CN112419199B - Patrol car anti-interference method based on Landweber smooth projection - Google Patents
Patrol car anti-interference method based on Landweber smooth projection Download PDFInfo
- Publication number
- CN112419199B CN112419199B CN202011405911.4A CN202011405911A CN112419199B CN 112419199 B CN112419199 B CN 112419199B CN 202011405911 A CN202011405911 A CN 202011405911A CN 112419199 B CN112419199 B CN 112419199B
- Authority
- CN
- China
- Prior art keywords
- patrol car
- projection
- landweber
- image
- patrol
- 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.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 22
- 239000011159 matrix material Substances 0.000 claims description 18
- 238000002945 steepest descent method Methods 0.000 claims description 12
- 230000003044 adaptive effect Effects 0.000 claims description 6
- 238000004364 calculation method Methods 0.000 claims description 6
- 230000009466 transformation Effects 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 4
- 238000000638 solvent extraction Methods 0.000 claims description 3
- 238000005286 illumination Methods 0.000 abstract description 6
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- 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
- G06T2207/20021—Dividing image into blocks, subimages or windows
-
- 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
- G06T2207/20024—Filtering details
-
- 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
- G06T2207/20064—Wavelet transform [DWT]
-
- 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
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30248—Vehicle exterior or interior
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Processing (AREA)
Abstract
The invention designs a patrol car anti-interference method based on Landweber smooth projection. The method aims at solving the problems that when the patrol car collects images, the patrol car is affected by illumination, physical shaking of the car body and the like, so that shaking and non-uniform illumination noise can occur in the collected images of the patrol car, and the decision accuracy of the patrol car can be further affected. The invention provides a patrol car anti-interference method based on Landweber smooth projection, which utilizes a gyroscope sensor to detect the real-time shaking angle of the patrol car, uses the Landweber projection method and natural image sparse characteristics to establish a patrol car anti-interference system taking Landweber projection and image sparse representation as cores, eliminates interference noise such as shaking and non-uniform illumination of images collected by the patrol car, improves intelligent decision precision of patrol, and increases anti-interference capability of the patrol car.
Description
Technical Field
The invention relates to the fields of system anti-interference and the like, in particular to a patrol car anti-interference method based on Landweber smooth projection.
Background
With the continuous development of society, more and more large-scale enterprise factory buildings use automatic patrol cars to replace manual patrol, so that patrol efficiency can be greatly improved, various conditions can be frequently met in the automatic patrol process, the patrol cars are affected by problems such as illumination and physical shake of the car body when images are collected, shake and uneven illumination noise can occur in collected images of the patrol cars, and decision accuracy of the patrol cars can be further affected. Aiming at the problems, the invention provides the patrol car anti-interference method based on Landweber smooth projection, which eliminates interference noise such as shake and uneven illumination of images collected by the patrol car, improves intelligent decision accuracy of patrol, and increases anti-interference capability of the patrol car.
Disclosure of Invention
In order to solve the above-mentioned problems. The invention provides a patrol car anti-interference method based on Landweber smooth projection. To achieve this object:
The invention provides a patrol car anti-interference method based on Landweber smooth projection, which is characterized by comprising the following steps of:
Step 1: collecting real-time shake angles of the patrol car by using a gyroscope sensor;
step 2: uniformly partitioning the acquired image x into blocks with the size of 256 multiplied by 256;
Step 3: establishing a random matrix theta, and unitizing each column of the random matrix theta, wherein the size of the random matrix is 64 multiplied by 256;
step 4: projecting the acquired image block to a low-dimensional space w using a random projection matrix;
Step 5: estimating the image block by using a back projection method to obtain a prediction block xx;
Step 6: denoising the predicted image block xx by using a wiener filter to obtain a denoised predicted block xx0;
step 7: predicting an image by using a steepest descent method to obtain xx1;
Step 8: performing wavelet transformation on the prediction block xx1 to obtain a wavelet coefficient d, and performing adaptive threshold denoising on the wavelet transformation coefficient to obtain a denoised wavelet coefficient ds;
step 9: carrying out wavelet reconstruction on the denoised wavelet coefficient ds to obtain a wavelet reconstruction image xx2;
Step 10: predicting the image by using a steepest descent method to obtain xx3;
Step 11: judging whether the image prediction result is converged, if so, denoising the image block to be xx3, otherwise, making xx=xx3, and continuously repeating the steps 6-10.
As a further improvement of the present invention, the calculation formula of the linear projection of the image block to the low-dimensional space w in the step 4 is as follows:
w＝Θx (1)
Where Θ is the random projection matrix.
As a further improvement of the present invention, the calculation formula of the back projection in the step 5 is as follows:
xx＝ΘTw (2)
where Θ T represents the transpose of the matrix Θ.
As a further improvement of the invention, the step 6 wiener filtering is expressed as:
xx0＝Wiener(xx,σ) (3)
Wherein Wiener (·) is a Wiener filter function, θ is the patrol car shake angle measured in step 1.
As a further improvement of the present invention, the formula of the steepest descent method in step 7 is as follows:
Where (Θ T)-1 represents the inverse of Θ T.
As a further improvement of the present invention, the adaptive threshold denoising formula in step 8 is:
λ＝median(d) (8)
Wherein media (·) is the median function.
As a further improvement of the present invention, the formula of the steepest descent method in the step 10 is as follows:
wherein, Consistent with the parameters of step 7.
As a further improvement of the present invention, the step 11 specifically includes:
where D is 1, which means that there is no convergence, otherwise, convergence, ε is the convergence threshold.
The patrol car anti-interference method based on Landweber smooth projection has the beneficial effects that:
1. the invention utilizes wiener filtering to increase the robustness of the system.
2. The invention utilizes Landweber smooth projection to increase the fidelity of the signal.
3. The method has low algorithm complexity and strong real-time performance.
4. The hardware system of the invention has simple realization and low cost.
Drawings
FIG. 1 is a flow chart of a system;
Detailed Description
The invention provides a patrol car anti-interference method based on Landweber smooth projection.
The invention is further described below with reference to the drawings and detailed description:
as shown in fig. 1, first, a real-time shake angle of a patrol car is acquired using a gyro sensor; uniformly partitioning the acquired image x into blocks with the size of 256 multiplied by 256; establishing a random matrix theta, and unitizing each column of the random matrix theta, wherein the size of the random matrix is 64 multiplied by 256; then, using a random projection matrix, projecting the acquired image block into a low-dimensional space w;
the calculation formula of the linear projection of the image block to the low-dimensional space w is as follows:
w＝Θx (1)
Where Θ is the random projection matrix.
Then, estimating the image block by using a back projection method to obtain a prediction block xx; denoising the predicted image block xx by using a wiener filter to obtain a denoised predicted block xx0; in addition, predicting an image by using a steepest descent method to obtain xx1;
The back projection calculation formula is:
xx＝ΘTw (2)
where Θ T represents the transpose of the matrix Θ.
Wiener filtering is expressed as:
xx0＝Wiener(xx,σ) (3)
Wherein Wiener (·) is a Wiener filter function, θ is the patrol car shake angle measured in step 1.
The formula of the steepest descent method is as follows:
Where (Θ T)-1 represents the inverse of Θ T.
Finally, carrying out wavelet transformation on the prediction block xx1 to obtain a wavelet coefficient d, and carrying out adaptive threshold denoising on the wavelet transformation coefficient to obtain a denoised wavelet coefficient ds; in addition, carrying out wavelet reconstruction on the denoised wavelet coefficient ds to obtain a wavelet reconstruction image xx2; predicting the image by using a steepest descent method to obtain xx3; judging whether the image prediction result is converged, if so, denoising the image block to be xx3, otherwise, enabling xx=xx3, and continuing updating iteration.
The adaptive threshold denoising formula is:
λ＝median(d) (8)
Wherein media (·) is the median function.
The formula of the steepest descent method is as follows:
wherein, Consistent with the parameters of step 7.
The step 11 specifically comprises the following steps:
where D is 1, which means that there is no convergence, otherwise, convergence, ε is the convergence threshold.
The above description is only of the preferred embodiment of the present invention, and is not intended to limit the present invention in any other way, but is intended to cover any modifications or equivalent variations according to the technical spirit of the present invention, which fall within the scope of the present invention as defined by the appended claims.
Claims (3)
1. A patrol car anti-interference method based on Landweber smooth projection comprises the following specific steps:
Step 1: collecting a real-time shaking angle theta of the patrol car by using a gyroscope sensor;
step 2: uniformly partitioning the acquired image x into blocks with the size of 256 multiplied by 256;
Step 3: establishing a random matrix theta, and unitizing each column of the random matrix theta, wherein the size of the random matrix is 64 multiplied by 256;
step 4: projecting the acquired image block to a low-dimensional space w using a random projection matrix;
The calculation formula of the image block linear projection to the low-dimensional space w in the step4 is as follows:
w＝Θx (1)
Wherein Θ is a random projection matrix;
Step 5: estimating the image block by using a back projection method to obtain a prediction block xx;
Step 6: denoising the predicted image block xx by using a wiener filter to obtain a denoised predicted block xx0;
the step 6 wiener filtering is expressed as:
xx0＝Wiener(xx,σ) (3)
wherein Wiener (·) is a Wiener filter function, θ is the patrol car shake angle measured in step 1;
step 7: predicting an image by using a steepest descent method to obtain xx1;
The formula of the steepest descent method in the step 7 is as follows:
Wherein (Θ T)-1 represents the inverse of Θ T;
Step 8: performing wavelet transformation on the prediction block xx1 to obtain a wavelet coefficient d, and performing adaptive threshold denoising on the wavelet transformation coefficient to obtain a denoised wavelet coefficient ds;
the adaptive threshold denoising formula in the step 8 is as follows:
λ＝median(d) (8)
Wherein media (·) is the median function;
step 9: carrying out wavelet reconstruction on the denoised wavelet coefficient ds to obtain a wavelet reconstruction image xx2;
Step 10: predicting the image by using a steepest descent method to obtain xx3;
the formula of the steepest descent method in the step 10 is as follows:
wherein, Consistent with the parameters of the step 7;
Step 11: judging whether the image prediction result is converged, if so, denoising the image block to be xx3, otherwise, making xx=xx3, and continuously repeating the steps 6-10.
2. The patrol car anti-interference method based on Landweber smooth projection according to claim 1, wherein the method comprises the following steps:
The back projection calculation formula in the step 5 is as follows:
xx＝ΘTw (2)
where Θ T represents the transpose of the matrix Θ.
3. The patrol car anti-interference method based on Landweber smooth projection according to claim 1, wherein the method comprises the following steps:
The step 11 specifically comprises the following steps:
where D is 1, which means that there is no convergence, otherwise, convergence, ε is the convergence threshold.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011405911.4A CN112419199B (en) | 2020-12-04 | 2020-12-04 | Patrol car anti-interference method based on Landweber smooth projection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011405911.4A CN112419199B (en) | 2020-12-04 | 2020-12-04 | Patrol car anti-interference method based on Landweber smooth projection |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112419199A CN112419199A (en) | 2021-02-26 |
| CN112419199B true CN112419199B (en) | 2024-10-25 |
Family
ID=74830212
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011405911.4A Active CN112419199B (en) | 2020-12-04 | 2020-12-04 | Patrol car anti-interference method based on Landweber smooth projection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112419199B (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104574338A (en) * | 2015-01-26 | 2015-04-29 | 西安交通大学 | Remote sensing image super-resolution reconstruction method based on multi-angle linear array CCD sensors |
| CN109982090A (en) * | 2019-03-21 | 2019-07-05 | 西安电子科技大学 | A kind of adaptive splits' positions cognitive method of sample rate of combination gray level entropy and blind deconvolution |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2008172321A (en) * | 2007-01-09 | 2008-07-24 | Olympus Imaging Corp | Image pickup device for performing electric image recovery processing |
| US9445115B2 (en) * | 2013-11-21 | 2016-09-13 | Bae Systems Information And Electronic Systems Integration Inc. | Coded image system and method thereof |
| CN110264415B (en) * | 2019-05-24 | 2020-06-12 | 北京爱诺斯科技有限公司 | Image processing method for eliminating jitter blur |
-
2020
- 2020-12-04 CN CN202011405911.4A patent/CN112419199B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104574338A (en) * | 2015-01-26 | 2015-04-29 | 西安交通大学 | Remote sensing image super-resolution reconstruction method based on multi-angle linear array CCD sensors |
| CN109982090A (en) * | 2019-03-21 | 2019-07-05 | 西安电子科技大学 | A kind of adaptive splits' positions cognitive method of sample rate of combination gray level entropy and blind deconvolution |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112419199A (en) | 2021-02-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Jin et al. | Adaptive Wiener filtering of noisy images and image sequences | |
| CN102968770A (en) | Method and device for eliminating noise | |
| CN103217409A (en) | Raman spectral preprocessing method | |
| CN103024248A (en) | Motion-adaptive video image denoising method and device | |
| CN101916433A (en) | Denoising method of strong noise pollution image on basis of partial differential equation | |
| CN107818547B (en) | A kind of minimizing technology towards the spiced salt and Gaussian mixed noise in twilight image sequence | |
| CN117527570B (en) | Sensor cluster position optimization method based on edge reinforcement learning | |
| CN113643201A (en) | Image denoising method of self-adaptive non-local mean value | |
| CN112541869A (en) | Retinex image defogging method based on matlab | |
| CN112419199B (en) | Patrol car anti-interference method based on Landweber smooth projection | |
| CN112327259A (en) | Method and device for eliminating interference signals in SAR image | |
| CN117056677A (en) | Transient electromagnetic signal denoising method for improving variational modal decomposition based on sparrow algorithm | |
| CN110657807B (en) | Indoor positioning displacement measurement method for detecting discontinuity based on wavelet transformation | |
| CN111598793A (en) | Method and system for defogging image of power transmission line and storage medium | |
| CN102314675B (en) | Wavelet high-frequency-based Bayesian denoising method | |
| CN115409872A (en) | Underwater camera image optimization method | |
| CN102143303A (en) | Image denoising method in transmission line intelligent monitoring system | |
| CN111641825A (en) | 3D denoising method and denoising device embedded into HEVC (high efficiency video coding) coding process | |
| CN109087257B (en) | Airspace increment image filtering method based on parameter estimation framework | |
| CN102572201A (en) | Method and system for removing overlapped curves from image | |
| CN111628750B (en) | Nonlinear filtering method for random resonance matching in trap | |
| CN109859128B (en) | Interaction system switching filtering method based on Bayesian estimation switching rule | |
| CN111638501A (en) | Spectral line enhancement method for self-adaptive matching stochastic resonance | |
| CN108122411B (en) | Tracking frequency self-adaptive optimization method based on vehicle speed prediction | |
| CN113780301A (en) | Self-adaptive denoising machine learning application method for defending against attack |
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 |