CN103164843A - Medical image colorizing method - Google Patents
Medical image colorizing method Download PDFInfo
- Publication number
- CN103164843A CN103164843A CN2011104244548A CN201110424454A CN103164843A CN 103164843 A CN103164843 A CN 103164843A CN 2011104244548 A CN2011104244548 A CN 2011104244548A CN 201110424454 A CN201110424454 A CN 201110424454A CN 103164843 A CN103164843 A CN 103164843A
- Authority
- CN
- China
- Prior art keywords
- mark
- gradient
- points
- current process
- image
- 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
Images
Landscapes
- Image Analysis (AREA)
- Measuring And Recording Apparatus For Diagnosis (AREA)
Abstract
The invention discloses a medical image colorizing method. The medical image colorizing method includes the steps: obtaining black and white digital video image data of a medical image; conducting color initialization to the black and white digital video image data; conducting region segmentation to the video image data with colors initialized and dividing a picture into independent closed regions; conducting region clustering to the divided regions according to a feature of each real target; and conducting respective colorization processing to each target region according an initial color and a result of the region clustering. The medical image colorizing method is beneficial for target segmentation and color propagation, is high in system efficiency, and has stability.
Description
Technical field
The present invention relates to image processing method, particularly a kind of medical image colorize method.
Background technology
Numeral CR medical radioactive image is radiated doctor's favor deeply with its high gray resolution, powerful computer picture post-processing function, little radiation dose, without advantages such as film diagnosis, the strange land consultation of doctors, has become the new focus of medical imaging technology.Yet present most of digital CR medical image is still simple black and white grayscale image.For human eye, the world is multicoloured, color is being taken on important role in the cognitive process of the mankind to the world, far above the black and white gray scale, simple black and white grayscale image is because the disappearance of color makes a big impact to Picture Showing power to the resolution characteristic of colour for human eye.Than black-and-white image, color makes the chromatic image content abundanter, and details is more clear, and effect is more true to nature, can give prominence to details individual in image by colorize, can embody real scene.
The colorize of intelligence is different from " manual colouring ", and the high-speed cruising that depends on computing machine can be increased work efficiency greatly.Again, the mode of the colorize technology simple and effective of intelligence also has important application in medical science and industrial circle.Such as the colorize processing of the painted of satellite imagery picture and medical X-ray picture etc.
Summary of the invention
The objective of the invention is to overcome the deficiencies in the prior art, a kind of black and white medical image colorize method is proposed, realize the content of image is carried out the Region Segmentation cluster by mode identification technology, then adopt the method for color expansion, complete the application of the colorize of medical image.
For achieving the above object, a kind of medical image colorize method of the present invention comprises:
Obtain the black and white digital video image data of medical image;
Described black and white digital video image data are carried out the color initialization;
Vedio data after the color initialization is carried out Region Segmentation, picture is divided into independently closed region;
Region clustering is carried out according to the characteristics of each realistic objective in zone after cutting apart;
Divide other colorize to process according to the result of initial color and region clustering to each target area.
Described color initialization is that subjective vision characteristics according to the people are the suitable color of goal-setting in medical image.
Described Region Segmentation adopts watershed algorithm.
The step of described Region Segmentation comprises:
Adopt the Gaussian smoothing operator to carry out smoothing processing to original image;
The gradient of each point in computed image, two-dimensional gradient are calculated and are got both direction gradient square root sum square, and the scanning entire image obtains the histogram of gradient image and the probability distribution of each gradient;
Travel through each pixel and draw each pixel in the position of sequence in array, pixel is sorted;
To the minimum layer of gradient, from top to bottom, from left to right each pixel is carried out mark;
Each pixel of each gradient layer of sequential loop obtains the watershed divide point;
Each closed region label is identical according to the watershed divide line that obtains is divided and is come, and realizes image segmentation.
Described Gaussian smoothing operator template is:
Described two-dimensional gradient calculates and adopts the sobel operator:
Described to the minimum layer of gradient, from top to bottom, the principle of from left to right each pixel being carried out mark is: if there is not the point of mark in four adjoint points up and down of current process points, current process points is new mark; If there is gauge point in four adjoint points up and down of current process points, current process points label equals this mark.
The described step of obtaining watershed divide point is, each pixel of each gradient layer of sequential loop carries out mark according to following criterion:
If there is not the point of mark in four adjoint points up and down of current process points, current process points is new mark;
If four adjoint points up and down of current process points only have a mark, this mark is assigned to current process points;
If there is two or more mark in four adjoint points up and down of current process points, current process points is the watershed divide point.
Whether described region clustering is with its merging according to the gray difference deciding degree between adjacent area, the combination in several zones has consisted of actual target area, wherein the process of zone merging is each zone to initial segmentation, other zones that searching is adjacent, if the difference of area grayscale mean value each other less than certain predetermined threshold T, merges them.
Beneficial effect of the present invention will fully be set forth hereinafter.
Description of drawings
Fig. 1 is the process flow diagram of a kind of specific embodiment of the present invention;
Fig. 2 is gradient calculation principle schematic in a kind of specific embodiment of the present invention.
Embodiment
The medical image colorize method that the present invention proposes is based on the pattern clustering technology content in black-and-white image is analyzed, and carries out Region Segmentation, region clustering, adopts the method for color expansion to complete.The common way of color expansion is coated suitable multi-color cord bar and is indicated look as the part in black white image, then sets certain color expansion mode according to the local similarity of color, makes local color expand to entire image.Initial color can adopt application scenario concrete according to medical image to preset.
The below will solve a kind of embodiment of the present invention and describe technical scheme of the present invention in detail.
The major function of video acquisition is to utilize camera or scanner from black and white film or photo acquisition image and digitizing.
Because there are the problems such as fuzzy, noise in the video image of dynamic acquisition, can first do pre-service for guaranteeing image quality, shake as removing interframe defective and interframe, and noise reduction and the processing such as greyscale transformation in case of necessity, carry out the interframe luminance proportion.
This step is to complete the committed step of the colorize processing of black white image.It is specially:
1, color initialization
In a kind of preferred embodiment of the present invention, this step needs artificial the participation, is the suitable color of goal-setting in medical image according to people's subjective vision characteristics, and the simple identification significant points gets final product, and gets final product putting a color dot on the inhomogeneity object.
2, Region Segmentation
The purpose of carrying out Region Segmentation is that picture is divided into independently closed region, and then carries out detection and the correction of color in regional.Figure line dividing method commonly used has a lot, as watershed algorithm, based on the method for Active contour etc.
Utilizing watershed algorithm in this patent method is mainly the characteristic of utilizing its over-segmentation, can access the marked region of sealing.This paper adopts the method based on the immersion model of Vincent and Soille proposition.Key step is summarized as follows:
A, original image is carried out smoothing processing.But the Gaussian smoothing operator can cause obscurity boundary when suppressing noise, so the selection of Gauss's standard deviation merits attention, select the too small noise that is unfavorable for suppressing, make over-segmentation worsen, cross the fuzzy actual boundary of conference, choose the Gauss operator template according to actual conditions here as follows:
The gradient of each point in B, computed image, two-dimensional gradient calculates and gets both direction gradient square root sum square, adopts the sobel operator here:
When using the Sobel operator, can be detecting template S
1And S
2Regard one " framework " as, be enclosed within on each pixel to be detected.As shown in Figure 2, rectangle ABCD is a zone in image, and wherein each little lattice is 1 pixel.For example in figure, the gray-scale value of pixel 21 is a
5, the gray-scale value of its surrounding pixel as shown in FIG., 21 horizontal edge value EH (21) are as follows for calculating pixel:
EH(21)＝(-1)×a
1+0×a
2+1×a
3+(-2)×a
4+0×a
5+2×a
6+(-1)×a
7+0×a
8+1×a
9
But the vertical edge value EV (21) of same calculating pixel 21.
The computing formula of edge strength EI is:
In C, step 2, the scanning entire image can obtain the histogram of gradient image and the probability distribution of each gradient.Again travel through each pixel and draw each pixel in the position of sequence in array, pixel is sorted.It should be noted that in computation process according to the threshold value of setting, it is zero point that gradient all is considered to gradient less than the point of this threshold value, can greatly reduce computing time like this.Herein, this threshold value is made as 15.
D, seed generative process.To the minimum layer of gradient, from top to bottom, from left to right each pixel is carried out mark, principle is: if there is not the point of mark in four adjoint points up and down of current process points, current process points is new mark; If there is gauge point in four adjoint points up and down of current process points, current process points label equals this mark.
E, immersion process.Each pixel of each gradient layer of sequential loop, carry out mark according to following criterion:
If there is not the point of mark in four adjoint points up and down of current process points, current process points is new mark;
If four adjoint points up and down of current process points only have a mark, this mark is assigned to current process points;
If there is two or more mark in four adjoint points up and down of current process points, current process points is the watershed divide point;
F, each closed region label is identical according to the watershed divide line that obtains are divided and are come, and realize image segmentation.
3, region clustering
Image after watershed segmentation is complete, tend to form many excessive cut zone, whether just must carry out cluster according to the characteristics of each realistic objective, it be merged according to the gray difference deciding degree between adjacent area, the combination in several zones has consisted of actual target area.The process that the zone merges is each zone to initial segmentation, seeks other zones that are adjacent, if the difference of area grayscale mean value each other less than certain thresholding T, merges them.Here recommending to merge thresholding T value is 15.
4, color map
Divide other colorize to process according to the result of initial color and region clustering to each target area.
Innovation of the present invention and advantage
The black and white medical image colorize system that the present invention proposes utilizes Target Segmentation and color expansion, and system effectiveness is high, stable.
The black and white medical image colorize system module clear in structure that the present invention proposes, the each several part division of labor is clear and definite, and independence is strong, improves whole stability and reliability.
The artificial participation that the present invention proposes is few, helps greatly to improve the efficient of video color.
Those skilled in the art should understand, embodiments of the invention can be provided as method, system or computer program.Therefore, the present invention can adopt complete hardware implementation example, implement software example or in conjunction with the form of the embodiment of software and hardware aspect fully.And the present invention can adopt the form that wherein includes the upper computer program of implementing of computer-usable storage medium (including but not limited to magnetic disk memory and optical memory etc.) of computer usable program code one or more.
The present invention is that reference is described according to process flow diagram and/or the block scheme of method, equipment (system) and the computer program of the embodiment of the present invention.Should understand can be by the flow process in each flow process in computer program instructions realization flow figure and/or block scheme and/or square frame and process flow diagram and/or block scheme and/or the combination of square frame.Can provide these computer program instructions to the processor of multi-purpose computer, special purpose computer, Embedded Processor or other programmable data processing device to produce a machine, make the instruction of carrying out by the processor of computing machine or other programmable data processing device produce to be used for the device of realizing in the function of flow process of process flow diagram or a plurality of flow process and/or square frame of block scheme or a plurality of square frame appointments.
These computer program instructions also can be stored in energy vectoring computer or the computer-readable memory of other programmable data processing device with ad hoc fashion work, make the instruction that is stored in this computer-readable memory produce the manufacture that comprises command device, this command device is realized the function of appointment in flow process of process flow diagram or a plurality of flow process and/or square frame of block scheme or a plurality of square frame.
These computer program instructions also can be loaded on computing machine or other programmable data processing device, make on computing machine or other programmable devices and to carry out the sequence of operations step producing computer implemented processing, thereby be provided for realizing the step of the function of appointment in flow process of process flow diagram or a plurality of flow process and/or square frame of block scheme or a plurality of square frame in the instruction of carrying out on computing machine or other programmable devices.
Obviously, those skilled in the art can carry out various changes and modification and not break away from the spirit and scope of the present invention the present invention.Like this, if within of the present invention these are revised and modification belongs to the scope of claim of the present invention and equivalent technologies thereof, the present invention also is intended to comprise these changes and modification interior.
Claims (9)
1. medical image colorize method comprises:
Obtain the black and white digital video image data of medical image;
Described black and white digital video image data are carried out the color initialization;
Vedio data after the color initialization is carried out Region Segmentation, picture is divided into independently closed region;
Region clustering is carried out according to the characteristics of each realistic objective in zone after cutting apart;
Divide other colorize to process according to the result of initial color and region clustering to each target area.
2. medical image colorize method as claimed in claim 1 is characterized in that: described color initialization is that subjective vision characteristics according to the people are the suitable color of goal-setting in medical image.
3. method as claimed in claim 2, it is characterized in that: described Region Segmentation adopts watershed algorithm.
4. method as claimed in claim 3, it is characterized in that: the step of described Region Segmentation comprises:
Adopt the Gaussian smoothing operator to carry out smoothing processing to original image;
The gradient of each point in computed image, two-dimensional gradient are calculated and are got both direction gradient square root sum square, and the scanning entire image obtains the histogram of gradient image and the probability distribution of each gradient;
Travel through each pixel and draw each pixel in the position of sequence in array, pixel is sorted;
To the minimum layer of gradient, from top to bottom, from left to right each pixel is carried out mark;
Each pixel of each gradient layer of sequential loop obtains the watershed divide point;
Each closed region label is identical according to the watershed divide line that obtains is divided and is come, and realizes image segmentation.
5. method as claimed in claim 4, it is characterized in that: described Gaussian smoothing operator template is:
6. method as claimed in claim 4, it is characterized in that: described two-dimensional gradient calculates and adopts the sobel operator:
7. method as claimed in claim 4, it is characterized in that: described to the minimum layer of gradient, the principle of from top to bottom, from left to right each pixel being carried out mark is: if there is not the point of mark in four adjoint points up and down of current process points, current process points is new mark; If there is gauge point in four adjoint points up and down of current process points, current process points label equals this mark.
8. method as claimed in claim 4 is characterized in that: the described step of obtaining watershed divide point is, each pixel of each gradient layer of sequential loop carries out mark according to following criterion:
If there is not the point of mark in four adjoint points up and down of current process points, current process points is new mark;
If four adjoint points up and down of current process points only have a mark, this mark is assigned to current process points;
If there is two or more mark in four adjoint points up and down of current process points, current process points is the watershed divide point.
9. method as claimed in claim 4, it is characterized in that: whether described region clustering is with its merging according to the gray difference deciding degree between adjacent area, the combination in several zones has consisted of actual target area, wherein the process of zone merging is each zone to initial segmentation, other zones that searching is adjacent, if the difference of area grayscale mean value each other less than certain predetermined threshold T, merges them.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110424454.8A CN103164843B (en) | 2011-12-15 | 2011-12-15 | A kind of medical image colorize method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201110424454.8A CN103164843B (en) | 2011-12-15 | 2011-12-15 | A kind of medical image colorize method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN103164843A true CN103164843A (en) | 2013-06-19 |
| CN103164843B CN103164843B (en) | 2015-11-25 |
Family
ID=48587902
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201110424454.8A Active CN103164843B (en) | 2011-12-15 | 2011-12-15 | A kind of medical image colorize method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN103164843B (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105118076A (en) * | 2015-07-30 | 2015-12-02 | 上海应用技术学院 | Image colorization method based on over-segmentation and local and global consistency |
| CN107862726A (en) * | 2016-09-20 | 2018-03-30 | 西门子保健有限责任公司 | Color 2 D film medical imaging based on deep learning |
| CN109978964A (en) * | 2019-03-19 | 2019-07-05 | 广东智媒云图科技股份有限公司 | A kind of image formation method, device, storage medium and terminal device |
| CN112884866A (en) * | 2021-01-08 | 2021-06-01 | 北京奇艺世纪科技有限公司 | Coloring method, device, equipment and storage medium for black and white video |
| CN116934905A (en) * | 2023-09-18 | 2023-10-24 | 晨达(广州)网络科技有限公司 | Real-time processing method for network image |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101299277A (en) * | 2008-06-25 | 2008-11-05 | 北京中星微电子有限公司 | Method and system for colorizing black and white picture |
| CN102236886A (en) * | 2010-04-22 | 2011-11-09 | 秦跃有限公司 | System for enhancing contrastive gray scale manipulation of medical image and realizing colorization |
-
2011
- 2011-12-15 CN CN201110424454.8A patent/CN103164843B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101299277A (en) * | 2008-06-25 | 2008-11-05 | 北京中星微电子有限公司 | Method and system for colorizing black and white picture |
| CN102236886A (en) * | 2010-04-22 | 2011-11-09 | 秦跃有限公司 | System for enhancing contrastive gray scale manipulation of medical image and realizing colorization |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105118076A (en) * | 2015-07-30 | 2015-12-02 | 上海应用技术学院 | Image colorization method based on over-segmentation and local and global consistency |
| CN105118076B (en) * | 2015-07-30 | 2017-12-01 | 上海应用技术学院 | Based on over-segmentation and the local image colorization method with global coherency |
| CN107862726A (en) * | 2016-09-20 | 2018-03-30 | 西门子保健有限责任公司 | Color 2 D film medical imaging based on deep learning |
| CN107862726B (en) * | 2016-09-20 | 2021-04-23 | 西门子保健有限责任公司 | Deep learning based medical imaging of two-dimensional color film |
| CN109978964A (en) * | 2019-03-19 | 2019-07-05 | 广东智媒云图科技股份有限公司 | A kind of image formation method, device, storage medium and terminal device |
| CN109978964B (en) * | 2019-03-19 | 2023-01-13 | 广东智媒云图科技股份有限公司 | Image making method and device, storage medium and terminal equipment |
| CN112884866A (en) * | 2021-01-08 | 2021-06-01 | 北京奇艺世纪科技有限公司 | Coloring method, device, equipment and storage medium for black and white video |
| CN112884866B (en) * | 2021-01-08 | 2023-06-06 | 北京奇艺世纪科技有限公司 | Coloring method, device, equipment and storage medium for black-and-white video |
| CN116934905A (en) * | 2023-09-18 | 2023-10-24 | 晨达(广州)网络科技有限公司 | Real-time processing method for network image |
| CN116934905B (en) * | 2023-09-18 | 2023-11-17 | 晨达(广州)网络科技有限公司 | Real-time processing method for network image |
Also Published As
| Publication number | Publication date |
|---|---|
| CN103164843B (en) | 2015-11-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US20220044375A1 (en) | Saliency Map Enhancement-Based Infrared and Visible Light Fusion Method | |
| CN102968782B (en) | In a kind of coloured image, remarkable object takes method automatically | |
| CN102005054B (en) | Real-time infrared image target tracking method | |
| CN105654091A (en) | Detection method and apparatus for sea-surface target | |
| CN104463804B (en) | Image enhancement method based on intuitional fuzzy set | |
| CN104751142A (en) | Natural scene text detection algorithm based on stroke features | |
| CN103164843B (en) | A kind of medical image colorize method | |
| CN113762201B (en) | Mask detection method based on yolov4 | |
| CN110363719A (en) | A kind of cell layered image processing method and system | |
| CN109543632A (en) | A kind of deep layer network pedestrian detection method based on the guidance of shallow-layer Fusion Features | |
| CN103020933B (en) | A kind of multisource image anastomosing method based on bionic visual mechanism | |
| CN103927727A (en) | Method for converting scalar image into vector image | |
| CN105931180A (en) | Salient information guided image irregular mosaic splicing method | |
| CN103745470A (en) | Wavelet-based interactive segmentation method for polygonal outline evolution medical CT (computed tomography) image | |
| CN113177503A (en) | Arbitrary orientation target twelve parameter detection method based on YOLOV5 | |
| Yang et al. | Multi-source medical image fusion based on wasserstein generative adversarial networks | |
| Xia et al. | Cervical cancer cell detection based on deep convolutional neural network | |
| CN111223110A (en) | Microscopic image enhancement method and device and computer equipment | |
| CN115564782A (en) | 3D blood vessel and trachea segmentation method and system | |
| Chen et al. | A new watershed algorithm for cellular image segmentation based on mathematical morphology | |
| CN102096913A (en) | Multi-strategy image fusion method under compressed sensing framework | |
| Wang et al. | Rca-cyclegan: unsupervised underwater image enhancement using red channel attention optimized cyclegan | |
| CN104778683B (en) | A kind of multi-modality images dividing method based on Functional Mapping | |
| CN109542145A (en) | A kind of light path automatically collimating method of device of high power laser | |
| JP2014106713A (en) | Program, method, and information processor |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| C56 | Change in the name or address of the patentee | ||
| CP01 | Change in the name or title of a patent holder |
Address after: A 530 building Taihu international science and Technology Park in Jiangsu province Wuxi District Qingyuan Road 214135 10 floor Patentee after: WUXI ZHONGGAN MICROELECTRONIC CO., LTD. Address before: A 530 building Taihu international science and Technology Park in Jiangsu province Wuxi District Qingyuan Road 214135 10 floor Patentee before: Wuxi Vimicro Co., Ltd. |