CN106097248B - High-resolution image knowledge prior-based compressed sensing method and mixed vision system thereof - Google Patents
High-resolution image knowledge prior-based compressed sensing method and mixed vision system thereof Download PDFInfo
- Publication number
- CN106097248B CN106097248B CN201610428922.1A CN201610428922A CN106097248B CN 106097248 B CN106097248 B CN 106097248B CN 201610428922 A CN201610428922 A CN 201610428922A CN 106097248 B CN106097248 B CN 106097248B
- Authority
- CN
- China
- Prior art keywords
- dimensional image
- dimension
- image
- camera
- calculating
- 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 17
- 238000003384 imaging method Methods 0.000 claims abstract description 17
- 239000011159 matrix material Substances 0.000 claims abstract description 10
- 239000011521 glass Substances 0.000 claims description 6
- 230000000007 visual effect Effects 0.000 abstract description 11
- 238000005070 sampling Methods 0.000 abstract description 7
- 230000006835 compression Effects 0.000 description 3
- 238000007906 compression Methods 0.000 description 3
- 238000013507 mapping Methods 0.000 description 2
- 230000008447 perception Effects 0.000 description 2
- 230000001174 ascending effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000009977 dual effect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000007613 environmental effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 238000013341 scale-up Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformations in the plane of the image
- G06T3/40—Scaling of whole images or parts thereof, e.g. expanding or contracting
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
The invention belongs to the technical field of machine vision, and particularly relates to a high-resolution image knowledge prior-based compressed sensing method and a mixed vision system thereof. The invention comprises the following steps: setting the middle dimension of a high-dimensional image and a low-dimensional image in a hybrid vision system; calculating a high-dimensional image dimension reduction observation matrix, reducing the dimension of the high-dimensional image, calculating a low-dimensional image dimension increasing observation matrix, and increasing the dimension of the low-dimensional image; matching the two obtained pictures with the same dimensionality by using an SIFT matching algorithm and calculating a matching rate; repeating the steps until the optimal middle dimensionality which can obtain the highest matching rate is found. The method uses an image processing means, and aims at the characteristic that the target has imaging scale deviation in different visual primitives of a mixed visual system, and adopts a compression-sensing-based technology to reconstruct a panoramic nonlinear compressed image and perform dimensionality reduction on a perspective image based on a down-sampling measure.
Description
Technical Field
The invention belongs to the technical field of machine vision, and particularly relates to a high-resolution image knowledge prior-based compressed sensing method and a mixed vision system thereof.
Background
The vision technology has the characteristics of non-contact perception, rich acquired information quantity, strong anti-interference capability and the like, and is widely applied to the fields of environment understanding, target detection, positioning and the like. The conventional binocular vision and binocular panoramic vision have the defects of narrow visual field and limited action distance when the target is positioned, the heterogeneous dual-scale hybrid vision system formed by combining the conventional binocular vision and the conventional binocular panoramic vision gives consideration to the dual factors of the visual field and the visual distance, and the application fields of target tracking and positioning are realized in a large visual field range. However, the imaging principle of the visual primitives in the mixing system is different and the imaging scale of the target is different, so that the classical binocular vision positioning algorithm is invalid. The method effectively solves the problem of equal-dimension reconstruction of images among images with different scales in a multi-view stereoscopic vision system consisting of heterogeneous imaging visual units, and is the most basic and important technology for realizing environment three-dimensional information perception based on machine vision.
The classical shannon sampling theorem states that in order to recover the analog signal without distortion, the sampling frequency should be no less than twice the nyquist frequency (i.e., the highest frequency in the analog signal spectrum). But where no a priori information is utilized other than the assumption that the signal is of limited bandwidth. There is a large degree of redundancy in the collected data. The Compressed Sensing method (Compressed Sensing CS) fully utilizes the prior information that most signals can be sparsely represented on a set of predicted bases, and provides a new idea for the reconstruction of dimensions.
Disclosure of Invention
The invention aims to provide a high-resolution image knowledge prior-based compressed sensing method applying a machine vision means.
The invention also aims to provide a hybrid vision system based on the high-resolution image knowledge prior compressed sensing method.
The purpose of the invention is realized as follows:
A high-resolution image knowledge prior-based compressed sensing method comprises the following steps:
1) setting the middle dimension of a high-dimensional image and a low-dimensional image in a hybrid vision system;
2) Calculating a high-dimensional image dimension reduction observation matrix, reducing the dimension of the high-dimensional image, calculating a low-dimensional image dimension increasing observation matrix, and increasing the dimension of the low-dimensional image;
3) Matching the two obtained pictures with the same dimensionality by using an SIFT matching algorithm and calculating a matching rate;
4) Repeating the steps until the optimal middle dimensionality which can obtain the highest matching rate is found.
A hybrid vision system based on a high-resolution image knowledge priori compression sensing method comprises a panoramic camera at the upper half part and a perspective camera at the lower half part, wherein the panoramic camera comprises a hyperboloid all-dimensional imaging reflector 1 which is supported on a panoramic camera support base 3 through a panoramic camera annular transparent glass supporting cylinder 2, a 1394 camera 4 is vertically and upwards arranged on the panoramic camera support base 3 in the panoramic camera annular transparent glass supporting cylinder 2 below the all-dimensional imaging reflector 1, the 1394 camera 5 serving as the perspective camera is vertically connected with a rotatable connecting rod 6 and is vertically connected on the panoramic camera support base 3, and the perspective camera annular transparent glass supporting cylinder 7 supports the panoramic camera and is arranged on a perspective camera support base 8.
The invention has the beneficial effects that:
the method uses an image processing means, and aims at the characteristic that the target has imaging scale deviation in different visual primitives of a mixed visual system, and adopts a compression-sensing-based technology to reconstruct a panoramic nonlinear compressed image and perform dimensionality reduction on a perspective image based on a down-sampling measure.
Drawings
FIG. 1 is a basic block diagram of a high-resolution image knowledge prior-based compressed sensing technology of the present invention.
FIG. 2 is a schematic view of a hybrid vision apparatus of the present invention.
Detailed Description
The invention is further described below with reference to the accompanying drawings.
the invention discloses a high-resolution image knowledge prior-based compressed sensing technology, which mainly comprises the following steps: setting the middle dimension of a high-dimensional image and a low-dimensional image in mixed vision, calculating a high-dimensional image dimension reduction observation matrix and reducing the dimension of the high-dimensional image, calculating a low-dimensional image dimension increase observation matrix and increasing the dimension of the low-dimensional image, matching the two obtained pictures with the same dimension by using an SIFT matching algorithm and calculating the matching rate, and repeating the steps until the optimal middle dimension which can obtain the highest matching rate is found. The method adopts a scheme of machine vision and image processing, utilizes a compressive sensing theory to realize the reconstruction mapping from a low-dimensional panoramic target characteristic image to a high-dimensional image, simultaneously carries out resolution reduction sampling on a conventional large-scale image, and finds the optimal matching point of panoramic characteristic scale ascending and perspective characteristic scale descending by combining the matching precision test result, thereby realizing the precise matching algorithm of the positioned target characteristics under different scales and improving the positioning precision of three-dimensional information sensing of the mixed visual environment.
The invention adopts the technical scheme that the purpose of the invention is realized by: the imaging scales of the same interested target are greatly different by the heterogeneous dual-scale hybrid vision system, the corresponding relation between a space view field and an imaging area shows obvious nonlinear characteristics mainly due to the fact that panoramic vision passes through a space light domain compression effect in the imaging process of the hyperboloid reflector, and on the contrary, the target can be finely imaged through optical zooming operation in the imaging of a conventional vision system, so that how to realize accurate matching of the characteristics in images with different scales is an important step of positioning. The panoramic system has complete and continuous characteristics on environmental imaging, small-scale CCD digital discrete imaging is discrete compression sampling on spatial imaging information, reconstruction mapping from a low-dimensional panoramic target feature image to a high-dimensional image is realized by utilizing a compressive sensing theory, resolution reduction sampling is carried out on a conventional large-scale image, the optimal matching point of panoramic feature scale up and perspective feature scale down is found by combining the matching precision test result, and the accurate matching algorithm of the positioned target features under different scales is realized.
A compressed sensing technology based on high-resolution image knowledge prior is characterized in that: the method comprises the following main steps:
1) And estimating and setting the middle dimension of the high-dimensional image and the low-dimensional image in the mixed vision according to the prior knowledge.
2) And calculating a high-dimensional image dimension reduction observation matrix, reducing the dimension of the high-dimensional image, calculating a low-dimensional image dimension increasing observation matrix, and increasing the dimension of the low-dimensional image.
3) And matching the two obtained pictures with the same dimensionality by using an SIFT matching algorithm and calculating the matching rate.
4) Repeating the steps until the optimal middle dimensionality which can obtain the highest matching rate is found.
Claims (2)
1. A compressed sensing method based on high-resolution image knowledge prior is characterized by comprising the following steps:
1) Setting the middle dimension of a high-dimensional image and a low-dimensional image in a hybrid vision system;
2) calculating a high-dimensional image dimension reduction observation matrix, reducing the dimension of the high-dimensional image, calculating a low-dimensional image dimension increasing observation matrix, and increasing the dimension of the low-dimensional image;
3) Matching the two obtained pictures with the same dimensionality by using an SIFT matching algorithm and calculating a matching rate;
4) Repeating the steps until the optimal middle dimensionality which can obtain the highest matching rate is found.
2. the hybrid vision system applying the high-resolution image knowledge prior compressed sensing method according to claim 1, wherein: the panoramic camera comprises a panoramic camera body, the upper half part is a perspective camera, the panoramic camera body comprises a hyperboloid all-dimensional imaging reflector (1), the hyperboloid all-dimensional imaging reflector is supported on a panoramic camera support base (3) through a panoramic camera annular transparent glass supporting cylinder (2), a 1394 camera (4) is vertically and upwards arranged on the panoramic camera support base (3) in the panoramic camera annular transparent glass supporting cylinder (2) below the all-dimensional imaging reflector (1), a 1394 camera (5) serving as the perspective camera is vertically connected to a rotatable link rod (6) and is vertically connected to the panoramic camera support base (3), the panoramic camera is supported by the perspective camera annular transparent glass supporting cylinder (7) and is placed on a perspective camera support base (8).
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610428922.1A CN106097248B (en) | 2016-06-16 | 2016-06-16 | High-resolution image knowledge prior-based compressed sensing method and mixed vision system thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610428922.1A CN106097248B (en) | 2016-06-16 | 2016-06-16 | High-resolution image knowledge prior-based compressed sensing method and mixed vision system thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106097248A CN106097248A (en) | 2016-11-09 |
| CN106097248B true CN106097248B (en) | 2019-12-10 |
Family
ID=57235849
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610428922.1A Active CN106097248B (en) | 2016-06-16 | 2016-06-16 | High-resolution image knowledge prior-based compressed sensing method and mixed vision system thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106097248B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107644444B (en) * | 2017-09-07 | 2020-04-03 | 广东工业大学 | Single-image camera calibration method based on compressed sensing |
| CN110631556B (en) * | 2019-09-26 | 2021-09-07 | 湖州南太湖智能游艇研究院 | Distance measurement method of heterogeneous stereoscopic vision system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102783135A (en) * | 2010-03-03 | 2012-11-14 | 伊斯曼柯达公司 | Method and apparatus for providing a high resolution image using low resolution |
| CN103051884A (en) * | 2013-01-14 | 2013-04-17 | 哈尔滨工程大学 | Omni-directional visual monitoring system combining rough and fine modes |
| CN103971378A (en) * | 2014-05-29 | 2014-08-06 | 福州大学 | Three-dimensional reconstruction method of panoramic image in mixed vision system |
-
2016
- 2016-06-16 CN CN201610428922.1A patent/CN106097248B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102783135A (en) * | 2010-03-03 | 2012-11-14 | 伊斯曼柯达公司 | Method and apparatus for providing a high resolution image using low resolution |
| CN103051884A (en) * | 2013-01-14 | 2013-04-17 | 哈尔滨工程大学 | Omni-directional visual monitoring system combining rough and fine modes |
| CN103971378A (en) * | 2014-05-29 | 2014-08-06 | 福州大学 | Three-dimensional reconstruction method of panoramic image in mixed vision system |
Non-Patent Citations (1)
| Title |
|---|
| 压缩感知及其图像处理应用研究进展与展望;任越美等;《自动化学报》;20140831;第1563-1575页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106097248A (en) | 2016-11-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US11823363B2 (en) | Infrared and visible light fusion method | |
| Genovese et al. | A 360-deg digital image correlation system for materials testing | |
| CN109413407B (en) | High spatial resolution light field acquisition device and image generation method | |
| CN102221331B (en) | Measuring method based on asymmetric binocular stereovision technology | |
| CN102589516B (en) | Dynamic distance measuring system based on binocular line scan cameras | |
| CN112880563B (en) | Single-dimensional pixel combination mode equivalent narrow-area-array camera spatial position measuring method | |
| CN112254656B (en) | Stereoscopic vision three-dimensional displacement measurement method based on structural surface point characteristics | |
| CN106097248B (en) | High-resolution image knowledge prior-based compressed sensing method and mixed vision system thereof | |
| CN107727238A (en) | Infrared parallelly compressed imaging system and imaging method based on mask plate modulation | |
| CN107633482A (en) | A kind of super resolution ratio reconstruction method based on sequence image | |
| CN115187565A (en) | Underwater pier disease identification and positioning method and device, electronic equipment and storage medium | |
| CN107392855A (en) | Image Super-resolution Reconstruction method based on sparse autoencoder network Yu very fast study | |
| CN103914818A (en) | Omni-directional image sparse reconstruction method based on omni-directional total variation | |
| CN105300302A (en) | Diameter measurement method for Brinell hardness indent circle | |
| CN104123726A (en) | Blanking point based large forging measurement system calibration method | |
| CN110211091A (en) | A kind of full resolution pricture reconstructing method, device and crack nondestructive detection system | |
| CN109241932A (en) | A kind of thermal infrared human motion recognition method based on movement variogram phase property | |
| CN105528761B (en) | A kind of super-resolution reconstructing method of micro-image | |
| CN103186891A (en) | Hexagon image reconstruction method based on compressed sensing | |
| CN207832691U (en) | A kind of industrial camera shooting head moving device | |
| CN107092908B (en) | Automatic identification method based on plane embossed characters on train bogie | |
| CN207280599U (en) | Infrared parallelly compressed imaging system based on mask plate modulation | |
| CN106949881B (en) | A kind of mobile robot fast vision localization method | |
| CN113379844B (en) | Method for detecting large-range surface quality of airplane | |
| CN102298768A (en) | High-resolution image reconstruction method based on sparse samples |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |