CN106020827A - Medical image texture analysis system - Google Patents
Medical image texture analysis system Download PDFInfo
- Publication number
- CN106020827A CN106020827A CN201610346609.3A CN201610346609A CN106020827A CN 106020827 A CN106020827 A CN 106020827A CN 201610346609 A CN201610346609 A CN 201610346609A CN 106020827 A CN106020827 A CN 106020827A
- Authority
- CN
- China
- Prior art keywords
- image
- analysis system
- texture analysis
- module
- texture
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F8/00—Arrangements for software engineering
- G06F8/30—Creation or generation of source code
- G06F8/34—Graphical or visual programming
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/451—Execution arrangements for user interfaces
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Human Computer Interaction (AREA)
- Image Analysis (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The present invention relates to a medical image texture analysis system. By utilization of MATLAB programming, Fourier transform and autocorrelation function texture analysis methods are integrated in a GUI interface, a generated GUI module is compiled into an EXE file independent of the MATLAB software, and image analysis is achieved by directly operating the EXE file. The image texture analysis system comprises an image input module, an image adjusting module, an image processing module, a status module and an image output module. According to the medical image texture analysis system, multiphoton microscopic images of skin injury can be qualitatively analyzed and processed, and images and curve graphs which are visual and convenient for judgment can be obtained.
Description
Technical field
The present invention relates to medical imaging texture analysis field, particularly a kind of medical imaging Texture Analysis System.
Background technology
Image texture is area of pattern recognition important research content, has a very wide range of applications background in field of biomedical research.One key problem of texture analysis is texture description i.e. texture feature extraction, there are some researches show that natural texture distinguishes that most important three latitudes are periodicity, directivity, randomness, at present at the most multiple texture characteristic extracting method in analyzing image texture field, wherein Fourier transformation is exactly a kind of important signal processing texture analysis method, Fourier transformation frequency spectrum comprises abundant image information, can describe roughly texture pattern.Extracted the feature of projection Energy distribution situation by texture image Fourier spectrogram, roughness and the directivity of texture can be reflected well.For periodicity principle, Fourier frequency spectrum is made up of regular-scattered spike in a certain direction, can realize the qualitative analysis of texture;For the texture that directivity is stronger, spectrogram also remain obvious directivity.Auto-correlation function is then a kind of conventional spatial frequency texture analysis method, auto-correlation function and power spectrum is had to be a pair Fourier transform pairs by wiener-khintchine's theorem, and the One-dimensional power spectrum of image approximates exponentially curve-like, by image power spectral function derivation graph as auto-correlation function expression formula, the correlation curve of auto-correlation function is obtained by MATLAB establishment, extract correlation analysis parameter, the judgement of Texture-period (roughness) can be realized.The application of both texture analysis methods, distinguishes, to the natural texture of medical imaging, the effect that can reach more significant.
At present, texture analysis software application is mainly adopted MATLAB with ImageJ and carries out, but on the application function of its correspondence, still have many problems to need improvement badly.On the one hand, MATLAB software is relatively low to the processing speed of texture image, user interface is fairly simple, need user to use manual mode and single to search and call single function to realize texture recognition, the most integrated multiple texture analysis method is in the complete module of same man-machine friendly interface, it is impossible to directly invoke existing module and image texture carries out the analysis of multiple dimension;On the other hand, another common software ImageJ that analyzing image texture generally utilizes realizes, this software function is enriched, interface is simple, experimentation is widely used, but the research for this project shows functional redundancy in application, do not take into full account that the function interface of software is disposed and property easy to use, the most unified normalized analysis flow process, this is inconvenient for putting into clinical diagnosis.Additionally, the image of second_harmonic generation comprises heavy calculating process, carry out medical evaluation real-time, effective and treatment for the ease of clinic, need the raising of image texture analysis system perfect in shape and function badly.
Summary of the invention
In view of this, the purpose of the present invention is to propose to a kind of medical imaging Texture Analysis System, this system provides the method that texture image is analyzed, be respectively " Fourier transformation ", " auto-correlation function ", and be integrated in an interface display analysis.Thus, the present invention can carry out qualitative analysis and process to the multi-photon micro-image of skin trauma, draws image and curve chart directly perceived, to be easy to judgement.
The present invention uses below scheme to realize: a kind of medical imaging Texture Analysis System, MATLAB is utilized to program, by the texture analysis method collection of Fourier transformation and auto-correlation function together in a gui interface, and the GUI module of generation is compiled into the EXE file independent of MATLAB software, realize the analysis of image by directly running EXE file;Described image texture analysis system includes image input module, image adjustment module, image processing module, block of state, image output module.
Further, described image input module includes " opening " and " removing " button, in order to carry out input and the removing of image.
Further, described image adjustment module carries out the adjustment of brightness and contrast in order to select to process ground image.
Further, described image processing module includes Fourier transform unit, auto-correlation function unit, in order to image is carried out texture analysis.
Further, described Fourier transform unit regulates the spectrogram effect of image by the slider bar on slip Fourier spectrum figure side, it is determined by the directivity of the position judgment texture of S (θ) peak of curve, by the analysis of S (r) curve being obtained the qualitative analysis of grain roughness.
Further, described auto-correlation function unit observes crest and the alternate shape of trough and the function regularity of distribution with the amplitude of variation of distance and then the periodic distribution size of observation texture by auto-correlation function profile.
Further, described block of state is in order to show that image procossing runs process.
Further, described image output module connects the outfan of described image processing module, shows image relevant treatment curve, the availability of the image texture curve that feedback processing goes out.
Further, described image texture analysis system is provided with functions of shortcut key.
It is also preferred that the left the present invention (1) develops graphical analysis code by MATLAB software, realize to image procossing two kinds of methods respectively: Fourier transformation, auto-correlation function.(2) gone out the interface of program by the GUIDE interface programming module appropriate design in MATLAB, and independent Fourier transformation, auto-correlation function are integrated into a complete module by interface.(3) GUI module that exploitation is completed by the compiler carried by MATLAB software be compiled into can the EXE file of independent operating, a namely independent program, can isolated operation in the PC not installing MATLAB software.
Compared with prior art, the present invention has a following beneficial effect:
1, present system goes out the fresh code to analyzing image texture by MATLAB software development, employs new method and realizes the analysis to texture image.
2, integrated by program of the present invention, complicated procedure set is become man-machine friendly interface, the numbers such as texture analysis desired parameters can be obtained intuitively, simply by easy steps such as key operations, analyze periodicity and the directivity judging texture, set up the parameter index for monitoring and assess processes of wound repair.
3, the present invention uses MATLAB to develop, and can depart from MATLAB software independent operating in computer, so MATLAB being programmed unfamiliar user also can conveniently use native system.
4, present invention achieves the analyzing and processing of the Fourier transformation to texture image, auto-correlation function, greatly reduce the user lengthy and jumbled degree to the process of image.
5, the present invention realizes the same interface display of a multiple result of width picture, it is simple to after extracting the quantitative characteristic of image texture, and process, picture can store.
6, function expansibility of the present invention is strong, and the follow-up extension to other correlation functions of system has exploitativeness.
Accompanying drawing explanation
Fig. 1 is that embodiment of the present invention system processes schematic diagram to the Fourier transformation of texture image.
Fig. 2 is that embodiment of the present invention system processes schematic diagram to the auto-correlation function of texture image.
Detailed description of the invention
Below in conjunction with the accompanying drawings and embodiment the present invention will be further described.
As shown in Figure 1, present embodiments provide a kind of medical imaging Texture Analysis System, MATLAB is utilized to program, by the texture analysis method collection of Fourier transformation and auto-correlation function together in a gui interface, and the GUI module of generation is compiled into the EXE file independent of MATLAB software, realize the analysis of image by directly running EXE file;Described image texture analysis system includes image input module, image adjustment module, image processing module, block of state, image output module.
In the present embodiment, described image input module includes " opening " and " removing " button, in order to carry out input and the removing of image.By clicking on open button, optional power taking brain needing picture to be processed, the most optional two pictures, the picture chosen shows in image display box, and native system support is to JPG, the process of BMP, TIF format-pattern.When needing to remove already present picture in interface, clicking on " removing " can empty all images in interface, i.e. " original image " and the picture in " output image " module.
In the present embodiment, described image adjustment module is in order to select image to be processed to carry out the adjustment of brightness and contrast.Described image adjustment module includes brightness and contrast edit box, " arranging image 1 ", " arranging image 2 " button, this functions of modules is to be adjusted the picture shown in original image module, realize the reasonable disposition to its brightness Yu contrast, respectively image 1 and image 2 are carried out with " arranging image 2 " button by clicking on " arranging image 1 " adjustment of brightness and contrast.Handled image in image processing module is all the image after processing with contrast through image adjustment module picture luminance.
In the present embodiment, described image processing module includes Fourier transform unit, auto-correlation function unit, in order to image is carried out texture analysis, comprises " Fourier transformation " to picture processing, the big function of " auto-correlation " two.
In the present embodiment, described Fourier transform unit regulates the spectrogram effect of image by the slider bar on slip Fourier spectrum figure side, by the directivity of the position judgment texture to S (θ) peak of curve, by the analysis of S (r) curve being obtained the qualitative analysis of grain roughness.Picture after image adjustment module processes is carried out Fourier transformation, clicks on " Fourier transformation " and the Fourier spectrum figure of image and the normalizing energy distribution curve (as shown in Figure 1) in S (θ) and S (r) direction can be shown in output image module.The spectrogram effect of image can be regulated by the slider bar on slip Fourier spectrum figure side, the position being determined by S (θ) peak of curve can determine whether the directivity of texture, by S (r) curve is analyzed the qualitative analysis of available grain roughness.
In the present embodiment, described auto-correlation function unit observes crest and the alternate shape of trough and the function regularity of distribution with the amplitude of variation of distance and then the periodic distribution size of observation texture by auto-correlation function profile.Picture after image adjustment module processes is carried out auto-correlation processing, click on " auto-correlation function " button, auto-correlation processing figure and the curve chart (as shown in Figure 2) of image can be shown in output image module, observe crest and the alternate shape of trough and the function regularity of distribution periodic distribution size with the amplitude of variation observable texture of distance by auto-correlation function profile.
In the present embodiment, described block of state is in order to show that image procossing runs process.Described block of state includes dynamic text display box, and the function of this module is the state that display system processes image, shows that state, for " ready ... ", shows " in operation ... " in system operation during system off-duty, and program end of run shows " completing ... ".
In the present embodiment, described image output module connects the outfan of described image processing module, shows image relevant treatment curve, the availability of the image texture curve that feedback processing goes out.This module includes that image display box, Fourier's brightness of image regulate slider bar, the function of this module is the result of output image procossing, and additionally when image is carried out Fourier's image procossing, the Fourier spectrum figure left side there will be the slider bar to the regulation of picture centre brightness radius.The image of this module can preserve, and moves to mouse need above the picture of preservation, and right button occurs that image preserves pop-up box, can carry out picture storage.
In the present embodiment, described image texture analysis system is provided with functions of shortcut key, it is simple to be rapidly performed by image procossing." Ctrl+O ": open image shortcut, " Ctrl+F ": Fourier transformation shortcut, " Ctrl+C ": auto-correlation function shortcut, " Ctrl+A ": close auto-correlation function shortcut in length and breadth, " Ctrl+D ": remove image shortcut.
It is also preferred that the left the present invention (1) develops graphical analysis code by MATLAB software in the present embodiment, realize to image procossing two kinds of methods respectively: Fourier transformation, auto-correlation function.(2) gone out the interface of program by the GUIDE interface programming module appropriate design in MATLAB, and independent Fourier transformation, auto-correlation function are integrated into a complete module by interface.(3) GUI module that exploitation is completed by the compiler carried by MATLAB software be compiled into can the EXE file of independent operating, a namely independent program, can isolated operation in the PC not installing MATLAB software.
The foregoing is only presently preferred embodiments of the present invention, all impartial changes done according to scope of the present invention patent and modification, all should belong to the covering scope of the present invention.
Claims (9)
1. a medical imaging Texture Analysis System, it is characterized in that: utilize MATLAB to program, by the texture analysis method collection of Fourier transformation and auto-correlation function together in a gui interface, and the GUI module of generation is compiled into the EXE file independent of MATLAB software, realize the analysis of image by directly running EXE file;Described image texture analysis system includes image input module, image adjustment module, image processing module, block of state, image output module.
A kind of medical imaging Texture Analysis System the most according to claim 1, it is characterised in that: described image input module includes " opening " and " removing " button, in order to carry out input and the removing of image.
A kind of medical imaging Texture Analysis System the most according to claim 1, it is characterised in that: described image adjustment module carries out the adjustment of brightness and contrast in order to select to process ground image.
A kind of medical imaging Texture Analysis System the most according to claim 1, it is characterised in that: described image processing module includes Fourier transform unit, auto-correlation function unit, in order to image is carried out texture analysis.
A kind of medical imaging Texture Analysis System the most according to claim 4, it is characterized in that: described Fourier transform unit regulates the spectrogram effect of image by the slider bar on slip Fourier spectrum figure side, it is determined by the directivity of the position judgment texture of S (θ) peak of curve, by the analysis of S (r) curve being obtained the qualitative analysis of grain roughness.
A kind of medical imaging Texture Analysis System the most according to claim 4, it is characterised in that: described auto-correlation function unit observes crest and the alternate shape of trough and the function regularity of distribution with the amplitude of variation of distance and then the periodic distribution size of observation texture by auto-correlation function profile.
A kind of medical imaging Texture Analysis System the most according to claim 1, it is characterised in that: described block of state is in order to show that image procossing runs process.
A kind of medical imaging Texture Analysis System the most according to claim 1, it is characterised in that: described image output module connects the outfan of described image processing module, shows image relevant treatment curve, the availability of the image texture curve that feedback processing goes out.
A kind of medical imaging Texture Analysis System the most according to claim 1, it is characterised in that: described image texture analysis system is provided with functions of shortcut key.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610346609.3A CN106020827B (en) | 2016-05-24 | 2016-05-24 | Medical image texture analysis system |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610346609.3A CN106020827B (en) | 2016-05-24 | 2016-05-24 | Medical image texture analysis system |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106020827A true CN106020827A (en) | 2016-10-12 |
CN106020827B CN106020827B (en) | 2019-12-10 |
Family
ID=57092968
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610346609.3A Active CN106020827B (en) | 2016-05-24 | 2016-05-24 | Medical image texture analysis system |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106020827B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108122222A (en) * | 2016-12-14 | 2018-06-05 | 北京无线电计量测试研究所 | A kind of automatic displaying method on graphical analysis |
CN111220092A (en) * | 2019-12-10 | 2020-06-02 | 东南大学 | Method for constructing fringe filter in optical measurement |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101030297A (en) * | 2007-03-29 | 2007-09-05 | 杭州电子科技大学 | Method for cutting complexity measure image grain |
CN102122356A (en) * | 2011-03-16 | 2011-07-13 | 中国人民解放军第二军医大学 | Computer-aided method for distinguishing ultrasound endoscope image of pancreatic cancer |
-
2016
- 2016-05-24 CN CN201610346609.3A patent/CN106020827B/en active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101030297A (en) * | 2007-03-29 | 2007-09-05 | 杭州电子科技大学 | Method for cutting complexity measure image grain |
CN102122356A (en) * | 2011-03-16 | 2011-07-13 | 中国人民解放军第二军医大学 | Computer-aided method for distinguishing ultrasound endoscope image of pancreatic cancer |
Non-Patent Citations (1)
Title |
---|
XIEXIEMING: "《https://download.csdn.net/download/xiexieming/1795086》", 5 November 2009 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108122222A (en) * | 2016-12-14 | 2018-06-05 | 北京无线电计量测试研究所 | A kind of automatic displaying method on graphical analysis |
CN111220092A (en) * | 2019-12-10 | 2020-06-02 | 东南大学 | Method for constructing fringe filter in optical measurement |
CN111220092B (en) * | 2019-12-10 | 2021-04-23 | 东南大学 | Method for constructing fringe filter in optical measurement |
Also Published As
Publication number | Publication date |
---|---|
CN106020827B (en) | 2019-12-10 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107895367B (en) | Bone age identification method and system and electronic equipment | |
CN108564026B (en) | Network construction method and system for thyroid tumor cytology smear image classification | |
Dobrescu et al. | Medical images classification for skin cancer diagnosis based on combined texture and fractal analysis | |
DE102010018864A1 (en) | Thumb touch profile sensor device | |
KR20200116107A (en) | Automated monitoring of medical imaging procedures | |
KR20170128454A (en) | Systems and methods for deconvolutional network-based classification of cell images and videos | |
RU2013154102A (en) | FINGER RECOGNITION AND TRACKING SYSTEM | |
WO2013064605A2 (en) | Generate percentage of positive cells for biomarkers by normalizing and autothresholding the image intensity produced by immunohistochemistry technique | |
CN104424386A (en) | Multi-parameter magnetic resonance image based prostate cancer computer auxiliary identification system | |
Ring et al. | Dermatoscopy | |
CN111166362B (en) | Medical image display method and device, storage medium and electronic equipment | |
CN106020827A (en) | Medical image texture analysis system | |
EP3265956A1 (en) | Classification of cellular images and videos | |
Kang et al. | Real-time image processing system for endoscopic applications | |
Zhao et al. | Fine-grained diabetic wound depth and granulation tissue amount assessment using bilinear convolutional neural network | |
CN105069444A (en) | Gesture recognition device | |
CN102467237A (en) | Device and method for realizing mouse function by using non-contact gestures | |
Elloumi et al. | Fast and accurate mobile-aided screening system of moderate diabetic retinopathy | |
Kumar et al. | Automated white corpuscles nucleus segmentation using deep neural network from microscopic blood smear | |
Saroğlu et al. | Machine learning, IoT and 5G technologies for breast cancer studies: A review | |
Gupta | Real time wound segmentation/management using image processing on handheld devices | |
Rai et al. | Fragmentary shape recognition: A BCI study | |
Pandey et al. | An integrated approach for breast cancer classification | |
CN111161256A (en) | Image segmentation method, image segmentation device, storage medium, and electronic apparatus | |
Liang et al. | Recognizing focal liver lesions in contrast-enhanced ultrasound with discriminatively trained spatio-temporal model |
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 |