantitative evaluation Methods 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/80—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level
- G06V10/806—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level of extracted features
-
- 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
- G06T2207/20—Special algorithmic details
- G06T2207/20048—Transform domain processing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Databases & Information Systems (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a feature fusion smoke screen interference efficiency evaluation method based on a direction gradient histogram and electronic equipment. And respectively calculating the directional gradient histogram characteristics of the reference image and the disturbed image. And performing Riss transformation based on the directional gradient histogram characteristics of the reference image and the interfered image, and solving a similarity matrix, a cosine similarity matrix and a correlation coefficient matrix of the reference image and the interfered image. And fusing the matrixes to obtain a smoke screen interference efficiency evaluation value. And calculating and setting a tracking drift threshold, and performing feature amplification processing on the fused features by using the threshold. The invention can quantitatively evaluate the interference efficiency of the smoke screen on the photoelectric imaging reconnaissance guidance equipment from the angle of image characteristics focused by image tracking.
Description
Technical Field
The invention relates to a quantitative evaluation method for smoke screen interference efficiency, in particular to a quantitative evaluation method for smoke screen interference efficiency from the perspective of target capture, tracking and concerned image characteristics.
Background
The photoelectric detection and guidance device mainly images a target area through a photoelectric detector and then transmits the image to an information processor, and the information processor identifies the target from a background image for capturing and tracking by utilizing the image characteristics of the target based on priori knowledge or a deep learning mode.
The smoke screen interference is an effective means for implementing interference on photoelectric reconnaissance and guidance equipment in a modern battlefield, has the advantages of simple application mode, low cost and the like, is applied to a large amount of equipment in all countries and military parties, but currently, few effect evaluation means aiming at the smoke screen interference photoelectric reconnaissance and guidance equipment are mainly qualitative evaluation of power criteria, probability criteria and the like, and small part quantitative evaluation methods of shielding area, shielding time and the like, and the method is short of quantitative evaluation methods from the aspect of image characteristics of target capture tracking attention.
Disclosure of Invention
The technical problem to be solved by the invention is as follows: methods for quantitatively assessing smoke disturbance from the perspective of target capture tracking of image features of interest are currently lacking. The invention provides a feature fusion smoke screen interference efficiency evaluation method based on a direction gradient histogram and electronic equipment, which quantitatively evaluate the interference efficiency of a smoke screen on photoelectric imaging reconnaissance guidance equipment from the angle of image features concerned by image tracking.
In order to solve the technical problems, the technical scheme adopted by the invention is as follows:
the method for evaluating the interference efficiency of the feature fusion smoke screen based on the direction gradient histogram comprises the following steps:
step 1, collecting an image of a target area before smoke screen interference as a reference image R, and collecting an image of the target area after smoke screen interference as an interfered image D.
Step 2, respectively calculating referenceDirectional gradient histogram feature Hog of image and disturbed image R ，Hog D 。
And 3, performing Riss transformation based on the directional gradient histogram characteristics of the reference image and the interfered image, and solving a similarity matrix of the reference image and the interfered image.
And 4, solving a cosine similarity matrix based on the directional gradient histogram characteristics of the reference image and the interfered image.
And 5, solving a correlation coefficient matrix based on the directional gradient histogram characteristics of the reference image and the interfered image.
And 6, fusing the Rich transform similarity matrix, the cosine similarity matrix and the correlation coefficient matrix obtained by the calculation to obtain an image quality evaluation value under smoke screen interference.
And 7, calculating and setting a tracking drift threshold, and performing feature amplification processing on the fused features by using the threshold.
And 8, analyzing the relation between the image quality evaluation value and the smoke interference efficiency evaluation value under the smoke interference, and calculating to obtain the smoke interference efficiency evaluation value.
The present invention also provides an electronic device comprising a memory configured to store executable instructions; the processor is configured to execute the executable instructions stored in the memory to implement the above-described method for estimating and processing the smoke screen interference performance based on feature fusion of the histogram of oriented gradients.
With the development of target tracking algorithms, the calculation speed and the calculation precision of the target tracking method based on the relevant filtering are greatly improved under the addition of the HOG characteristics. Wherein, because the HOG is operated on the local grid cells of the image, the HOG can keep good invariance to the geometric and optical deformation of the image, and the obtained characteristic can be more compact due to the characteristic of blocking processing. Secondly, HOG describes the edge structure features, and can fully describe the structure information of the object. That is, the HOG feature is robust to motion blur, illumination transformation, color transformation, and the like, but less robust to deformation. And based on HOG characteristics, the image characteristic gradient similarity matrix is obtained by introducing an MS theory and using the Rich transform, so that the detection of the image edge characteristics can be enhanced. And because the cosine similarity has the characteristics of sensitivity to vector direction and insensitivity to logarithmic absolute value, the robustness of feature detection can be enhanced, and good detection precision is still maintained in a high-dimensional space. And the correlation coefficient is used for reflecting the degree of closeness of correlation of the HOG characteristics before and after the target interference, and directly reflecting the difference of the target characteristics before and after the smoke screen interference from the angle of the characteristic gray value.
According to the method, the edge and structure information of the target can be focused more by fusing the Rich transform similarity matrix, the cosine similarity matrix and the correlation coefficient matrix obtained by the directional gradient histogram characteristics, and good invariance and robustness to image geometry and optical deformation can be kept.
In the invention, the tracking drift threshold is calculated and set in the step 7, and the threshold is used for carrying out feature amplification processing on the fused features, so that the smoke screen interference evaluation value of the serious smoke screen interference part can be increased, and the interference effect of the smoke screen is more prominent. Meanwhile, the amplified smoke screen interference evaluation value is beneficial to better combining the objective evaluation value with the subjective evaluation value, and a qualitative evaluation result which is more visual and has actual reference value is obtained.
Therefore, the method can quantitatively evaluate the interference efficiency of the smoke screen on the photoelectric imaging reconnaissance guidance equipment from the angle of image characteristics focused by image tracking, focuses on the structure and edge information of an image target, and fuses the HOG characteristics with the Rich transform similarity matrix, the cosine similarity matrix and the correlation coefficient matrix, so that the evaluation result can keep good invariance to the geometric and optical deformation of the image, the robustness of the evaluation result on the deformation of the target is enhanced, and the evaluation accuracy is improved.
Drawings
FIG. 1 is a flow chart of a method for evaluating interference performance of a feature fusion smoke screen based on histogram of oriented gradients according to the present invention;
FIG. 2 is a histogram of oriented gradient features of reference and interference image extraction;
FIG. 3 is a representation of the Rich transform space;
FIG. 4 is a graphical illustration of the criticality of smoke screen interference on tracking effect.
Detailed Description
Specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.
As shown in fig. 1, the method for evaluating interference performance of feature fusion smoke screen based on histogram of oriented gradient according to the present invention comprises the steps of:
step 1, collecting an image of a target area before smoke interference as a reference image, and collecting an image of the target area after the smoke interference as an interfered image; let the reference image be R and the interference image be D.
Step 2, firstly, the directional gradient histogram characteristics of the images R and D are respectively solved to obtain Hog R ，Hog D As shown in fig. 2, the upper line in the drawing is a reference image, the lower line is an interference image, and the original image, the image normalization, the directional gradient map, and the directional gradient histogram are sequentially represented in each line from left to right.
Step 3, histogram feature Hog of the directional gradient R ，Hog D By a Rich transformation to give [ R, R 1 ,R 2 ]And [ D, D 1 ,D 2 ]Using the local characteristic similarity functions (local amplitude a, local direction θ and local phase ψ), the MS similarity matrix S is obtained M (R,D)。
Fig. 3 shows a spherical coordinate system of the rieus transformation space in a three-dimensional euclidean space, and R, R, R2 are projections of points in the spherical coordinate system on three axes. In this spatial domain, the local amplitude a, the local direction θ and the local phase ψ can be expressed as:
wherein, the first and the second end of the pipe are connected with each other,
θ R (x,y)∈[0,π)，ψ R (x,y)∈[0,π)。
the MS similarity of the images R and D at the pixel point (x, y) is:
wherein C 1 Is a very small positive number, preventing instability due to a denominator of zero or close to zero.
Constructional parameter S M As MS similarity matrix:
S M ＝S A ·S θ ·S ψ 。
step 4, histogram feature Hog of directional gradient R ，Hog D Respectively converted into one-dimensional arrays H of size MXN R ,H D The cosine similarity of the images R and D is:
step 5, histogram feature Hog of directional gradient R ，Hog D And (3) performing autocorrelation calculation to obtain a correlation coefficient matrix:
wherein the content of the first and second substances,
m and N are the size of the matrix.
Step 6, performing weighted fusion on the MS similarity matrix, the cosine similarity matrix and the correlation coefficient matrix obtained by the calculation to obtain an image quality evaluation index IQ _ EEMSSJ under smoke screen interference, wherein the value range is [0,1]:
IQ_EEMSSJ＝S M (R,D) α *CosSim(R,D) β *Corr(R,D) τ
where α, β, τ are fusion weight values of the three features, and α = β = τ =1 is set.
Step 7, selecting an interfered image in a smoke screen interference state when the target tracks and drifts, and calculating smoke screen interference pair tracking and tracking according to the calculation stepMarginal IQ _ EEMSSJ threshold for trace effects
If the IQ _ EEMSSJ value calculated by the interfered image is larger than the threshold value
The infrared seeker can effectively track the target under the smoke screen interference state, namely the smoke screen interference does not cause effective shielding influence on target tracking, the IQ _ EEMSSJ value is amplified to obtain EEMSSJ _ aug, and the image quality is actively enhanced.
If the IQ _ EEMSSJ value calculated by the interfered image is less than the threshold value
The infrared seeker cannot effectively track the target under the smoke interference state, namely the smoke interference effectively shields the target tracking, and the image quality is not actively enhanced.
Fig. 4 shows a diagram of critical states of smoke interference on tracking, wherein the tracking algorithm can stably track in fig. 1 and 3, and the tracking algorithm drifts in fig. 2 and 4. The 2 nd graph and the 4 th graph are the forms of the next frames of the 1 st graph and the 3 rd graph in the smoke interference process respectively, and it can be seen from the graphs that the smoke just effectively shields the target, resulting in tracking drift. Therefore, the frame image is selected as a critical state diagram to calculate the threshold value of IQ _ EEMSSJ
Step 8, evaluating smoke interference efficiency:
since the EEMSSJ _ aug index at this time is an evaluation of the image quality, and the smoke interference effect value is opposite to the image quality, i.e., the better the image quality, the worse the smoke interference effect, and the worse the image quality, the better the smoke interference effect. Therefore, the estimation of the smoke screen interference performance is expressed by the following formula, and the value range is [0,1]:
EEMSSJ＝1-EEMSSJ_aug。
Claims (9)
1. the method for evaluating and processing the interference efficiency of the feature fusion smoke screen based on the direction gradient histogram is characterized by comprising the following steps of:
step 1, collecting an image of a target area before smoke screen interference as a reference image R, and collecting an image of the target area after smoke screen interference as an interfered image D;
step 2, respectively calculating directional gradient histogram characteristics Hog of the reference image and the interfered image R ，Hog D ；
Step 3, performing Riss transformation based on the directional gradient histogram characteristics of the reference image and the interfered image, and solving a similar matrix of the reference image and the interfered image;
step 4, solving a cosine similarity matrix based on the directional gradient histogram characteristics of the reference image and the interfered image;
step 5, a correlation coefficient matrix is obtained based on the directional gradient histogram characteristics of the reference image and the interfered image;
step 6, fusing the calculated Riss transformation similarity matrix, cosine similarity matrix and correlation coefficient matrix to obtain a smoke screen interference efficiency evaluation value;
and 7, calculating and setting a tracking drift threshold value, and performing feature amplification processing on the fused features by using the tracking drift threshold value.
2. The method for estimating and processing smoke screen interference performance based on histogram of oriented gradients as claimed in claim 1, wherein the step 3 comprises:
characterizing histogram of oriented gradients by Hog R ，Hog D Performing a Reed-Solomon transformation to obtain [ R, R 1 ,R 2 ]And [ D, D 1 ,D 2 ]Obtaining local amplitude A, local direction theta and local phase psi by using similar function of local characteristicsMS similarity matrix S M (R,D)。
3. The histogram of oriented gradient-based feature fusion smoke screen interference performance assessment and processing method according to claim 2, wherein in the spatial domain, the local amplitude a, the local direction θ and the local phase ψ are expressed as:
wherein the content of the first and second substances,
θ R (x,y)∈[0,π)，ψ R (x,y)∈[0,π)。
the MS similarity of the images R and D at the pixel point (x, y) is:
wherein C is 1 Is a very small positive number;
constructional parameter S M As MS similarity matrix:
S M ＝S A ·S θ ·S ψ 。
4. the histogram of oriented gradients-based feature fusion smoke screen interference performance evaluation and processing method of claim 2, wherein the step 4 comprises:
characterizing histogram of oriented gradients by Hog R 、Hog D Respectively converted into one-dimensional arrays H of size MXN R ,H D The cosine similarity of the images R and D is:
5. the method according to claim 2, wherein the step 5 comprises:
characterizing histogram of oriented gradients by Hog R ，Hog D And (3) performing autocorrelation calculation to obtain a correlation coefficient matrix:
wherein the content of the first and second substances,
m and N are the size of the matrix.
6. The histogram of oriented gradients-based feature fusion smoke screen interference performance assessment and processing method according to claim 2, wherein the step 6 comprises:
weighting and fusing the MS similarity matrix, the cosine similarity matrix and the correlation coefficient matrix obtained by calculation to obtain an image quality evaluation index IQ _ EEMSSJ under smoke screen interference, wherein the value range is [0,1]:
IQ_EEMSSJ＝S M (R,D) α *CosSim(R,D) β *Corr(R,D) τ
where α, β, τ are fusion weight values of the three features, and α = β = τ =1 is set.
7. The method according to claim 2, wherein the step 7 comprises:
selecting an interfered image in a smoke screen interference state when the target is tracked and drifted, and calculating a critical IQ _ EEMSSJ threshold value of the smoke screen interference on tracking influence by using the methods of the steps 1 to 6
And makes the image quality evaluation index IQ _ EEMSSJ asThe following calculation processing:
if the IQ _ EEMSSJ value calculated by the interfered image is larger than the threshold value
Amplifying the IQ _ EEMSSJ value to obtain EEMSSJ _ aug, and actively enhancing the image quality;
if the IQ _ EEMSSJ value calculated by the interfered image is smaller than the threshold value
No active enhancement of the image quality is performed.
8. The histogram of oriented gradients based feature fusion smoke interference performance evaluation and processing method of claim 2, further comprising step 8 of evaluating smoke interference performance: the estimation of the smoke screen interference efficiency is represented by the following formula, and the value range is [0,1]:
EEMSSJ＝1-EEMSSJ_aug。
9. an electronic device, characterized in that: comprising a memory configured to store executable instructions; a processor configured to execute executable instructions stored in the memory to implement the method for estimating and processing smoke screen interference performance based on histogram of oriented gradients according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210956882.3A CN115345845A (en) | 2022-08-10 | 2022-08-10 | Feature fusion smoke screen interference efficiency evaluation and processing method based on direction gradient histogram and electronic equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210956882.3A CN115345845A (en) | 2022-08-10 | 2022-08-10 | Feature fusion smoke screen interference efficiency evaluation and processing method based on direction gradient histogram and electronic equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115345845A true CN115345845A (en) | 2022-11-15 |
Family
ID=83952392
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210956882.3A Pending CN115345845A (en) | 2022-08-10 | 2022-08-10 | Feature fusion smoke screen interference efficiency evaluation and processing method based on direction gradient histogram and electronic equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115345845A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117058670A (en) * | 2023-10-12 | 2023-11-14 | 深圳市华加生物科技有限公司 | Electronic tobacco tar flexibility evaluation method |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107507140A (en) * | 2017-08-03 | 2017-12-22 | 重庆大学 | The outdoor scene vehicle shadow disturbance restraining method of highway of feature based fusion |
| CN108765468A (en) * | 2018-05-30 | 2018-11-06 | 中国石油大学(华东) | A kind of method for tracking target and device of feature based fusion |
| CN110569723A (en) * | 2019-08-02 | 2019-12-13 | 西安工业大学 | Target tracking method combining feature fusion and model updating |
| CN111127387A (en) * | 2019-07-11 | 2020-05-08 | 宁夏大学 | Method for evaluating quality of non-reference image |
| CN111242912A (en) * | 2020-01-08 | 2020-06-05 | 北京电子工程总体研究所 | Smoke screen interference efficiency acquisition method |
| CN113610862A (en) * | 2021-07-22 | 2021-11-05 | 东华理工大学 | Screen content image quality evaluation method |
-
2022
- 2022-08-10 CN CN202210956882.3A patent/CN115345845A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107507140A (en) * | 2017-08-03 | 2017-12-22 | 重庆大学 | The outdoor scene vehicle shadow disturbance restraining method of highway of feature based fusion |
| CN108765468A (en) * | 2018-05-30 | 2018-11-06 | 中国石油大学(华东) | A kind of method for tracking target and device of feature based fusion |
| CN111127387A (en) * | 2019-07-11 | 2020-05-08 | 宁夏大学 | Method for evaluating quality of non-reference image |
| CN110569723A (en) * | 2019-08-02 | 2019-12-13 | 西安工业大学 | Target tracking method combining feature fusion and model updating |
| CN111242912A (en) * | 2020-01-08 | 2020-06-05 | 北京电子工程总体研究所 | Smoke screen interference efficiency acquisition method |
| CN113610862A (en) * | 2021-07-22 | 2021-11-05 | 东华理工大学 | Screen content image quality evaluation method |
Non-Patent Citations (1)
| Title |
|---|
| GUANGYI YANG: "RVSIM: a feature similarity method for full-reference image quality assessment", EURASIP JOURNAL ON IMAGE AND VIDEO PROCESSING, vol. 6, 31 December 2018 (2018-12-31), pages 1 - 15 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117058670A (en) * | 2023-10-12 | 2023-11-14 | 深圳市华加生物科技有限公司 | Electronic tobacco tar flexibility evaluation method |
| CN117058670B (en) * | 2023-10-12 | 2024-01-02 | 深圳市华加生物科技有限公司 | Electronic tobacco tar flexibility evaluation method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Xuan et al. | Object tracking in satellite videos by improved correlation filters with motion estimations | |
| CN110728697B (en) | Infrared dim target detection tracking method based on convolutional neural network | |
| CN112800964B (en) | Remote sensing image target detection method and system based on multi-module fusion | |
| CN110033472B (en) | Stable target tracking method in complex infrared ground environment | |
| Xue et al. | Low-rank approximation and multiple sparse constraint modeling for infrared low-flying fixed-wing UAV detection | |
| CN111429485B (en) | Cross-modal filtering tracking method based on self-adaptive regularization and high-reliability updating | |
| Guo et al. | An anchor-free network with density map and attention mechanism for multiscale object detection in aerial images | |
| CN115345845A (en) | Feature fusion smoke screen interference efficiency evaluation and processing method based on direction gradient histogram and electronic equipment | |
| CN116823694B (en) | Infrared and visible light image fusion method and system based on multi-focus information integration | |
| Ma et al. | MSMA-Net: An Infrared Small Target Detection Network by Multi-scale Super-resolution Enhancement and Multi-level Attention Fusion | |
| Zhao et al. | Deep learning-based laser and infrared composite imaging for armor target identification and segmentation in complex battlefield environments | |
| CN111127355A (en) | Method for finely complementing defective light flow graph and application thereof | |
| CN111091586A (en) | Rapid smoke dynamic shielding area detection and positioning method and application thereof | |
| CN116385915A (en) | Water surface floater target detection and tracking method based on space-time information fusion | |
| CN114842506A (en) | Human body posture estimation method and system | |
| Qian et al. | STRCFD: Small maneuvering object tracking via improved STRCF and redetection in near infrared videos | |
| Gao et al. | Aero-optical image and video restoration based on mean filter and adversarial network | |
| CN117523428B (en) | Ground target detection method and device based on aircraft platform | |
| Lu et al. | A Sar Image registration method based on SIFT Algorithm | |
| CN114648564B (en) | Visible light and infrared image optimization registration method and system for unsteady state target | |
| CN111476821B (en) | Target tracking method based on online learning | |
| Wu et al. | An Infrared Target Images Recognition and Processing Method Based on the Fuzzy Comprehensive Evaluation | |
| Zhang et al. | Research on binocular real-time ranging method in window area | |
| Mou et al. | Saliency detection of underwater target based on spatial probability | |
| Bai et al. | Robust Tracking for Motion Blur Based on Correlation Filter |
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 |