CN109615578A - Medical image interpolation processing method and device applied to various medical scenes - Google Patents
Medical image interpolation processing method and device applied to various medical scenes Download PDFInfo
- Publication number
- CN109615578A CN109615578A CN201811305785.8A CN201811305785A CN109615578A CN 109615578 A CN109615578 A CN 109615578A CN 201811305785 A CN201811305785 A CN 201811305785A CN 109615578 A CN109615578 A CN 109615578A
- Authority
- CN
- China
- Prior art keywords
- image
- psnr
- restored
- expected
- medical
- 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
- 238000003672 processing method Methods 0.000 title abstract description 14
- 238000012545 processing Methods 0.000 claims abstract description 99
- 238000011156 evaluation Methods 0.000 claims abstract description 37
- 238000000034 method Methods 0.000 claims description 51
- 238000004364 calculation method Methods 0.000 claims description 16
- 230000008569 process Effects 0.000 claims description 16
- 230000003044 adaptive effect Effects 0.000 claims description 5
- 238000003384 imaging method Methods 0.000 description 8
- 239000003814 drug Substances 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000009467 reduction Effects 0.000 description 3
- 230000008901 benefit Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 210000000813 small intestine Anatomy 0.000 description 2
- 208000027418 Wounds and injury Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000001413 cellular effect Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000013170 computed tomography imaging Methods 0.000 description 1
- 230000006378 damage Effects 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 239000012467 final product Substances 0.000 description 1
- 208000014674 injury Diseases 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 235000001968 nicotinic acid Nutrition 0.000 description 1
- 210000000056 organ Anatomy 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
- G06T3/4007—Scaling of whole images or parts thereof, e.g. expanding or contracting based on interpolation, e.g. bilinear interpolation
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16H—HEALTHCARE INFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR THE HANDLING OR PROCESSING OF MEDICAL OR HEALTHCARE DATA
- G16H30/00—ICT specially adapted for the handling or processing of medical images
- G16H30/20—ICT specially adapted for the handling or processing of medical images for handling medical images, e.g. DICOM, HL7 or PACS
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Radiology & Medical Imaging (AREA)
- Epidemiology (AREA)
- General Health & Medical Sciences (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Primary Health Care (AREA)
- Public Health (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a kind of Study of Medical Image Interpolation processing methods and its device applied to plurality of medical scene, it carries out color interpolation treated to go back the evaluation index parameter of original image by obtaining three kinds of common color interpolation algorithms in advance to sample image, so as to by the actual requirement parameter of itself and different medical scenes, as expected processing speed range goes back the prospective quality fraction range of original image and goes back the expection PSNR range of original image and is compared, to be matched to most suitable color interpolation algorithm, and then interpolation processing is carried out using the most suitable color interpolation algorithm.
Description
Technical field
The present invention relates to technical field of image processing, and with being related to, a kind of medical image applied to a variety of same medical scenes is slotting
It is worth processing method and its device.
Background technique
In Digital image technology, colorful optical filter array CFA technology is particularly significant, and sensor only passes through CFA ability
Generate image, structural reference bionics principle, for cellular optical filter.On CFA, a pixel has to three kinds substantially
One of color light.CFA colorful optical filter array generallys use RGB color model, and distribution pattern is green, blue or green, red.
In RGB color model, CFA mainly has Bayer type, stripe and Mosaic style three types.Wherein, have in Bayer format
Tri- Essential colour of RGB, compare other two kinds of colors, and human eye is more sensitive to green, thus in Bayer format G pixel quantity
It is the summation of other two kinds of pixels.And when color filter array acquisition image information, each pixel can only obtain a kind of component,
The component of other two kinds of missings then is estimated to obtain according to adjacent pixel value.This process is referred to as image interpolation.Pass through
Controlling computer and completing the algorithm of process of image interpolation is color interpolation algorithm.Color interpolation algorithm is adopted in medical digital images
There is very important application in the colored real-time display of collecting system.Classical interpolation algorithm has bilinear interpolation algorithm, color
Than constant method and adaptive-interpolation algorithm etc..
Color interpolation algorithm is mainly reflected in the reduction to medical image in the utilization in terms of medicine, is widely used in
In CT imaging, MRI imaging, ultrasonic imaging and fujinon electronic video endoscope imaging system, wherein again the most with fujinon electronic video endoscope imaging applications
Extensively.Main application equipment has infrared scope, fluorescence scope, confocal laser scope and contact Room Mirror etc..According to people
Body is by inspection position and by the difference of procuratorial organ's formula, and medical environment when instrument acquires medical image also differs widely, such as electronics small intestine
The examined position of mirror is behaved inside internal small intestine, and instrument need to carry light source and be shot into small enteron aisle, in point blank,
The phenomenon that inevitably generating image local exposure.This just needs to handle image using color interpolation algorithm, with
It avoids overexposure imaged image clarity and diagnose doctor can not.In terms of medicine, color interpolation algorithm is accurate
Medical diagnosis provide strong image information and support, also has for reduce that the secondary injury of patient provides in medical procedure
The guarantee of power.
But under different medical scenes, medical environment is different, and the requirement to interpolation algorithm is also different, therefore, not
With medical scene under the adaptable color interpolation algorithm of reasonable selection be very it is necessary to.In view of this, the present invention mentions
For a kind of medical image color interpolation processing method that can be applied to plurality of medical scene and its device, three kinds of base common color
Color interpolation process method, based on for bilinear interpolation algorithm, color ratio constant method and adaptive-interpolation algorithm, i.e., according to not
Same medical scene, recommends most suitable algorithm to handle medical image from these three color interpolation algorithms.
Summary of the invention
For the above technical problems, it is color to provide a kind of medical image that can be applied to plurality of medical scene by the present invention
Color interpolation process method.
In order to solve the above-mentioned technical problem, the technical solution adopted by the present invention are as follows:
A kind of Study of Medical Image Interpolation method applied to plurality of medical scene comprising step:
Sample image is obtained respectively after bilinear interpolation algorithm, color ratio constant method and adaptive-interpolation algorithm process
Three are gone back the evaluation index parameter of original image, and the evaluation index parameter includes processing speed, picture quality and image fidelity；
The demand parameter of image procossing needed for the current medical scene of user's input and the evaluation index parameter are carried out
Matching so that matching is suitble to the optimal color interpolation Processing Algorithm of the current medical scene, and is inserted using the optimal colour
Value Processing Algorithm carries out color interpolation processing to the image that user currently uploads；
Wherein, the demand parameter includes expected processing speed range, goes back the prospective quality fraction range of original image and go back
The expection PSNR range of original image.
Further, sample image is obtained respectively to calculate by bilinear interpolation algorithm, color ratio constant method and adaptive-interpolation
Method treated the step of going back the evaluation index parameter of original image, specifically include step:
Duration spent by individual sample image of three kinds of timing color interpolation algorithm process respectively, obtains processing speed；
The respective average mass fraction of original image is gone back described in calculating separately three, obtains picture quality；Wherein, the quality
Score is respectively by least 20 observer's typings；
The respective Y-PSNR PSNR of original image is gone back described in calculating separately three, obtains image fidelity.
Further, every average mass fraction for going back original imageCalculation formula are as follows:
Wherein, N is the total number of grades of mass fraction, N=5；Ci is the quality point of i-stage mass fraction
Number, i=1,2,3,4,5；Ki is the number that the observer of original image i-stage mass fraction is gone back described in imparting.
Further, the calculation formula of every Y-PSNR PSNR for going back original image are as follows:
Wherein, MSE is mean square error, and calculation isM is figure
The height of picture, N are the width of image, fc(m, n) andIt is sample head portrait respectively and goes back original image and set at (m, n) in place
The component value of Color Channel c.
Further, described by the demand parameter of image procossing needed for the current medical scene of user's input and institute's commentary
Valence index parameter compares, to obtain the step of being suitble to the color interpolation algorithm of the current medical scene, specifically includes
Step:
The mass fraction of original image is gone back described in judging three respectively whether in the prospective quality fraction range, Yi Jisuo
The Y-PSNR PSNR for going back original image is stated whether within the scope of the expected PSNR；
If wherein at least having the mass fraction for going back original image in the prospective quality fraction range, and peak value noise
Also again within the scope of the expected PSNR, then continue whether judgement described at least one gone back the corresponding processing speed of original image than PSNR
In the range of the expected processing speed, if so, obtaining the optimal color interpolation Processing Algorithm for being suitble to current medical scene.
Based on the above-mentioned Study of Medical Image Interpolation processing method applied to plurality of medical scene, the present invention also provides one kind
It can be applied to the Study of Medical Image Interpolation processing unit of plurality of medical scene comprising:
Evaluation parameter obtains module, is used to obtain respectively and passes through bilinear interpolation algorithm, color ratio constant method and adaptively insert
Value-based algorithm treated three evaluation index parameters for going back original image, the evaluation index parameter includes processing speed, image matter
Amount and image fidelity；
Interpolation processing module, the demand parameter of image procossing needed for the current medical scene for inputting user with it is described
Evaluation index parameter is matched, so that matching is suitble to the optimal color interpolation Processing Algorithm of the current medical scene, and benefit
Color interpolation processing is carried out to the image that user currently uploads with the optimal color interpolation Processing Algorithm；Wherein, the demand
Parameter includes expected processing speed range, goes back the prospective quality fraction range of original image and go back the expection PSNR range of original image.
Wherein, the evaluation parameter obtains module and specifically includes:
Timing unit, for distinguishing duration spent by individual sample image of three kinds of timing color interpolation algorithm process, i.e.,
Obtain processing speed；
Picture quality acquiring unit is obtained for going back the respective average mass fraction of original image described in calculating separately three
Picture quality；Wherein, the mass fraction is respectively by least 20 observer's typings；
Image fidelity acquiring unit, for going back the respective Y-PSNR PSNR of original image described in calculating separately three,
Obtain image fidelity.
Wherein, every average mass fraction for going back original imageCalculation formula are as follows:
Wherein, N is the total number of grades of mass fraction, N=5；Ci is the quality point of i-stage mass fraction
Number, i=1,2,3,4,5；Ki is the number that the observer of original image i-stage mass fraction is gone back described in imparting.
Wherein, the calculation formula of every Y-PSNR PSNR for going back original image are as follows:
Wherein, MSE is mean square error, and calculation isM is figure
The height of picture, N are the width of image, fc(m, n) andIt is sample head portrait respectively and goes back original image and set at (m, n) in place
The component value of Color Channel c.
Wherein, the interpolation processing module specifically includes:
Recommendation unit, for whether going back the mass fraction of original image in the prospective quality score described in judging three respectively
In range and whether the Y-PSNR PSNR for going back original image is within the scope of the expected PSNR；If wherein at least having
One is gone back the mass fraction of original image in the prospective quality fraction range, and the Y-PSNR PSNR also expection again
Within the scope of PSNR, then continues judgement described at least one and whether go back the corresponding processing speed of original image in the expected processing speed
In the range of, if so, obtaining the optimal color interpolation Processing Algorithm for being suitble to current medical scene；
Image processing unit, for being carried out in conjunction with the optimal color interpolation Processing Algorithm to the image that user currently uploads
Color interpolation processing.
The invention has the beneficial effects that:
The invention discloses a kind of Study of Medical Image Interpolation processing methods and its device applied to plurality of medical scene, lead to
It crosses to obtain three kinds of common color interpolation algorithms in advance and sample image is carried out color interpolation treated going back the evaluation of original image and refer to
Parameter is marked, so as to by the actual requirement parameter of itself and different medical scenes, processing speed range as expected goes back the pre- of original image
Phase mass fraction range and the expection PSNR range for going back original image are compared, and are calculated to be matched to most suitable color interpolation
Method, and then interpolation processing is carried out using the most suitable color interpolation algorithm.
Detailed description of the invention
Fig. 1 is an a kind of implementation of medical image color interpolation processing method applied to plurality of medical scene of the invention
The flow chart of example；
Fig. 2 a is sample image；
Fig. 2 b, Fig. 2 c and Fig. 2 d are respectively that sample image passes through bilinear interpolation algorithm, color ratio constant method and adaptively inserts
Value-based algorithm treated image；
Fig. 3 is an a kind of implementation of medical image color interpolation processing unit applied to plurality of medical scene of the invention
The functional block diagram of example.
Specific embodiment
With reference to the accompanying drawing, the present invention is described in detail.
In order to make the objectives, technical solutions, and advantages of the present invention clearer, with reference to the accompanying drawings and embodiments, right
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.
It is a kind of medical image color interpolation processing method applied to plurality of medical scene of the invention referring to Fig. 1
The flow chart of one embodiment, specifically, the Study of Medical Image Interpolation method in the present embodiment comprising steps of
S101 obtains sample image by bilinear interpolation algorithm, color ratio constant method and adaptive-interpolation algorithm respectively
Three after reason are gone back the evaluation index parameter of original image.
In the present embodiment, when in advance by a sample image pass through respectively bilinear interpolation algorithm, color ratio constant method and from
Interpolation algorithm processing is adapted to, when obtaining three and going back original image, it is also necessary to evaluate the quality for going back original image, the present embodiment
In, then using subjective estimate method and objective evaluation and every kind of color interpolation algorithm process single image institute time-consuming come pair
Also original image is evaluated, and obtains corresponding evaluation index parameter, comprising: processing speed, picture quality and image fidelity.
Wherein, processing speed then refers to every kind of color interpolation algorithm process single image institute time-consuming, therefore, can directly adopt
Three kinds of color interpolation algorithms are recorded with modes such as timers to handle a sample image respectively respectively time-consuming for institute, and are made
For one of evaluation index parameter.
Wherein, subjective estimate method refers to multiple observers (including professional observer and non-viewing person) according to international
Pyatyi quality yardstick and scale (referring to table one) is interfered, goes back the quality of original image to going back original image and evaluating in the form of mass fraction
Then quality calculates the average mark of mass fraction given by all observers, i.e., the average mark is as measurement also original image
The index of the picture quality of picture.Therefore, in the present embodiment, being averaged for mass fraction is provided using multiple observers (at least 20)
Score is evaluated original image is gone back, and by the average mark (or picture quality) as one of evaluation index parameter.
Wherein, objective evaluation, which refers to, obtains the method that a certain numerical value again evaluates image by mathematical computations, and
Gap size also between original image and sample image (i.e. original image), that is, the image fidelity for going back original image is in objective evaluation
It is most common, and the measurement standard of image fidelity is then Y-PSNR PSNR, therefore, is believed in the present embodiment using peak value
It makes an uproar than PSNR, i.e., image ratio is directed to and objectively evaluates to going back original image, and by Y-PSNR PSNR (or image fidelity
Degree) as one of evaluation index parameter.
The international Pyatyi quality yardstick of table one and obstruction scale table
| Mass fraction | Interfere scale | Quality yardstick |
| 5 | Picture quality quality is not seen at all | Very well |
| 4 | It can be seen that picture quality variation but do not interfere to watch | It is good |
| 3 | Picture quality is apparent to degenerate | Generally |
| 2 | Picture quality has obstruction to viewing | Difference |
| 1 | Picture quality has serious obstruction to viewing | It is very poor |
In one embodiment, above-mentioned steps S101 specifically includes step:
I, respectively duration spent by individual sample image of three kinds of timing color interpolation algorithm process.
II calculates separately three according to formula (1) and goes back the respective average mass fraction of original image (i.e. average mark), obtains
Picture quality.
In the present embodiment, the average value of every mass fraction for going back original imageCalculation formula are as follows:
Wherein, N is the total number of grades of mass fraction, N=5；Mass fraction of the ci for i-stage mass fraction, i=1,2,3,
4,5；Ki is the number that the observer of original image i-stage mass fraction is gone back described in imparting.In one embodiment, mass fraction
Provided respectively by least 22 observers, and the quantity of professional observer and amateur observer answer it is roughly equal.
III calculates separately three according to formula (2) and goes back the respective Y-PSNR PSNR of original image, obtains image fidelity
Degree.
In the present embodiment, the calculation formula of every Y-PSNR PSNR for going back original image are as follows:
Wherein, MSE is mean square error, and calculation isM is figure
The height of picture, N are the width of image, fc(m, n) andIt is sample head portrait respectively and goes back original image and set face at (m, n) in place
The component value of chrominance channel c.
S103 carries out the demand parameter of image procossing needed for the current medical scene of user's input and evaluation index parameter
Matching, to be matched to the optimal color interpolation Processing Algorithm of suitable current medical scene.
In the present embodiment, due to different medical scenes, it is different to the requirement for going back original image, carries out to medical image color
The demand of the algorithm of color interpolation processing is also different, and therefore, it is necessary to corresponding professionals according to different medical scene typing phases
The demand parameter answered, such as: it is expected that processing speed range, going back the prospective quality fraction range of original image and going back the expection of original image
PSNR range.
Wherein, which, which refers to, wishes to handle individual sample image institute time-consuming under Medical scene
It is long, the usual time-consuming a length of numerical intervals of institute.Such as real time imagery is needed under usual ultrasonic imaging scene, and therefore, the medicine
Expection processing speed under scene is 0.1-0.09s；And under CT scene, real time imagery is not needed, therefore, its settable expection
Processing speed is 0.5-5s.
Wherein, it is contemplated that mass fraction range, which refers to, wishes that individual sample image obtains after treatment under Medical scene
The specific mass fraction range going back original image and capable of reaching, namely go back the fine or not degree of original image.For example, ultrasonic imaging field
The prospective quality score for needing to go back original image under scape reaches 3-3.5, and needs to go back the prospective quality score of original image under CT scene
Reach 3.5-3.9.
Wherein, it is contemplated that PSNR range refer to wish that individual sample image under Medical scene obtains after treatment also
The specific expected PSNR that original image can reach, namely go back the fidelity of original image.For example, being needed under ultrasonic imaging scene also
The expection PSNR range of original image reaches 11-11.2G, and the prospective quality score for needing to go back original image under CT scene reaches
11.2-11.4G。
In one embodiment, step S103 specifically includes step:
The mass fraction of original image is gone back described in judging three respectively whether in prospective quality fraction range, and also original image
Whether the Y-PSNR PSNR of picture is within the scope of expected PSNR；For example, if
If wherein at least having the mass fraction for going back original image in the prospective quality fraction range, and peak value noise
Also again within the scope of the expected PSNR, then continue whether judgement described at least one gone back the corresponding processing speed of original image than PSNR
In the range of the expected processing speed, if so, obtaining the optimal color interpolation Processing Algorithm for being suitble to current medical scene.
S105 carries out the image that user currently uploads using optimal color interpolation Processing Algorithm obtained in step S103
Color interpolation processing.
It is illustrated below with reference to Study of Medical Image Interpolation processing method of the specific example to the present embodiment:
Sample image is passed through bilinear interpolation algorithm, color ratio constant method and adaptive-interpolation algorithm by a referring to fig. 2 respectively
Processing, obtains going back original image, referring to fig. 2 b, Fig. 2 c and Fig. 2 d；
Getting the processing speed that three are gone back original image is respectively 0.093s, 0.536s and 2.422s, 22 observers couple
It goes back the mass fraction that original image provides for three and every reduction is calculated as shown in following table two, and according to the mass fraction in table two
Corresponding average (quality) score of image, which is respectively as follows:, goes back the flat of original image (Fig. 2 b) by what bilinear interpolation algorithm was handled
Equal score is 3.14；It is 3.31 by the average mark for going back original image (Fig. 2 c) that color ratio constant method is handled；It is adaptive to insert
It is 3.73 that value-based algorithm was handled, which goes back the average mark of original image (Fig. 2 d),；It gets three and goes back R, G of original image, B peak
Signal-to-noise ratio calculates every reduction fidelity for going back original image as shown in following table three, and according to table three, i.e. Y-PSNR PSNR divides
Not are as follows: the Y-PSNR PSNR for going back original image (Fig. 2 b) handled by bilinear interpolation algorithm is 11.049G；By
What color ratio constant method was handled goes back the Y-PSNR PSNR of original image (Fig. 2 c) as 9.737G；Adaptive-interpolation algorithm process
The obtained Y-PSNR PSNR for going back original image (Fig. 2 d) is 11.274, finally obtains evaluation result as shown in following table four.
Two or three, table are gone back the image quality score table of original image
Three or three, table are gone back R, G of original image, B peak signal-to-noise ratio (db)
Four or three, table are gone back the evaluation result of original image
In one embodiment, it is 0.09- that the demand parameter of ultrasonic imaging scene, which is respectively as follows: expected processing speed,
0.1s, it is contemplated that mass fraction reaches 3-3.5, it is contemplated that PSNR range reaches 11-11.2G.By parameters and above-mentioned evaluation knot
Parameters in fruit table compare it is recognized that while, the PSNR of adaptive-interpolation algorithm is greater than in expected PSNR range most
Big value, average mass fraction are greater than the maximum value in prospective quality fraction range, and processing speed is less than expected processing speed model
Maximum value in enclosing, in other words, although what adaptive-interpolation algorithm process obtained goes back the picture quality of original image beyond expection
(i.e. very true to nature or distortion is seldom), and image fidelity also exceeds expection, but its processing speed is but unsatisfactory for demand, because
This, does not recommend；Similarly, the picture quality for going back original image that color ratio constant method is handled reaches expected, but its image fidelity
Degree does not reach expected but；And the picture quality, image fidelity and place for going back original image that bilinear interpolation algorithm is handled
Reason speed reaches expected, and therefore, under ultrasonic scene, recommendation bilinear interpolation algorithm is optimal/most suitable color interpolation
Processing method carries out interpolation processing to image.
In one embodiment, it is 0.5-5s that the demand parameter of CT or MRI field scape, which is respectively as follows: expected processing speed, in advance
Phase mass fraction reaches 3.5-3.9, it is contemplated that PSNR range reaches 11.2-11.4G.By parameters and above-mentioned evaluation result table
In parameters compare it is found that the picture quality for going back original image and image fidelity that color ratio constant method is handled are equal
Do not reach expected, therefore, does not recommend；What bilinear interpolation algorithm was handled goes back the picture quality and image fidelity of original image
Degree is not up to expected, and therefore, is not recommended；And the picture quality for going back original image that adaptive-interpolation algorithm process obtains reaches pre-
Phase, and image fidelity also reaches expected, processing speed also meet demand, therefore, it is recommended that adaptive-interpolation method be it is optimal/
Most suitable color interpolation processing method；I.e. under ultrasonic scene, it is optimal/most suitable colored slotting for recommending adaptive-interpolation method
It is worth processing method and interpolation processing is carried out to image.
Embodiment two
Based on the above-mentioned Study of Medical Image Interpolation processing method applied to plurality of medical scene, the present invention also provides one kind
Applied to the Study of Medical Image Interpolation processing unit of plurality of medical scene, it is carried out in the following with reference to the drawings and specific embodiments in detail
Explanation.
It is a kind of Study of Medical Image Interpolation processing unit applied to plurality of medical scene of the invention, specifically referring to Fig. 3
The Study of Medical Image Interpolation processing unit on ground, the present embodiment includes:
Evaluation parameter obtains module 31, for being obtained respectively by bilinear interpolation algorithm, color ratio constant method and adaptive
Interpolation algorithm treated three evaluation index parameters for going back original image；In the present embodiment, which includes processing
Speed, picture quality and image fidelity；
Interpolation processing module 32, the demand parameter of image procossing needed for the current medical scene for inputting user with
The evaluation index parameter stated is matched, so that it is matched to the optimal color interpolation Processing Algorithm of suitable current medical scene, and
Color interpolation processing is carried out to the image that user currently uploads using the optimal color interpolation Processing Algorithm；It, should in the present embodiment
Demand parameter includes expected processing speed range, goes back the prospective quality fraction range of original image and go back the expection PSNR model of original image
It encloses.
In one embodiment, which obtains module and specifically includes:
Timing unit 311, for distinguishing duration spent by individual sample image of three kinds of timing color interpolation algorithm process,
Processing speed is arrived to obtain the final product；
Picture quality acquiring unit 312 is obtained for going back the respective average mass fraction of original image described in calculating separately three
To picture quality；Specifically, every mass fraction for going back original image be respectively by least 20 observer's typings, and every also
The average mass fraction of original imageIt is the average value of the mass fraction provided by least 20 observers, calculation formula
Are as follows:Wherein, N is the total number of grades of mass fraction, N=5；Ci is the mass fraction of i-stage mass fraction, i
=1,2,3,4,5；Ki is the number that the observer of original image i-stage mass fraction is gone back described in imparting；
Image fidelity acquiring unit 313 is gone back the respective Y-PSNR PSNR of original image for calculating separately three, is obtained
To image fidelity；Specifically, the calculation formula of every Y-PSNR PSNR for going back original image are as follows:
Wherein, MSE is mean square error, and calculation isM is figure
The height of picture, N are the width of image, fc(m, n) andIt is sample head portrait respectively and goes back original image and set face at (m, n) in place
The component value of chrominance channel c.
In one embodiment, which specifically includes:
Recommendation unit 321, for whether going back the mass fraction of original image in above-mentioned prospective quality described in judging three respectively
In fraction range, and the Y-PSNR PSNR of original image is gone back whether within the scope of above-mentioned expected PSNR；If wherein at least having
One is gone back the mass fraction of original image in above-mentioned prospective quality fraction range, and the Y-PSNR PSNR also expection again
Within the scope of PSNR, then continue to judge that whether this at least one gone back the corresponding processing speed of original image in the range for being expected processing speed
It is interior, if so, obtaining the optimal color interpolation Processing Algorithm for being suitble to current medical scene；
Image processing unit 322, the optimal color interpolation Processing Algorithm for being matched in conjunction with recommendation unit is to user
The image currently uploaded carries out color interpolation processing.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all in essence of the invention
Made any modifications, equivalent replacements, and improvements etc., should all be included in the protection scope of the present invention within mind and principle.
Claims (10)
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811305785.8A CN109615578A (en) | 2018-11-05 | 2018-11-05 | Medical image interpolation processing method and device applied to various medical scenes |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811305785.8A CN109615578A (en) | 2018-11-05 | 2018-11-05 | Medical image interpolation processing method and device applied to various medical scenes |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109615578A true CN109615578A (en) | 2019-04-12 |
Family
ID=66003004
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811305785.8A Pending CN109615578A (en) | 2018-11-05 | 2018-11-05 | Medical image interpolation processing method and device applied to various medical scenes |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109615578A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114066828A (en) * | 2021-11-03 | 2022-02-18 | 深圳市创科自动化控制技术有限公司 | Image processing method and system based on multifunctional bottom layer algorithm |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101059867A (en) * | 2006-12-20 | 2007-10-24 | 哈尔滨工业大学（威海） | Support vector machine based image interpolation algorithm |
| CN107993196A (en) * | 2017-12-12 | 2018-05-04 | 苏州大学 | Image interpolation method and system based on prediction verification |
-
2018
- 2018-11-05 CN CN201811305785.8A patent/CN109615578A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101059867A (en) * | 2006-12-20 | 2007-10-24 | 哈尔滨工业大学（威海） | Support vector machine based image interpolation algorithm |
| CN107993196A (en) * | 2017-12-12 | 2018-05-04 | 苏州大学 | Image interpolation method and system based on prediction verification |
Non-Patent Citations (3)
| Title |
|---|
| 邓彩: "图像插值算法研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
| 陆志芳: "基于边缘信息的图像插值技术研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
| 高翔: "基于DCT域的图像插值技术研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114066828A (en) * | 2021-11-03 | 2022-02-18 | 深圳市创科自动化控制技术有限公司 | Image processing method and system based on multifunctional bottom layer algorithm |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Longere et al. | Perceptual assessment of demosaicing algorithm performance | |
| US8027533B2 (en) | Method of automated image color calibration | |
| CN104203081B (en) | Method for combining a plurality of eye images into a multi-focus full-light image | |
| CN103561629A (en) | Endoscope device | |
| US11436728B2 (en) | Endoscope system | |
| CN107396079B (en) | white balance adjusting method and device | |
| JP2001258044A (en) | Medical use image processing unit | |
| Li et al. | Deep learning approach for hyperspectral image demosaicking, spectral correction and high-resolution RGB reconstruction | |
| CN109615578A (en) | Medical image interpolation processing method and device applied to various medical scenes | |
| US11521319B2 (en) | Endoscope system | |
| CN109310299A (en) | Fujinon electronic video endoscope processor and electronic endoscope system | |
| CN109310303A (en) | Electronic endoscope processor and electronic endoscope system | |
| JP2000278710A (en) | Device for evaluating binocular stereoscopic vision picture | |
| Jin et al. | Softcopy quality ruler method: implementation and validation | |
| JP3969688B2 (en) | Binocular stereoscopic image evaluation apparatus, binocular stereoscopic image evaluation method, and recording medium | |
| Babcock et al. | Eye tracking observers during color image evaluation tasks | |
| EP2204980B1 (en) | Image processing system, imaging system, and microscope imaging system | |
| Shrestha et al. | Simultaneous multispectral imaging and illuminant estimation using a stereo camera | |
| Kambara et al. | Development of color correction system for medical images using color charts | |
| Babilon et al. | Spectral reflectance estimation of organic tissue for improved color correction of video-assisted surgery | |
| Winter et al. | Physically motivated enhancement of color images for fiber endoscopy | |
| JP2006030014A (en) | Color estimation system and color estimation method | |
| US20130003016A1 (en) | Calibration device for use with a fundus camera | |
| Constantinou et al. | A comparison of color correction algorithms for endoscopic cameras | |
| CN119055168B (en) | Image processing method for endoscope and endoscope |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190412 |