CN105809159A - Imaging-based visual weight graph extraction method - Google Patents
Imaging-based visual weight graph extraction method Download PDFInfo
- Publication number
- CN105809159A CN105809159A CN201610081221.5A CN201610081221A CN105809159A CN 105809159 A CN105809159 A CN 105809159A CN 201610081221 A CN201610081221 A CN 201610081221A CN 105809159 A CN105809159 A CN 105809159A
- Authority
- CN
- China
- Prior art keywords
- image
- imaging
- vision weight
- fourier
- filtering
- 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
- 238000003384 imaging method Methods 0.000 title claims abstract description 33
- 238000000605 extraction Methods 0.000 title claims abstract description 11
- 230000000007 visual effect Effects 0.000 title abstract description 13
- 238000000034 method Methods 0.000 claims abstract description 25
- 238000001914 filtration Methods 0.000 claims abstract description 19
- 238000001228 spectrum Methods 0.000 claims abstract description 14
- 230000003287 optical effect Effects 0.000 claims description 10
- 230000009466 transformation Effects 0.000 claims description 3
- 238000004458 analytical method Methods 0.000 abstract description 5
- 230000000694 effects Effects 0.000 abstract 1
- 238000005516 engineering process Methods 0.000 description 4
- 230000008569 process Effects 0.000 description 3
- 238000001514 detection method Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 238000004088 simulation Methods 0.000 description 2
- 239000000654 additive Substances 0.000 description 1
- 230000000996 additive effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000000205 computational method Methods 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000012634 optical imaging Methods 0.000 description 1
- 239000011148 porous material Substances 0.000 description 1
- 230000026676 system process Effects 0.000 description 1
- 230000009885 systemic effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/40—Analysis of texture
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/10—Image enhancement or restoration using non-spatial domain filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/10—Image acquisition
- G06V10/12—Details of acquisition arrangements; Constructional details thereof
- G06V10/14—Optical characteristics of the device performing the acquisition or on the illumination arrangements
- G06V10/147—Details of sensors, e.g. sensor lenses
-
- 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
- G06T2207/20056—Discrete and fast Fourier transform, [DFT, FFT]
-
- 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
- G06T2207/20092—Interactive image processing based on input by user
- G06T2207/20104—Interactive definition of region of interest [ROI]
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Vascular Medicine (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses an imaging-based visual weight graph extraction method. A system brought forward by the invention is based on a single-optical-path double Fourier lens imaging, can directly carry out imaging analysis of a visual weight on a scene, and can effectively and rapidly obtain a high-precision visual weight graph. The method comprises the following steps: 1, constructing a single-optical-path double Fourier lens imaging system; and 2, generating a visual weight image. According to the method, the single-optical-path double Fourier lens imaging system is constructed, different filtering of two times is realized at a frequency spectrum surface, different low-pass filtering effects are realized, a filtering result is obtained by use of an industrial camera, finally, the visual weight graph is obtained through an absolute value of a gray-scale difference between images, and the processing result is close to that of human eye visual system. The corresponding visual weight graph can be rapidly obtained as long as the system aims at the scene (it should be satisfied that an object distance is greater than a lens focal length by more than ten times).
Description
Technical field
The present invention relates to optical imagery and image processing techniques, particularly relate to a kind of vision weight figure extracting method based on imaging.
Background technology
Visual information in the form of light, is received by the mankind by human eye, is collected by imageing sensor, and this is the topmost information source that people recognize the world.After visual information is received by imageing sensor, it is the formation of the computer picture that we recognize.The analysis of image and process have become as a very important technology, wherein analyze the vision weight of image the unusual people of Image Engineering of today.
Analyzing the vision weight of image, namely according to original image, obtain the image of reflection zones of different difference weight distribution, the numerical value of this image is between 0~1, more big, shows that human eye more pays close attention to this region, more little then contrary.This is the function using information theory to solve simulation human eye, automatically realizes graphical analysis work.Therefore, the analysis of image vision weight is intended to obtain high-quality vision weight figure, reflects the human eye degree of concern of zones of different in image.
The analysis of this vision weight has very important application, the automatic detection of such as target, the retrieval of image, and compression of image and video etc. can use vision weight figure auxiliary to realize efficiently.At present, vision weight analyzes method such as significance detection technique, in every respect also cannot be satisfactory.On the one hand, the resolution of vision weight figure own is not high enough;On the other hand, the relative complex typically via computer that obtains of vision weight figure calculates, and efficiency depends on the computing capability of hardware.How setting up the computational methods of the fast vision weight map more meeting human visual system is also one of difficult problem of processing of current visual image information.
Summary of the invention
The problem that this invention address that is to provide a kind of vision weight figure extracting method based on imaging, utilizes the operation on monochromatic light road and simple later stage computing, can effectively simplify the complexity of image vision weight extraction, improve the speed of system process images.
For solving the problem described in background technology, the present invention proposes a kind of vision weight figure extracting method based on imaging, and the method is simple but highly effective, and the method flow process specifically includes that
Step (1) builds the double; two fourier lense imaging system in monochromatic light road;
For scenery, first carry out the structure of the double; two fourier lense imaging system in monochromatic light road of vision weight figure extraction, and be filtered and gather acquisition operation, finally utilize industrial camera to obtain low-pass filtering image g1With g2;
The acquisition of step (2) vision weight figure;
After industrial camera obtains image, the gray value of two figure is subtracted each other and seeks absolute value, it is achieved the extraction of vision weight figure, it is thus achieved that vision weight figure;Namely g is being obtained1With g2On basis, utilize S=| g1-g2| just can obtain the final vision weight image S of image.
The double; two fourier lense imaging system in structure monochromatic light road described in step (1), and it is as follows with the concrete grammar gathering the operations such as acquisition to carry out image filtering:
Scenery: the natural landscape that analyze and system wish to analyze the object obtaining vision weight figure, the distance (object distance) of scenery and first fourier lense is much larger than the focal length (object distance > 10 times focal length) of fourier lense;
Optical Fourier transform: scene light generates the Fourier spectrum of image through first fourier lense focal plane behind, generates position at frequency plane place;
Space filtering: on the frequency plane generated, places precise pinhole and realizes space filtering, filtering high frequency;Different cut-off frequencies is realized by adjusting the size of aperture radius;Select two pinhole size respectively A at frequency spectrum place1、A2um;
The image of imaging image planes generates: placing fourier lense after frequency plane, filtered frequency spectrum again passes by Fourier transformation and generates spatial image, is gathered by industrial camera, for two various sizes of wave filter A1With A2, it is thus achieved that image g1With g2。
The present invention, by adopting the means of optical imagery, coordinates the simple additive operation of hardware, it is achieved the extraction of vision weight figure, specific as follows:
1, the double; two fourier lense system of structure incoherent light optical imagery, for d0Remote scenery, this distance is far longer than the focal distance f of fourier lense, and frequency plane is on the back focal plane of first fourier lense, operates for frequency spectrum below and provides frequency plane position;Then continue with fourier lense inversion and change to imaging surface, obtain image with image camera.
2, frequency spectrum operation.Analyze the different frequency information obtaining scenery, 1 frequency plane generated places aperture, the i.e. spatial filter of amplitude type, it is achieved image spectrum filters and conducting.The low-pass filtering image of different cut-off frequency can be obtained by adjusting the size of aperture radius.Here adopting two various sizes of apertures, aperture is D respectively1With D2。
3, the extraction of vision weight figure.Image camera obtains image: utilizes the aperture of in 22 different pore sizes to realize space filtering, obtains the image that two width cut-off frequencies are different;Two width images are subtracted each other in data, its absolute value and the defined vision weight figure of the present invention.
Compared with prior art, the technical program has the advantage that
The extracting method of the existing vision weight figure utilized on significance basis is all directly use computerized algorithm, and the hardware requirements such as CPU is relatively high, and time cost also can promote;And the method based on imaging that the present invention proposes, utilize the concurrency of optical oomputing, high efficiency, it is achieved stable vision weight figure obtains.The characteristic of another aspect of the present invention is in that, only uses monochromatic light road and to scenery direct imaging and analyze, and conventional extracting method is all be calculated image on computers obtaining.In general, one is efficient, is efficiently in that the concurrency of optical oomputing;But high-resolution, the inventive method is obtained in that the vision weight figure being the systemic resolution limit to the maximum;Three is stable, is stably the simple in construction due to whole system, can be directly integrated in existing imaging system such that it is able to more effective simulation biological vision process;Four be directly scenery is analyzed, rather than to the image gathered, for the application place mat basis in later stage.
Accompanying drawing explanation
Fig. 1 is the concrete operations flow chart of the inventive method;
Fig. 2 is the double; two fourier lense imaging system in monochromatic light road that the inventive method builds;
Fig. 3 a-3c is one group of lab diagram involved by a concrete embodiment, wherein:
Fig. 3 a is the image 1 that imaging surface obtains;
Fig. 3 b is the image 2 that imaging surface obtains;
Fig. 3 c is the vision weight figure obtained.
Detailed description of the invention
In order to obtain the visual weight distribution of image, the present invention utilizes optical imaging modalities, utilizes the real-time of the Calculation of Optical Path, quickly realizes the extraction of vision weight figure.
Below in conjunction with accompanying drawing, by specific embodiment, technical scheme is carried out clear, complete description.
The present invention proposes the operation framework of a kind of vision weight figure extracting method based on imaging as it is shown in figure 1, it mainly comprises the steps of
1, the double; two fourier lense imaging system in structure monochromatic light road, as in figure 2 it is shown, scene light enters system, and then is filtered operating with acquisition etc..Whole imaging optical path system includes being positioned at d0Scenery at a distance, two focal lengths are the fourier lense of f, 2 precise pinhole (wave filter), industry image camera.Wish to analyze the object obtaining vision weight figure as in figure 2 it is shown, natural landscape and scenery are systems;Scenery light path is through the acquisition face of double; two fourier lense system imagings to image camera;Frequency plane in double; two fourier lense systems uses two different precise pinhole to be filtered operation, selects different frequencies for image, thus obtaining filtering image.This processing procedure wherein comprised includes:
Optical Fourier transform: scene light generates the Fourier spectrum of image through first fourier lense focal plane behind, generates frequency plane place in such as Fig. 2, the position;
Space filtering: on the frequency plane generated, i.e. Fig. 2 intermediate frequency spectrum face place, place precise pinhole and realize space filtering, filtering high frequency.Different cut-off frequencies is realized by adjusting the size of aperture radius.Select two pinhole size respectively A at frequency spectrum place1、A2Um (micron);
The image of imaging image planes generates: placing fourier lense after frequency plane, filtered frequency spectrum again passes by Fourier transformation and generates spatial image, is gathered by industrial camera, for two various sizes of wave filter A1With A2, it is thus achieved that image g1With g2, respectively as best seen in figs. 3a and 3b.
2, vision weight image generates.After industrial camera obtains image, the gray value of two figure is subtracted each other and seeks absolute value, it is achieved the extraction of vision weight figure, it is thus achieved that vision weight figure.Namely, after obtaining Fig. 3 a and Fig. 3 b, S=is utilized | g1-g2| just can obtain the final vision weight image S of image, as shown in Figure 3 c.
In the legend of the present invention, the distance d of scenery0More much bigger than fourier lense focal distance f (more than 10 times), equipment or relevant parameter that used needs indicate are as follows:
Adopt image being processed with utilizing a kind of vision weight figure extracting method energy fast and stable based on imaging of the embodiment of the present invention, obtain good vision weight and extract result, refer to Fig. 2 and Fig. 3 a-3c, scenery in Fig. 2 is object to be analyzed, Fig. 3 c is the vision weight figure extracted, and is comparatively intactly highlighted in flower region interested for human eye.
Although the present invention is with preferred embodiment openly as above; but it is not for limiting the present invention; any those skilled in the art are without departing from the spirit and scope of the present invention; may be by the method for the disclosure above and technology contents and technical solution of the present invention is made possible variation and amendment; therefore; every content without departing from technical solution of the present invention; according to any simple modification, equivalent variations and modification that above example is made by the technical spirit of the present invention, belong to the protection domain of technical solution of the present invention.
Claims (2)
1. the vision weight figure extracting method based on imaging, it is characterised in that comprise the steps:
The double; two fourier lense imaging system in step (1) monochromatic light road;
For scenery, first carry out the structure of the double; two fourier lense imaging system in monochromatic light road of vision weight figure extraction, and be filtered and gather acquisition operation, finally utilize industrial camera to obtain low-pass filtering image g1With g2;
The acquisition of step (2) vision weight figure;
After industrial camera obtains image, the gray value of two figure is subtracted each other and seeks absolute value, it is achieved the extraction of vision weight figure, it is thus achieved that vision weight figure;Namely g is being obtained1With g2On basis, utilize S=| g1-g2| just can obtain the final vision weight image S of image.
2. a kind of vision weight figure extracting method based on imaging as claimed in claim 1, it is characterized in that, the double; two fourier lense imaging system in structure monochromatic light road described in step (1), and it is as follows with the concrete grammar gathering the operations such as acquisition to carry out image filtering:
Scenery: the natural landscape that analyze and system wish to analyze the object obtaining vision weight figure, and the distance of scenery and first fourier lense is more than the focal length of 10 times of fourier lenses;
Optical Fourier transform: scene light generates the Fourier spectrum of image through first fourier lense focal plane behind, generates position at frequency plane place;
Space filtering: on the frequency plane generated, places precise pinhole and realizes space filtering, filtering high frequency;Different cut-off frequencies is realized by adjusting the size of aperture radius;Select two pinhole size respectively A at frequency spectrum place1、A2um;
The image of imaging image planes generates: placing fourier lense after frequency plane, filtered frequency spectrum again passes by Fourier transformation and generates spatial image, is gathered by industrial camera, for two various sizes of wave filter A1With A2, it is thus achieved that image g1With g2。
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610081221.5A CN105809159A (en) | 2016-02-04 | 2016-02-04 | Imaging-based visual weight graph extraction method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610081221.5A CN105809159A (en) | 2016-02-04 | 2016-02-04 | Imaging-based visual weight graph extraction method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN105809159A true CN105809159A (en) | 2016-07-27 |
Family
ID=56466259
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610081221.5A Pending CN105809159A (en) | 2016-02-04 | 2016-02-04 | Imaging-based visual weight graph extraction method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN105809159A (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108921053A (en) * | 2018-06-15 | 2018-11-30 | 杭州电子科技大学 | A kind of scene objects automatically analyze detection processing method and device |
CN110460756A (en) * | 2019-08-12 | 2019-11-15 | 杭州电子科技大学 | A kind of scene removes rain image processing method and device automatically in real time |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101718589A (en) * | 2009-11-14 | 2010-06-02 | 张青川 | Optical readout method for infrared thermal imagery imager |
CN102109680A (en) * | 2011-01-07 | 2011-06-29 | 深圳大学 | Method and device for producing diffraction-free Bessel beam array in random order based on phase hologram |
US20140085715A1 (en) * | 2012-09-21 | 2014-03-27 | The Board Of Trustees Of The University Of Illinois | Diffraction Phase Microscopy with White Light |
CN103973976A (en) * | 2014-04-14 | 2014-08-06 | 杭州电子科技大学 | Saliency extraction device and method with optical imaging adopted |
-
2016
- 2016-02-04 CN CN201610081221.5A patent/CN105809159A/en active Pending
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101718589A (en) * | 2009-11-14 | 2010-06-02 | 张青川 | Optical readout method for infrared thermal imagery imager |
CN102109680A (en) * | 2011-01-07 | 2011-06-29 | 深圳大学 | Method and device for producing diffraction-free Bessel beam array in random order based on phase hologram |
US20140085715A1 (en) * | 2012-09-21 | 2014-03-27 | The Board Of Trustees Of The University Of Illinois | Diffraction Phase Microscopy with White Light |
CN103973976A (en) * | 2014-04-14 | 2014-08-06 | 杭州电子科技大学 | Saliency extraction device and method with optical imaging adopted |
Non-Patent Citations (1)
Title |
---|
李灿等: "基于液晶纯相位光调制器的4f系统去噪方法", 《光学技术》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108921053A (en) * | 2018-06-15 | 2018-11-30 | 杭州电子科技大学 | A kind of scene objects automatically analyze detection processing method and device |
CN108921053B (en) * | 2018-06-15 | 2021-05-07 | 杭州电子科技大学 | Scene target automatic analysis detection processing method and device |
CN110460756A (en) * | 2019-08-12 | 2019-11-15 | 杭州电子科技大学 | A kind of scene removes rain image processing method and device automatically in real time |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Gallo et al. | 3D reconstruction of small sized objects from a sequence of multi-focused images | |
Godse et al. | Wavelet based image fusion using pixel based maximum selection rule | |
Li et al. | Region-based multi-focus image fusion using the local spatial frequency | |
CN103973976B (en) | The conspicuousness extraction element and method of a kind of utilization optical imagery | |
CN104200450A (en) | Infrared thermal image resolution enhancing method | |
Ruchay et al. | Accuracy analysis of 3D object shape recovery using depth filtering algorithms | |
CN105809159A (en) | Imaging-based visual weight graph extraction method | |
Dziubich et al. | Depth images filtering in distributed streaming | |
Quan et al. | A depth enhancement strategy for kinect depth image | |
CN203840438U (en) | Device for extracting saliency by using optical imaging | |
Cao et al. | Depth image vibration filtering and shadow detection based on fusion and fractional differential | |
Martišek et al. | Multifocal image processing | |
Hemmat et al. | Fast planar segmentation of depth images | |
Ahmed et al. | Enhancement of depth map through weighted combination of guided image filters in shape-from-focus | |
Wang et al. | Pavement crack detection combining non-negative feature with fast LoG in complex scene | |
Lei et al. | Study on the method to process the images of the laser initiative illumination | |
Lee et al. | Spatially resolved stereoscopic surface profiling by using a feature-selective segmentation and merging technique | |
Mori et al. | Pseudo normal image generation for anomaly detection on road surface | |
Sakila et al. | Image enhancement using morphological operations | |
Bayram et al. | A graph signal filtering-based approach for detection of different edge types on airborne lidar data | |
Tian et al. | Spectral residual method of saliency detection based on the two-dimensional fractional Fourier transform domain | |
Beckman et al. | Automated volumetric damage detection and quantification using region-based convolution neural networks and an inexpensive depth camera | |
Bethmann et al. | Multi-image semi-global matching in object space | |
Li et al. | Sparse 3D point clouds segmentation considering 2D image feature extraction with deep learning | |
Ma et al. | A light field depth estimation algorithm by edge disparity extraction |
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 |
Application publication date: 20160727 |
|
WD01 | Invention patent application deemed withdrawn after publication |