CN117252875A - Medical image processing method, system, medium and equipment based on hyperspectral image - Google Patents
Medical image processing method, system, medium and equipment based on hyperspectral image Download PDFInfo
- Publication number
- CN117252875A CN117252875A CN202311531442.4A CN202311531442A CN117252875A CN 117252875 A CN117252875 A CN 117252875A CN 202311531442 A CN202311531442 A CN 202311531442A CN 117252875 A CN117252875 A CN 117252875A
- Authority
- CN
- China
- Prior art keywords
- image
- hyperspectral
- channel
- blood
- image processing
- 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
- 238000003672 processing method Methods 0.000 title claims abstract description 20
- 239000008280 blood Substances 0.000 claims abstract description 67
- 210000004369 blood Anatomy 0.000 claims abstract description 67
- 238000002310 reflectometry Methods 0.000 claims abstract description 37
- 238000012545 processing Methods 0.000 claims abstract description 24
- 238000002329 infrared spectrum Methods 0.000 claims abstract description 23
- 230000002159 abnormal effect Effects 0.000 claims abstract description 18
- 238000012512 characterization method Methods 0.000 claims abstract description 17
- 238000001429 visible spectrum Methods 0.000 claims abstract description 11
- 230000003595 spectral effect Effects 0.000 claims description 17
- 238000003860 storage Methods 0.000 claims description 13
- 238000004458 analytical method Methods 0.000 claims description 12
- 239000003086 colorant Substances 0.000 claims description 12
- 238000004590 computer program Methods 0.000 claims description 10
- 238000012360 testing method Methods 0.000 claims description 6
- 230000002708 enhancing effect Effects 0.000 claims description 5
- 238000002474 experimental method Methods 0.000 claims description 5
- 238000010606 normalization Methods 0.000 claims description 5
- 238000005070 sampling Methods 0.000 claims description 5
- 230000035945 sensitivity Effects 0.000 claims description 4
- 238000000034 method Methods 0.000 abstract description 39
- 230000008569 process Effects 0.000 abstract description 19
- 230000001965 increasing effect Effects 0.000 abstract description 4
- 230000003287 optical effect Effects 0.000 abstract description 4
- 239000003550 marker Substances 0.000 abstract 1
- 206010028980 Neoplasm Diseases 0.000 description 18
- 238000009499 grossing Methods 0.000 description 14
- 238000012216 screening Methods 0.000 description 14
- 230000006870 function Effects 0.000 description 11
- 238000009826 distribution Methods 0.000 description 7
- 238000010586 diagram Methods 0.000 description 6
- 239000003814 drug Substances 0.000 description 4
- 230000008859 change Effects 0.000 description 3
- 238000012937 correction Methods 0.000 description 3
- 229940079593 drug Drugs 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 241000167854 Bourreria succulenta Species 0.000 description 2
- 206010034972 Photosensitivity reaction Diseases 0.000 description 2
- 238000000701 chemical imaging Methods 0.000 description 2
- 235000019693 cherries Nutrition 0.000 description 2
- 238000010219 correlation analysis Methods 0.000 description 2
- 230000036211 photosensitivity Effects 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- 206010009944 Colon cancer Diseases 0.000 description 1
- 208000001333 Colorectal Neoplasms Diseases 0.000 description 1
- 206010058467 Lung neoplasm malignant Diseases 0.000 description 1
- 208000005718 Stomach Neoplasms Diseases 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 201000011510 cancer Diseases 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000593 degrading effect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000003745 diagnosis Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 206010017758 gastric cancer Diseases 0.000 description 1
- 201000005202 lung cancer Diseases 0.000 description 1
- 208000020816 lung neoplasm Diseases 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000877 morphologic effect Effects 0.000 description 1
- 230000001575 pathological effect Effects 0.000 description 1
- 238000010183 spectrum analysis Methods 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
- 201000011549 stomach cancer Diseases 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
- 208000024891 symptom Diseases 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0012—Biomedical image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/90—Determination of colour characteristics
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10024—Color image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10048—Infrared image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30096—Tumor; Lesion
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30004—Biomedical image processing
- G06T2207/30101—Blood vessel; Artery; Vein; Vascular
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A40/00—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production
- Y02A40/10—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in agriculture
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- Radiology & Medical Imaging (AREA)
- Quality & Reliability (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention relates to the technical field of hyperspectral images, in particular to a medical image processing method, a medical image processing system, a medical image processing medium and medical image processing equipment based on hyperspectral images. The hyperspectral image processing method comprises the following steps: adjusting the image acquisition parameters according to the relation between the abnormal external representation of blood and the image acquisition parameters to obtain a hyperspectral image of a blood sample; performing color standardization on the hyperspectral image according to the relation between the reflectivity calibration parameter and the RGB color duty ratio of the image; and distributing the infrared spectrum of the hyperspectral image to a red channel, and distributing the visible spectrum to a green channel and a blue channel, so as to enhance the near infrared spectrum band information of the hyperspectral image. The method is applied to the blood hyperspectral image processing process, the influence of external characterization is eliminated, the influence of color on a judgment result is avoided, the optical information of the marker is increased, and the accuracy is improved.
Description
Technical Field
The invention relates to the technical field of hyperspectral images, in particular to a medical image processing method, a medical image processing system, a medical image processing medium and medical image processing equipment based on hyperspectral images.
Background
Medical images are images reflecting the internal structure of the human body, and are one of the main bases of modern medical diagnosis. Hyperspectral imaging is becoming an emerging technology for biomedical visualization, and research in the biomedical field is receiving attention. Hyperspectral imaging can capture the slight spectral differences of tissues under different pathological conditions, provide analysis information about the physiological, morphological and biochemical components of the tissues, and further provide more auxiliary information for medical analysis.
Collecting hyperspectral images of blood is one way to obtain diagnostic results by processing the hyperspectral images, but external characterization of blood can be affected by sample processing, collection time, drug interference and the like in the blood sample collection process. Meanwhile, the color of human blood has different manifestations under different states of the human body, common human blood comprises light red blood, dark purple blood, cherry red blood and the like, and the color of the collected hyperspectral image can influence the judgment result.
Disclosure of Invention
Aiming at the defects existing in the prior art, the embodiment of the invention aims to provide a medical image processing method based on hyperspectral images so as to solve the influence of parameter selection on the hyperspectral images.
In order to achieve the above object, the embodiment of the present invention provides the following technical solutions:
a medical image processing method based on hyperspectral images, comprising:
adjusting the image acquisition parameters according to the relation between the abnormal external representation of blood and the image acquisition parameters to obtain a hyperspectral image of a blood sample;
performing color standardization on the hyperspectral image according to the relation between the reflectivity calibration parameter and the RGB color duty ratio of the image;
and distributing the infrared spectrum of the hyperspectral image to a red channel, and distributing the visible spectrum to a green channel and a blue channel, so as to enhance the near infrared spectrum band information of the hyperspectral image.
Optionally, adjusting the image acquisition parameter includes:
establishing a hyperspectral image database, wherein the hyperspectral image database comprises: hyperspectral images acquired for blood samples under different preset conditions;
carrying out blood external characterization analysis on the hyperspectral image, and extracting information related to abnormal external characterization;
determining the association relation between abnormal external blood characterization and image acquisition parameters;
and adjusting image acquisition parameters according to the association relation.
Optionally, the image acquisition parameters include light source type intensity, exposure time of the camera, spectral resolution, photosensitivity and sampling mode.
Optionally, performing color normalization on the hyperspectral image includes:
collecting blood hyperspectral images with different colors and under a collecting state, carrying out RGB color ratio analysis on all hyperspectral images, setting different types of control tests to carry out reflectivity calibration experiments, determining the relation between reflectivity calibration parameters and the RGB color ratio of the images, setting standardized RGB color ratio, and adjusting the RGB color ratio of the hyperspectral images to be in an RGB color ratio space of normal blood.
Optionally, the setting of the different types of control tests includes:
the relationship between the ratio of the three colors of RGB and the related reflectivity calibration parameters, and the relationship between the ratio of the two colors of RGB and the related reflectivity calibration parameters.
Alternatively, the infrared spectrum is assigned to the red channel, which is green and the green channel is blue.
Optionally, enhancing the hyperspectral image near infrared spectrum band information includes:
reading an infrared image and a visible light image, and loading the infrared image and the visible light image into a memory;
creating a new image according to the sizes and the channel numbers of the infrared image and the visible light image, and storing the channel allocated result;
traversing each pixel of the infrared image for the red channel, assigning a pixel value of the infrared image to the red channel of the new image; traversing each pixel of a red light channel of the visible light image for a green channel, and assigning a pixel value thereof to the green channel of the new image; traversing each pixel of a green light channel of the visible light image for the blue channel, and assigning a pixel value thereof to the blue channel of the new image;
and according to the data type and the display requirement of the image, adjusting the range of the pixel value of the new image, and storing the image.
According to other embodiments, the present disclosure further adopts the following technical solutions:
a hyperspectral image based medical image processing system, comprising:
the image acquisition module is configured to adjust the image acquisition parameters according to the relation between the abnormal external representation of blood and the image acquisition parameters, and acquire a hyperspectral image of the blood sample;
a color normalization module configured to color normalize the hyperspectral image according to a relationship between a reflectance calibration parameter and an image RGB color duty cycle;
and the information enhancement module is configured to allocate the infrared spectrum of the hyperspectral image to a red channel, allocate the visible spectrum to a green channel and a blue channel and enhance the near infrared spectrum band information of the hyperspectral image.
According to other embodiments, the present disclosure further adopts the following technical solutions:
a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, performs the hyperspectral image processing method as described above.
According to other embodiments, the present disclosure further adopts the following technical solutions:
a terminal device comprising a processor and a computer readable storage medium, the processor configured to implement instructions; the computer readable storage medium is for storing a plurality of instructions adapted to be loaded by a processor and to perform the hyperspectral image processing method as described above.
One or more technical solutions provided in the embodiments of the present invention at least have the following technical effects or advantages:
according to the method, the influence relation between abnormal external representation of blood and adjustment of the acquisition parameters is determined through experiments, and the acquisition parameters for acquiring the hyperspectral image of the blood are set according to the influence relation. And then comparing the color value of the acquired image with the color value under the standard light source according to the principle that the hyperspectral images with different spectral reflectivities show different colors, analyzing the relation between the spectral reflectivities and the adjustment of the color values, and adjusting according to the reflectance parameters to obtain the blood hyperspectral image with standard color distribution under normal conditions, namely the color standardized hyperspectral image. Finally, an image processing technology combining infrared and visible light wave bands is used for enhancing infrared information and combining the infrared information and the visible light information. When the method is applied to the blood hyperspectral image processing process, the influence of external characterization on the tumor early screening process is eliminated, meanwhile, the influence of color on the tumor early screening judgment result is avoided, the optical information of tumor markers in the tumor early screening process is increased, and the accuracy of the tumor early screening is improved.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the invention.
FIG. 1 is a flowchart of a hyperspectral image processing method according to an embodiment of the present invention;
FIG. 2 is a comparison of the color registration of a hyperspectral image provided by the first embodiment of the present invention;
the mutual spacing or dimensions are exaggerated for the purpose of showing the positions of the various parts, and the schematic illustrations are used for illustration only.
Detailed Description
It should be noted that the following detailed description is exemplary and is intended to provide further explanation of the invention. Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. It is noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of exemplary embodiments in accordance with the present disclosure. As used herein, the singular is also intended to include the plural unless the context clearly indicates otherwise, and furthermore, it is to be understood that the terms "comprises" and/or "comprising" when used in this specification are taken to specify the presence of stated features, steps, operations, devices, components, and/or combinations thereof.
As described in the background, the selection of parameters affects the hyperspectral image, and the following description will take early tumor screening as an example:
the early screening of tumors means that a rapid and simple method is used for screening a very small number of high-risk tumor groups from a large number of target groups which look healthy and have not developed symptoms, so that tumors can be found early, the risk of morbidity is reduced, and especially cancer seeds with high morbidity and mortality and long development period, such as lung cancer, gastric cancer, colorectal cancer and the like, can be found early.
The hyperspectral image of blood is collected, and the hyperspectral image is processed to obtain a tumor early screening result, but the external representation of the blood can be influenced by the conditions of sample processing, collection time, medicine interference and the like in the blood sample collection process. Meanwhile, the color of human blood has different manifestations under different states of the human body, common human blood comprises light red blood, dark purple blood, cherry red blood and the like, and the color of the collected hyperspectral image can influence the judgment result of early screening of tumors.
Example 1
As shown in fig. 1, the present embodiment proposes a medical image processing method based on hyperspectral image, which is applied to blood hyperspectral image processing process, including:
step S100: adjusting the image acquisition parameters according to the relation between the abnormal external representation of blood and the image acquisition parameters to obtain a hyperspectral image of a blood sample, wherein the method specifically comprises the following steps:
step S101: and establishing a hyperspectral image database. Wherein the hyperspectral image database comprises: and hyperspectral images acquired for blood samples under different preset conditions.
The method comprises the steps of establishing a blood hyperspectral image database collected under different collection conditions, adding specific medicines and the like, collecting hyperspectral images of each blood sample, and recording corresponding collection parameters, wherein the image collection parameters comprise parameters which influence the quality and characteristics of the collected hyperspectral images, such as light source type intensity, exposure time of a camera, spectral resolution, photosensitivity, sampling mode and the like.
Step S102: analysis of external characterization of blood. And carrying out blood external characterization analysis on the hyperspectral image, and extracting information related to abnormal external characterization.
Using the collected hyperspectral image, an external characterization of the blood is analyzed. This may include using image processing and analysis algorithms, such as spectral analysis, feature extraction, etc., to extract information related to blood anomaly characteristics.
Step S103: parameter adjustment and correlation analysis. And determining the association relation between the abnormal external representation of blood and the image acquisition parameters.
Through experimental data analysis, the relationship between abnormal external characteristics of blood (namely abnormal characteristics of blood) and acquisition parameters is determined. This may involve methods of statistical analysis, machine learning, etc. to determine which acquisition parameters are relevant to a particular external characterization, and the degree of correlation between them.
The relationship between the abnormal external characterization of blood and the acquisition parameters may be manifested in several aspects:
light source type intensity: a stronger light source may provide a higher signal-to-noise ratio, making the image clearer and more detailed, but an excessively strong light source may cause overexposure of the image, so that the details of the bright portion are lost, resulting in distortion of the image.
Exposure time of camera: longer exposure times may increase the light information in the image, helping to capture darker details, but too long exposure times may cause the image to blur or overexposure, lose detail, and fail to recover.
Spectral resolution: higher spectral resolution may provide more detailed spectral information that helps to distinguish features of different wavelengths, but too low a spectral resolution may result in aliasing and blurring of the spectral features that are difficult to accurately identify and analyze.
Sensitivity of: higher sensitivity can obtain brighter images under low light conditions, increasing the brightness of the image. While too high a sensitivity may introduce more noise, degrading the quality and accuracy of the image.
Sampling mode: different sampling patterns can affect the resolution, speed and cost of the image. The point-by-point scanning mode can obtain high-resolution images, but the speed is slower. Linear scanning or area scanning can increase acquisition speed but image resolution may be lower.
Step S104: and (5) automatically adjusting parameters. And adjusting image acquisition parameters according to the association relation.
Based on the result of the correlation analysis, a module for automatically adjusting the acquisition parameters is designed. The module can automatically adjust acquisition parameters according to the detected external representation of blood so as to optimize the quality of hyperspectral images and eliminate the influence of the external representation on the tumor early screening process. For example, if the external characterization indicates that the blood sample is subject to some drug disturbance, the module may automatically adjust the collection parameters to reduce the effects of such disturbance.
The parameter automatic adjustment is integrated on a chip by designing a function, a new acquisition parameter is output by the acquisition parameter and the acquired image parameter, the function in the chip is mainly used for comparing the acquisition parameter with the acquired image parameter, and the new acquisition parameter is output according to the relation between the acquisition parameter acquired in the steps and the acquired image parameter. The comparison of the two parameters is adopted to set the interval of the signal-to-noise ratio and the spectral resolution of the output function, and the new acquisition parameters are adjusted, so that hyperspectral images with similar parameters such as the signal-to-noise ratio and the spectral resolution of the output function are output. I.e. to normalize the above parameters. This chip may be integrated onto the hyperspectral camera lens.
Step S200: and carrying out color normalization on the hyperspectral image according to the relation between the reflectivity calibration parameter and the RGB color duty ratio of the image.
Collecting blood hyperspectral images with different colors and under a collecting state, carrying out RGB color ratio analysis on all hyperspectral images, carrying out a reflectivity calibration experiment, knowing the relation between the change of parameters (including reflectivity calibration parameters, reflectivity correction coefficients and curve smoothing parameters) and the change of the RGB color ratio of an original image in the reflectivity calibration process, setting a standardized RGB color ratio, and carrying out standardization on the colors of the hyperspectral images, namely, adjusting the RGB color ratio to be in the RGB color ratio space of normal blood. Wherein the reflectivity calibration parameters include a reflectivity correction coefficient and a curve smoothing number.
The realization process is to use the same standardized white board for collection and calibration, set different types of control tests, and respectively set the relation between the ratio of the three colors of RGB and the related reflectivity calibration parameters, and the relation control test between the condition of the ratio of the two colors of RGB and the related reflectivity calibration parameters. Finally, the relation between parameter adjustment and RGB color duty ratio change is obtained.
According to the relation between the obtained color duty ratio and parameter adjustment, a module for carrying out automatic color standardization on the hyperspectral image output by the steps is designed, the module can be integrated into a chip, the process is consistent with the acquisition parameter adjustment, matlab or python can be used for writing as an automatic parameter adjustment program, and finally the module is packaged.
The specific process of color standardization of each sample comprises the following steps:
s201, collecting standard white boards before collecting each batch of hyperspectral images, wherein the standard white boards mainly serve as background boards for reflectivity calibration, and the reflectivity calibration is carried out on the basis of the background boards, the standard white boards mainly serve as standardized white collecting boards, and calibration files for reflectivity calibration are generated after collection so as to carry out subsequent calibration.
S202, inputting the hyperspectral image obtained in the step S100 into an RGB color distribution program, generating an RGB distribution histogram, obtaining RGB duty ratio information according to the RGB distribution histogram, and outputting a numerical value, wherein R represents more red pixels, and B and G represent blue and green duty ratios equivalent to each other as shown in FIG. 2.
S203, carrying out reflectivity calibration according to the acquired RGB distribution histogram and RGB duty ratio information, wherein the reflectivity calibration can adjust the duty ratio information, and a standard reflection whiteboard in the reflectivity calibration provides an absolute reflectivity value (provided by national institute of metrology) to obtain a hyperspectral image with an adjusted color channel duty ratio.
Numerical calculation formula after reflectivity calibration:
。
where Iref is the calculated reflectance calibrated value, iraw is the raw data value for the given pixel, idark is the dark background value for the given pixel, and Iwhite is the intensity value for the whiteboard reference. The reflectance calibration procedure itself changes the RGB color channel duty cycle.
After the reflectivity calibration, the influence of a dark background value is reduced by adding a reflectivity correction coefficient (%), the whole reflectivity of the hyperspectral image is corrected, the hyperspectral data is further standardized, the value of a spectrum curve of the hyperspectral data is standardized to be between 0 and 1, the value of the emissivity of each pixel point is corrected to a certain extent, the occupation ratio difference of each reflected color channel is reduced, and the function of adjusting the occupation ratio of RGB color channels is achieved.
And S204, adjusting the smoothing parameters of the added curve, adjusting the hyperspectral image, reducing the influence of noise, highlighting the main trend and characteristics in the spectral line of the reflectivity through smoothing treatment, and smoothing the data, so that the subsequent analysis and treatment are more accurate and stable, the smoothing process can make the data more gentle and continuous, the complexity of the subsequent treatment is reduced, the further analysis and modeling are more accurate, and finally the color standardized hyperspectral image is obtained.
The curve smoothing adopts a Savitzky-Golay smoothing method, which is a smoothing method based on polynomial fitting, and can effectively reduce high-frequency noise in hyperspectral data while retaining main characteristics of a spectrum. The main steps of the Savitzky-Golay smoothing method comprise: 1. a window of fixed size is moved to cover each data point in the data. 2. Polynomial fitting is performed within each window, and polynomial coefficients are calculated. 3. The smoothed data points are calculated using the fitted coefficients.
The following is a formula for calculating the general form of Savitzky-Golay smoothing:
given a sequence (x 0, x1, x2, …, xn) containing observed data, smoothed data points (y 0, y1, y2, …, yn) are estimated by Savitzky-Golay smoothing. The general formula for the smoothing process is as follows:
。
where yi is the smoothed data point. xi+j is a data point within the window, where j takes on the value-m to m. cj is a polynomial coefficient, and is calculated by polynomial fitting according to data points in a window. These coefficients are typically pre-calculated and stored for direct use in practical applications. m represents the size of the fit window and can be selected as desired. A larger value of m will result in a stronger smoothing effect, but may also result in excessively smoothed data, losing some detail information. In general, the value of m depends on the nature of the data and the noise level.
Step S300: and distributing the infrared spectrum of the hyperspectral image to a red channel, and distributing the visible spectrum to a green channel and a blue channel, so as to enhance the near infrared spectrum band information of the hyperspectral image.
Information enhancement is performed on the near infrared spectrum band by the CIR color scheme technique, which assigns the infrared spectrum to the red channel and the visible spectrum to the green and blue channels, thereby forming a specific color display. Specifically, the infrared spectrum is assigned to the red channel, which is green, and the green channel is blue. By the distribution mode, the difference between the infrared spectrum and the visible spectrum is highlighted, so that information contained in the infrared spectrum is emphasized, and meanwhile, the distribution of the visible spectrum provides information in a visible light range in the CIR image, so that the image is more comprehensive. Therefore, when the CIR image is displayed, cells in blood can be red or magenta, and other substances are blue-green or cyan and are biased to red, so that the information quantity which can be represented by a near infrared spectrum band is enhanced, and the subsequent blood hyperspectral image research and application are facilitated. It will be appreciated that the selection of the channels to which the visible spectrum is assigned may be adjusted according to specific needs and applications, and that other color channels may sometimes be selected for assignment to achieve better visual effects or to highlight specific information.
The process for enhancing the infrared information specifically comprises the following steps:
step S301: the infrared image and the visible light image are read, and the infrared image and the visible light image are loaded into the memory by using corresponding functions in an image processing library or a programming language.
Step S302: creating a new image, and creating a new image according to the sizes and the channel numbers of the infrared image and the visible light image, wherein the new image is used for storing the channel distribution result.
Step S303: channel allocation: traversing each pixel of the infrared image for the red channel, assigning a pixel value of the infrared image to the red channel of the new image; traversing each pixel of a red light channel of the visible light image for a green channel, and assigning a pixel value thereof to the green channel of the new image; for the blue channel, each pixel of the green channel of the visible image is traversed, assigning its pixel value to the blue channel of the new image.
Step S304: adjusting the range according to the requirement: and according to the data type and the display requirement of the image, adjusting the range of the pixel value of the new image, and storing the image. For example, the pixel values are scaled to a range of 0 to 255 to facilitate proper display.
Wherein the corresponding function may be:
python (using PIL library or OpenCV library):
reading an image: image.open (), cv2.Imread (); creating a new image: image. New (), np. Zeros (); acquiring pixel values: img, getpixel (), img [ y, x ]; setting a pixel value: img, putpixel (), img [ y, x ] =value; saving an image: img.save (), cv2.Imwrite ().
MATLAB：
Reading an image: imread (); creating a new image: zeros (), uint8 (); acquiring pixel values: img (y, x,:); setting a pixel value: img (y, x,:) =value; saving an image: imwrite ().
According to the method, the influence relation between abnormal external representation of blood and adjustment of the acquisition parameters is determined through experiments, and the acquisition parameters for acquiring the hyperspectral image of the blood are set according to the influence relation. And then comparing the color value of the acquired image with the color value under the standard light source according to the principle that different spectral reflectivities can generate different colors, analyzing the relation between spectral reflectivities and color value adjustment, and adjusting according to the reflectivity parameters to obtain the blood hyperspectral image with the standard color under the normal condition. Finally, an image processing technology combining infrared and visible light wave bands is used for enhancing infrared information and combining the infrared information and the visible light information. When the method is applied to the blood hyperspectral image processing process, the influence of external characterization on the tumor early screening process is eliminated, meanwhile, the influence of color on the tumor early screening judgment result is avoided, the optical information of tumor markers in the tumor early screening process is increased, and the accuracy of the tumor early screening is improved.
Example two
Based on the same inventive concept, another embodiment of the present invention provides a medical image processing system based on hyperspectral image, including:
the image acquisition module is configured to adjust the image acquisition parameters according to the relation between the abnormal external representation of blood and the image acquisition parameters, and acquire a hyperspectral image of the blood sample;
a color normalization module configured to color normalize the hyperspectral image according to a relationship between a reflectance calibration parameter and an image RGB color duty cycle;
and the information enhancement module is configured to allocate the infrared spectrum of the hyperspectral image to a red channel, allocate the visible spectrum to a green channel and a blue channel and enhance the near infrared spectrum band information of the hyperspectral image.
For system embodiments, the description is relatively simple as it is substantially similar to method embodiments, and reference is made to the description of method embodiments for relevant points.
Example III
An object of the present embodiment is to provide a computer-readable storage medium.
A computer readable storage medium having stored thereon a computer program which when executed by a processor performs the steps of the method of embodiment one.
Example IV
In one embodiment of the disclosure, a terminal device is provided, including a processor and a computer readable storage medium, where the processor is configured to implement instructions; the computer readable storage medium is for storing a plurality of instructions adapted to be loaded by a processor and to perform the steps of the method of embodiment one.
It will be apparent to those skilled in the art that embodiments of the present disclosure may be provided as a method, system, or computer program product. Accordingly, the present disclosure may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present disclosure may take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) having computer-usable program code embodied therein.
The present disclosure is described with reference to flowchart illustrations and/or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the disclosure. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks.
These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks.
While the specific embodiments of the present disclosure have been described above with reference to the drawings, it should be understood that the present disclosure is not limited to the embodiments, and that various modifications and changes can be made by one skilled in the art without inventive effort on the basis of the technical solutions of the present disclosure while remaining within the scope of the present disclosure.
Claims (10)
1. A medical image processing method based on hyperspectral images, characterized by comprising:
adjusting the image acquisition parameters according to the relation between the abnormal external representation of blood and the image acquisition parameters to obtain a hyperspectral image of a blood sample;
performing color standardization on the hyperspectral image according to the relation between the reflectivity calibration parameter and the RGB color duty ratio of the image;
and distributing the infrared spectrum of the hyperspectral image to a red channel, and distributing the visible spectrum to a green channel and a blue channel, so as to enhance the near infrared spectrum band information of the hyperspectral image.
2. The hyperspectral image based medical image processing method as recited in claim 1, wherein adjusting the image acquisition parameters includes:
establishing a hyperspectral image database, wherein the hyperspectral image database comprises: hyperspectral images acquired for blood samples under different preset conditions;
carrying out blood external characterization analysis on the hyperspectral image, and extracting information related to abnormal external characterization;
determining the association relation between abnormal external blood characterization and image acquisition parameters;
and adjusting image acquisition parameters according to the association relation.
3. The hyperspectral image based medical image processing method as claimed in claim 2 wherein the image acquisition parameters include light source type intensity, exposure time of camera, spectral resolution, sensitivity and sampling mode.
4. The hyperspectral image based medical image processing method as recited in claim 1, wherein color normalizing the hyperspectral image includes:
collecting blood hyperspectral images with different colors and under a collecting state, carrying out RGB color ratio analysis on all hyperspectral images, setting different types of control tests to carry out reflectivity calibration experiments, determining the relation between reflectivity calibration parameters and the RGB color ratio of the images, setting standardized RGB color ratio, and adjusting the RGB color ratio of the hyperspectral images to be in an RGB color ratio space of normal blood.
5. The hyperspectral image based medical image processing method as claimed in claim 4, wherein the setting of different types of control tests includes:
the relationship between the ratio of the three colors of RGB and the related reflectivity calibration parameters, and the relationship between the ratio of the two colors of RGB and the related reflectivity calibration parameters.
6. The hyperspectral image based medical image processing method as claimed in claim 1, wherein the infrared spectrum is assigned to a red channel, the red channel is green, and the green channel is blue.
7. The hyperspectral image based medical image processing method as claimed in claim 1, wherein enhancing the hyperspectral image near infrared spectral band information includes:
reading an infrared image and a visible light image, and loading the infrared image and the visible light image into a memory;
creating a new image according to the sizes and the channel numbers of the infrared image and the visible light image, and storing the channel allocated result;
traversing each pixel of the infrared image for the red channel, assigning a pixel value of the infrared image to the red channel of the new image; traversing each pixel of a red light channel of the visible light image for a green channel, and assigning a pixel value thereof to the green channel of the new image; traversing each pixel of a green light channel of the visible light image for the blue channel, and assigning a pixel value thereof to the blue channel of the new image;
and according to the data type and the display requirement of the image, adjusting the range of the pixel value of the new image, and storing the image.
8. A hyperspectral image based medical image processing system, comprising:
the image acquisition module is configured to adjust the image acquisition parameters according to the relation between the abnormal external representation of blood and the image acquisition parameters, and acquire a hyperspectral image of the blood sample;
a color normalization module configured to color normalize the hyperspectral image according to a relationship between a reflectance calibration parameter and an image RGB color duty cycle;
and the information enhancement module is configured to allocate the infrared spectrum of the hyperspectral image to a red channel, allocate the visible spectrum to a green channel and a blue channel and enhance the near infrared spectrum band information of the hyperspectral image.
9. A computer-readable storage medium, on which a computer program is stored, which program, when being executed by a processor, performs the medical image processing method according to any one of claims 1 to 7.
10. A terminal device comprising a processor and a computer readable storage medium, the processor configured to implement instructions; a computer readable storage medium for storing a plurality of instructions adapted to be loaded by a processor and to perform the medical image processing method according to any of claims 1-7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311531442.4A CN117252875B (en) | 2023-11-17 | 2023-11-17 | Medical image processing method, system, medium and equipment based on hyperspectral image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311531442.4A CN117252875B (en) | 2023-11-17 | 2023-11-17 | Medical image processing method, system, medium and equipment based on hyperspectral image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117252875A true CN117252875A (en) | 2023-12-19 |
| CN117252875B CN117252875B (en) | 2024-02-09 |
Family
ID=89129843
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311531442.4A Active CN117252875B (en) | 2023-11-17 | 2023-11-17 | Medical image processing method, system, medium and equipment based on hyperspectral image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117252875B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117474815A (en) * | 2023-12-25 | 2024-01-30 | 山东大学 | Hyperspectral image calibration method and system |
| CN117830114A (en) * | 2024-01-02 | 2024-04-05 | 瀚湄信息科技(上海)有限公司 | Hemoglobin enhancement method and device based on white light LED illumination and electronic equipment |
Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006270334A (en) * | 2005-03-23 | 2006-10-05 | Olympus Corp | Shading correction method and image inspection device |
| CN103033512A (en) * | 2012-07-24 | 2013-04-10 | 南京农业大学 | Device and method for recognizing hatching egg incubation based on hyperspectrum |
| CN108090883A (en) * | 2018-01-04 | 2018-05-29 | 中煤航测遥感集团有限公司 | High spectrum image preprocess method, device and electronic equipment |
| CN110441249A (en) * | 2019-09-10 | 2019-11-12 | 四川轻化工大学 | Method for building pit mud total acid prediction model based on hyperspectral image technology |
| CN111655129A (en) * | 2018-02-02 | 2020-09-11 | 宝洁公司 | Hyperspectral imaging system and method of use |
| CN113409193A (en) * | 2021-06-18 | 2021-09-17 | 北京印刷学院 | Super-resolution reconstruction method and device for hyperspectral image |
| CN114878508A (en) * | 2022-05-07 | 2022-08-09 | 上海大学 | Hyperspectral imaging-based method for detecting surface water content of cultural relic in earthen site |
| CN114998109A (en) * | 2022-08-03 | 2022-09-02 | 湖南大学 | Hyperspectral imaging method, system and medium based on dual RGB image fusion |
| US20220292718A1 (en) * | 2021-03-11 | 2022-09-15 | Microsoft Technology Licensing, Llc | Fiducial marker based field calibration of a device |
| US11461899B1 (en) * | 2021-10-29 | 2022-10-04 | Guangdong Polytechnic Normal University | Method for detecting infection stage of anthracnose pathogenic with pre-analysis capacity |
| CN115728236A (en) * | 2022-11-21 | 2023-03-03 | 山东大学 | Hyperspectral image acquisition and processing system and working method thereof |
| CN115753691A (en) * | 2022-08-23 | 2023-03-07 | 合肥工业大学 | Water quality parameter detection method based on RGB reconstruction hyperspectrum |
| CN115901644A (en) * | 2022-12-14 | 2023-04-04 | 山东深蓝智谱数字科技有限公司 | Method for establishing wheat stripe rust spectrum library |
| CN116840233A (en) * | 2023-05-31 | 2023-10-03 | 重庆中烟工业有限责任公司 | Cigarette appearance detection method and device, electronic equipment and storage medium |
| CN116840110A (en) * | 2023-06-15 | 2023-10-03 | 正大青春宝药业有限公司 | Quality detection method for Guanxinning based on hyperspectral imaging technology and application |
-
2023
- 2023-11-17 CN CN202311531442.4A patent/CN117252875B/en active Active
Patent Citations (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006270334A (en) * | 2005-03-23 | 2006-10-05 | Olympus Corp | Shading correction method and image inspection device |
| CN103033512A (en) * | 2012-07-24 | 2013-04-10 | 南京农业大学 | Device and method for recognizing hatching egg incubation based on hyperspectrum |
| CN108090883A (en) * | 2018-01-04 | 2018-05-29 | 中煤航测遥感集团有限公司 | High spectrum image preprocess method, device and electronic equipment |
| CN111655129A (en) * | 2018-02-02 | 2020-09-11 | 宝洁公司 | Hyperspectral imaging system and method of use |
| CN110441249A (en) * | 2019-09-10 | 2019-11-12 | 四川轻化工大学 | Method for building pit mud total acid prediction model based on hyperspectral image technology |
| US20220292718A1 (en) * | 2021-03-11 | 2022-09-15 | Microsoft Technology Licensing, Llc | Fiducial marker based field calibration of a device |
| CN113409193A (en) * | 2021-06-18 | 2021-09-17 | 北京印刷学院 | Super-resolution reconstruction method and device for hyperspectral image |
| US11461899B1 (en) * | 2021-10-29 | 2022-10-04 | Guangdong Polytechnic Normal University | Method for detecting infection stage of anthracnose pathogenic with pre-analysis capacity |
| CN114878508A (en) * | 2022-05-07 | 2022-08-09 | 上海大学 | Hyperspectral imaging-based method for detecting surface water content of cultural relic in earthen site |
| CN114998109A (en) * | 2022-08-03 | 2022-09-02 | 湖南大学 | Hyperspectral imaging method, system and medium based on dual RGB image fusion |
| CN115753691A (en) * | 2022-08-23 | 2023-03-07 | 合肥工业大学 | Water quality parameter detection method based on RGB reconstruction hyperspectrum |
| CN115728236A (en) * | 2022-11-21 | 2023-03-03 | 山东大学 | Hyperspectral image acquisition and processing system and working method thereof |
| CN115901644A (en) * | 2022-12-14 | 2023-04-04 | 山东深蓝智谱数字科技有限公司 | Method for establishing wheat stripe rust spectrum library |
| CN116840233A (en) * | 2023-05-31 | 2023-10-03 | 重庆中烟工业有限责任公司 | Cigarette appearance detection method and device, electronic equipment and storage medium |
| CN116840110A (en) * | 2023-06-15 | 2023-10-03 | 正大青春宝药业有限公司 | Quality detection method for Guanxinning based on hyperspectral imaging technology and application |
Non-Patent Citations (3)
| Title |
|---|
| TAE BYUNG CHUN等: "Impact of Atmospheric Correction on the Ship Detection Using Airborne Hyperspectral Image", 《IGARSS 2019 - 2019 IEEE INTERNATIONAL GEOSCIENCE AND REMOTE SENSING SYMPOSIUM》, pages 2190 - 2192 * |
| 张宝萍: "基于多变量光谱数据分析方法的乳腺癌血清拉曼光谱特征研究", 《光谱学与光谱分析》, vol. 42, no. 03, pages 426 - 434 * |
| 胡光辉: "基于机器视觉哈密瓜外部品质与成熟度分析研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》, vol. 2015, no. 05, pages 138 - 1077 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117474815A (en) * | 2023-12-25 | 2024-01-30 | 山东大学 | Hyperspectral image calibration method and system |
| CN117474815B (en) * | 2023-12-25 | 2024-03-19 | 山东大学 | Hyperspectral image calibration method and system |
| CN117830114A (en) * | 2024-01-02 | 2024-04-05 | 瀚湄信息科技(上海)有限公司 | Hemoglobin enhancement method and device based on white light LED illumination and electronic equipment |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117252875B (en) | 2024-02-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN117252875B (en) | Medical image processing method, system, medium and equipment based on hyperspectral image | |
| US7646905B2 (en) | Scoring estrogen and progesterone receptors expression based on image analysis | |
| AU2003236675B2 (en) | Method for quantitative video-microscopy and associated system and computer software program product | |
| JP6086949B2 (en) | Image analysis method based on chromogen separation | |
| US9779503B2 (en) | Methods for measuring the efficacy of a stain/tissue combination for histological tissue image data | |
| US9418414B2 (en) | Image measurement apparatus, image measurement method and image measurement system | |
| WO2013080868A1 (en) | Image processing device, image processing method, and image processing program | |
| AU2003236675A1 (en) | Method for quantitative video-microscopy and associated system and computer software program product | |
| JP2004151101A (en) | Method and system for identifying object of interest in biological specimen | |
| JP5833631B2 (en) | Method for setting one or more ranges of one or more inspection parameters in an optical inspection system | |
| JP6872566B2 (en) | Processes and systems for identifying bacterial gram types | |
| CN107862659A (en) | Image processing method, device, computer equipment and computer-readable recording medium | |
| CN110495888B (en) | Standard color card based on tongue and face images of traditional Chinese medicine and application thereof | |
| JP5210571B2 (en) | Image processing apparatus, image processing program, and image processing method | |
| CN116509326A (en) | Tongue image multispectral image generation method, device, equipment and storage medium | |
| EP3053514A1 (en) | Organ imaging apparatus | |
| Kibria et al. | Smartphone-based point-of-care urinalysis assessment | |
| JP2009025147A (en) | Device and program for image processing | |
| CN112801112B (en) | Image binarization processing method, device, medium and equipment | |
| US20220392071A1 (en) | Image processing apparatus and image-based test strip identification method | |
| CN117849043A (en) | Urine test paper analysis device and detection method thereof | |
| Long et al. | A Practical Dental Image Enhancement Network for Early Diagnosis of Oral Dental Disease | |
| CN115824982A (en) | Optical POCT color interpretation method, system and device | |
| CN117969046A (en) | LED light source defect type detection method and system and electronic equipment | |
| CN115035286A (en) | Image processing method, image processing apparatus, terminal, and readable storage medium |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |