WO2023043036A1 - 영상 처리 방법 및 장치 - Google Patents
영상 처리 방법 및 장치 Download PDFInfo
- Publication number
- WO2023043036A1 WO2023043036A1 PCT/KR2022/010640 KR2022010640W WO2023043036A1 WO 2023043036 A1 WO2023043036 A1 WO 2023043036A1 KR 2022010640 W KR2022010640 W KR 2022010640W WO 2023043036 A1 WO2023043036 A1 WO 2023043036A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- ray image
- layer
- image processing
- map
- noise
- Prior art date
Links
- 238000003672 processing method Methods 0.000 title claims abstract description 21
- 238000012545 processing Methods 0.000 claims abstract description 86
- 238000000034 method Methods 0.000 claims abstract description 27
- 230000008569 process Effects 0.000 claims abstract description 23
- 230000002708 enhancing effect Effects 0.000 claims abstract description 9
- 238000013507 mapping Methods 0.000 claims abstract description 6
- 230000009466 transformation Effects 0.000 claims description 23
- 210000003484 anatomy Anatomy 0.000 claims description 16
- 238000006243 chemical reaction Methods 0.000 claims description 13
- 238000004458 analytical method Methods 0.000 claims description 9
- 238000010191 image analysis Methods 0.000 claims description 9
- 238000012937 correction Methods 0.000 claims description 8
- 238000009499 grossing Methods 0.000 claims description 4
- 230000001629 suppression Effects 0.000 claims description 4
- 230000008859 change Effects 0.000 claims description 3
- 238000010606 normalization Methods 0.000 claims description 3
- 230000001131 transforming effect Effects 0.000 claims description 3
- 239000010410 layer Substances 0.000 description 45
- 238000004891 communication Methods 0.000 description 20
- 238000010586 diagram Methods 0.000 description 12
- 230000006866 deterioration Effects 0.000 description 5
- 230000000694 effects Effects 0.000 description 5
- 238000003745 diagnosis Methods 0.000 description 4
- 238000011551 log transformation method Methods 0.000 description 3
- 230000005855 radiation Effects 0.000 description 3
- 230000003044 adaptive effect Effects 0.000 description 2
- 230000015556 catabolic process Effects 0.000 description 2
- 238000006731 degradation reaction Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 239000011229 interlayer Substances 0.000 description 2
- 210000000988 bone and bone Anatomy 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 210000004197 pelvis Anatomy 0.000 description 1
- 238000012805 post-processing Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 239000004984 smart glass Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- A—HUMAN NECESSITIES
- A61—MEDICAL OR VETERINARY SCIENCE; HYGIENE
- A61B—DIAGNOSIS; SURGERY; IDENTIFICATION
- A61B6/00—Apparatus or devices for radiation diagnosis; Apparatus or devices for radiation diagnosis combined with radiation therapy equipment
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/90—Dynamic range modification of images or parts thereof
- G06T5/92—Dynamic range modification of images or parts thereof based on global image properties
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
- G06T7/73—Determining position or orientation of objects or cameras using feature-based methods
-
- 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
- G06T2207/10116—X-ray 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
- G06T2207/20—Special algorithmic details
- G06T2207/20016—Hierarchical, coarse-to-fine, multiscale or multiresolution image processing; Pyramid transform
Definitions
- the present disclosure relates to image processing, and more particularly, to an image processing method and apparatus for reducing a radiation exposure dose or minimizing deterioration of image quality of an image obtained through radiation regardless of the dose.
- X-ray medical devices are devices that take pictures of the inside of a human body for radiographic examination.
- an image obtained from an X-ray medical device that is, an X-ray image
- an X-ray image has an uneven brightness level and is difficult to distinguish by region, making it difficult to precisely diagnose or accurately determine the X-ray image.
- an X-ray image usually has a wide dynamic range of 16 bits or more, and information on a region of interest is concentrated in a narrow region, making it difficult to apply conventional general image processing methods as they are.
- An object of the present disclosure is to provide an image processing method and apparatus for processing X-ray images taken with a minimum dose to prevent or minimize image quality deterioration.
- Another object of the present disclosure is to provide an image processing method and apparatus for preventing or minimizing image quality degradation of an X-ray image obtained regardless of exposure dose.
- the present disclosure processes an X-ray image obtained by considering noise characteristics in which details of a boosted image are considered according to a body part to be photographed, photographing conditions, etc., thereby preventing or minimizing image quality degradation. Another object is to provide an image processing method and apparatus.
- An X-ray image processing apparatus may include an image analyzer configured to analyze a received original X-ray image; a conversion unit which converts the original X-ray image into multi-scale and separates the original X-ray image in units of frequencies based on the analysis result; a noise prediction unit generating a noise prediction map by enhancing and combining specific frequency parts; an edge map processing unit generating an edge map based on the generated noise prediction map; a contrast processing unit enhancing contrast of the X-ray image based on the generated edge map; an inverse transform unit performing a flattening process and an edge correction process on the contrast-enhanced X-ray image and inversely transforming it; and a controller configured to tone-map the inversely transformed X-ray image and control output.
- An X-ray image processing method includes receiving an original X-ray image; analyzing the received original X-ray image; converting an original X-ray image into a multi-scale based on the result of the analysis and separating the original X-ray image in units of frequencies; generating a noise prediction map by enhancing and combining specific frequency parts; generating an edge map based on the generated noise prediction map; enhancing contrast of the X-ray image based on the generated edge map; performing a flattening process and an edge correction process on the contrast-enhanced X-ray image and inversely transforming the image; and tone-mapping and outputting the inversely transformed X-ray image.
- noise processing is primarily performed on the multi-scale converted X-ray image through various processes, and noise processing is additionally performed in the inverse transformation process to increase the accuracy of noise removal.
- noise processing is additionally performed in the inverse transformation process to increase the accuracy of noise removal.
- FIG. 1 is a schematic diagram of an image processing system according to an embodiment of the present invention.
- FIG. 2 illustrates an example of an actual implementation of the image processing system of FIG. 1 .
- FIG. 3 is a configuration block diagram of an image processing device according to an embodiment of the present invention.
- FIG. 4 is a flowchart illustrating an image processing method according to an embodiment of the present invention.
- FIG. 5 is a diagram for explaining image analysis according to an embodiment of the present invention.
- 6 and 7 are diagrams for explaining noise prediction according to an embodiment of the present invention.
- FIG. 8 is a diagram for explaining edge map generation according to an embodiment of the present invention.
- FIG. 10 is a diagram for explaining inverse transformation according to an embodiment of the present invention.
- FIG. 11 is a diagram for explaining inverse transform according to another embodiment of the present invention.
- FIG. 12 is a diagram for explaining noise and contrast control according to an embodiment of the present invention.
- FIG. 13 is a diagram for explaining an original image and a processed image according to an embodiment of the present invention.
- a noise map is generated by predicting noise characteristics for each brightness of an X-ray image, and edge detection, standard deviation, correlation between layers
- a more robust edge map can be created by utilizing various information such as correlation.
- noise included in the X-ray image may be minimized by improving the accuracy of noise removal by processing noise twice.
- noise prediction information when noise is removed, since noise prediction information, edge strength, and characteristics of each target part are considered together, a result more adaptive to the signal can be derived, and it is independent of low-dose X-ray images or dose, that is, It can provide consistent quality correction function even at high dose or standard dose.
- the 'images' described herein refer to radiation, in particular, X-ray images obtained from an X-ray machine, but are not limited thereto.
- the image processing method prevents deterioration of image quality while reducing the dose of exposure to the human body (eg, bones of a subject to be photographed) through X-rays of not only standard dose but also low dose, or Describe how to minimize it.
- FIG. 1 is a schematic diagram of an image processing system 1 according to an embodiment of the present invention.
- FIG. 2 illustrates an example of an actual implementation of the image processing system 1 of FIG. 1 .
- FIG. 3 is a block diagram of the image processing device 150 according to an embodiment of the present invention.
- an image processing system 1 may include an image capture device 100 , an image processing device 150 , and an image output device 180 .
- the image output device 180 may be one component of the image processing device 150 .
- the image capture device 100 may include an X-ray tube 110 and a Digital X-ray Detector (DXD) 120 .
- DXD Digital X-ray Detector
- the X-ray tube 110 radiates a certain amount of X-rays to a photographing target (eg, a body part).
- a photographing target eg, a body part
- the digital X-ray detector 120 detects an X-ray image based on the X-ray radiated onto the object to be photographed.
- the image capture device 100 may transmit an image detected through the digital X-ray detector 120 to the image processing device 150 .
- the image acquisition device 100 and the image processing device 150 may perform data communication through a wired/wireless network.
- the image processing device 150 may process the X-ray image received through the image capture device 100 and provide it through the display 180 .
- the image processing device 150 When an X-ray image is input, the image processing device 150 according to an embodiment of the present disclosure displays a background region, an anatomy region, a collimation region, a metal object, and the like.
- information is stored by segmenting, and it is multi-frequency (multi-frequency), that is, separated by frequency unit, and image analysis information is used to create a noise prediction map by frequency (or by brightness), predicted noise information, and image analysis information.
- Edge map generation, noise suppression and contrast boost for each layer and/or region are first performed using inter-layer correlation, etc., and then, brightness (object thickness) and additional
- edge correction considering noise removal and edge directionality (for example, in a specific frequency unit or in a specific layer)
- automatic contrast control is performed to maintain not only more natural picture quality but also consistency of picture quality, that is, consistency between images. can do.
- the image processing unit 160 includes a communication interface unit 310, an image analysis unit 320, a conversion unit 330, an enhancement unit 340, an inverse transformation unit 350, a control unit 360, and the like. It can be implemented including.
- the configuration shown in FIG. 3 may be partially omitted or modularized together with other components, or configurations not shown in FIG. 3 may be added.
- the communication interface unit 301 provides an interfacing environment for data communication with the image capture device 100 and may receive an X-ray image obtained from the image capture device 100 .
- the communication interface unit 301 may perform short range communication with the image capture device 100 .
- the communication interface unit 301 is Bluetooth (BluetoothTM), BLE (Bluetooth Low Energy), RFID (Radio Frequency Identification), infrared communication (Infrared Data Association; IrDA), UWB (Ultra Wideband), ZigBee, NFC ( Near Field Communication), Wi-Fi (Wireless-Fidelity), Wi-Fi Direct, and wireless USB (Wireless Universal Serial Bus) technologies may be used to support short-distance communication.
- the communication interface unit 310 may be used between the image acquisition device 100 and a wireless communication system, between the image processing device 150 and the display 180, or the image processing device 150 through wireless local area networks. ) and a network in which a server is located may support wireless communication.
- the image analyzer 320 may analyze an X-ray image received through the communication interface 310 .
- the image analysis is, for example, as shown in (b) of FIG.
- noise and edges may be separated with respect to the anatomy region 510 .
- the image processing device 150 may perform primary processing by suppressing the separated noise and further emphasizing the edge.
- the conversion unit 330 enhances the original X-ray image data or raw X-ray image data received through the communication interface unit 310 for each frequency unit characteristic. You can do multi-scale transformation so that you can do .
- the transform unit 330 may perform log transformation on the original X-ray image before performing the multi-scale transformation.
- the enhancement unit 340 may include a noise estimation unit 341, an edge map processing unit 343, a range control unit 345, a contrast processing unit 347, and the like.
- the conversion unit 330 and the enhancement unit 340 may be modularized.
- the enhancement unit 340 may be implemented in the form of a plurality of modules.
- a result of image analysis through the image analyzer 320 may be transferred to and utilized by at least one component of the enhancement unit 340 .
- the noise prediction unit 341 may generate noise prediction information from an X-ray image subjected to multi-scale conversion based on a result of analyzing the original X-ray image.
- the edge map processing unit 343 may generate an edge map based on a result of analyzing the original X-ray image.
- the range controller 345 performs range control for each layer of the multi-scale converted X-ray image based on the result of analyzing the original X-ray image and the noise prediction information generated by the noise predictor 341.
- the range control is, for example, related to the consistency of image quality, and as shown in FIG. 9, contrast normalization in certain ranges in which a result is derived using a reference value Therefore, it can be seen as an operation that enables consistent image quality processing regardless of dose.
- the reference value represents a reference value according to characteristics of each layer
- the reference value may be different in each layer.
- a value calculated in advance by an experiment or the like may be applied as the reference value.
- the contrast processor 347 may perform detail contrast enhancement through base layer contrast and noise control based on the analysis result of the original X-ray image.
- the inverse transformation unit 350 may inverse transform the X-ray image processed through the contrast processing unit 347 again.
- controller 360 may control the operation of the image processing unit 160 . Therefore, the control unit 360 can appropriately control the operation of the corresponding component through data communication with each component of the image processing unit 160 .
- the control unit 360 may control data received, generated, and processed through the data communication to be stored in the DB 170.
- the database (DB) 170 may temporarily store received X-ray images. Unlike shown in FIG. 1 , the database (DB) 170 does not necessarily have to be built into the image processing device 150, and may be located externally or remotely.
- the display 180 may receive and output an X-ray image processed through the image processing unit 160 .
- the display 180 may be in the form of a fixed device such as a monitor, TV, or signage, or a mobile device such as a mobile phone, a tablet PC, or a laptop computer.
- the display 180 may be a dedicated device for outputting an image according to the present disclosure.
- the display 180 is a wearable device capable of (or interlocking) exchanging data with another display device (not shown), such as a smart watch or smart glass. , a head-mounted display (HMD), and a mobile terminal such as a smart phone.
- another display device such as a smart watch or smart glass.
- HMD head-mounted display
- a mobile terminal such as a smart phone.
- the communication interface unit 310 may detect (or recognize) a communicable wearable device around the image processing device 150 .
- the controller 340 transmits at least a portion of data processed by the image processing device 150 to communication. It can be transmitted to the wearable device through the interface unit 301.
- a user of the wearable device may use data processed on the display 180 through the wearable device.
- the image processing system 1, the image acquisition device 100, or the image processing device 150 illustrated in FIGS. 1 and 3 are only one embodiment of the present disclosure. Some of the illustrated components may be integrated, added, or omitted according to specifications of an actually implemented system or device.
- two or more components may be combined into one component, or one component may be subdivided into two or more components as needed.
- functions performed by each component are for explaining an embodiment of the present invention, and the specific operation or device does not limit the scope of the present invention.
- FIG. 4 is a flowchart illustrating an image processing method according to an embodiment of the present invention.
- the image processing device 150 may analyze the original X-ray image obtained from the image capture device 100 through the received communication interface unit 301 (S101).
- the image processing device 150 segments the original X-ray image into an anatomy region 510, a direct exposure region 520, and a collimation region 530, as shown in (b) of FIG. 5, and the segmented Among the regions, the anatomy region 510 may be analyzed and the analysis result may be stored.
- the image processing device 150 applies a segmentation technique to the original X-ray image shown in (a) of FIG. 5 as shown in (b) of FIG.
- Region 520 and collimation region 530 can be predicted and a map can be generated, and then enhancement can be performed only for a desired region (eg, only for an edge region for anatomy region 510). there is.
- the image processing device 150 may first transform the original X-ray image (S103).
- the first transformation may mean, for example, log transformation, but is not limited thereto.
- the image processing device 150 may perform a second conversion process on the first converted X-ray image (S105).
- the second transformation process is, for example, as shown in FIG. 6, a multi-scale transformation in which n layers in frequency units are separated from the original X-ray image is an example, and the present disclosure It is not limited to this.
- the image processing device 150 may generate a noise prediction map from the second converted X-ray image.
- the image processing device 150 may use a Gaussian-Laplacian pyramid structure as shown in FIG. 6 for multi-scale transformation.
- the image processing device 150 may generate the noise prediction map as illustrated in FIGS. 6 and 7 .
- X-ray images may have different points that require processing or improvement according to target regions or target regions due to their characteristics.
- the image processing device 150 divides into frequency units on a multi-scale (or multi-frequency) basis and adjusts a parameter for each frequency to obtain, for example, suitable for a target region. image can be obtained.
- the present disclosure uses a Gaussian-Laplacian pyramid that can be used in a multi-scale conversion (or multi-frequency conversion) method, but is not limited thereto.
- an X-ray image may be converted into a root square.
- global contrast of a low signal level may be degraded.
- the use of the log transformation described above is one example, but is not limited thereto.
- the logarithmic transformation since the deviation of noise changes according to intensity, in the present disclosure, it is possible to predict the deviation of noise for each brightness and use it as a brightness variable prediction function.
- noise in another layer may be corrected based on information predicted in a layer having the greatest influence on noise, without independent prediction for each layer.
- the transform unit 330 may employ a Gaussian-Laplacian pyramid structure as shown in FIG. 6 for multi-scale transform processing, but is not limited thereto.
- the conversion unit 330 may form n layers (where n is a natural number), ie, a multi-layer structure, based on an original X-ray image.
- the n may be determined according to the setting of the image processing device 150. For example, if n is 10, a total of 10 images, that is, L1 to L10 may be processed.
- noise prediction may be performed using Gaussian information corresponding to brightness in layers corresponding to a high frequency region, that is, L1 to L3.
- the noise prediction unit 341 includes four noise prediction modules each predicting noise, but the number of the modules is not limited to what is shown. That is, the number of noise prediction modules included in the noise prediction unit 341 may be different from that shown.
- the output value Lap [0] of the L1 layer is input to the first noise estimation module (L0 noise estimation), and the output value Lap [1] of the L2 layer is input to the second noise estimation module. It is input to the noise estimation module (L1 noise estimation), the output value Lap [2] of the L3 layer is input to the third noise estimation module (L2 noise estimation), and the output value Lap [3] of the L4 layer is input to the fourth noise estimation module ( L3 noise estimation).
- the noise prediction unit 341 provides the noise information to the next noise prediction module based on the noise information predicted by the first noise prediction module, so that noise in another layer can be predicted.
- the noise information predicted by the first noise prediction module is used as a reference because the input of the first noise prediction module actually includes the most noise and thus has the greatest influence on noise prediction.
- the present disclosure is not limited thereto, and a noise prediction value in another layer may be further referred to as a noise reference value.
- the edge map processing unit 343 may generate an edge map based on the analysis result of the original X-ray image (S109).
- the contrast processor 347 may perform detail contrast enhancement through base layer contrast and noise control based on the analysis result of the original X-ray image (S111).
- the range controller 345 performs range control for each layer of the second converted X-ray image based on a result of analyzing the original X-ray image and noise prediction information generated by the noise predictor 341.
- the image processing device 150 may form a multi-featured edge map.
- the multi-feature may refer to information about noise, edge, contrast, layer, etc., but is not limited thereto.
- the edge map processing unit 343 may predict the noise level of each Laplacian using Gaussian information corresponding to brightness, anatomy, and information for each layer in the high-frequency layers (Layer 0 to 3 and 4).
- a local standard deviation value can be calculated, but when an edge map is generated only with the calculated local standard deviation value, the directional information of the edge is not considered and the area around the edge may be messy. can
- the image processing device 150 independently generates a map for each layer without considering inter-layer correlation, noise removal may be limited for higher frequency layers.
- a more robust map may be generated by reflecting local edge information and correlation between layers in a specific frequency unit, in addition to the calculated local standard deviation value.
- the correlation may be performed by combining an edge map and a contrast map, and dividing the combined value based on noise prediction information predicted in FIG. 7 . This may be, for example, to determine whether or not to suppress intensity (intensity) for a map according to noise.
- the image processing device 150 may perform map-based layer contrast & noise control & local contrast coherence processing.
- the image processing device 150 performs range adjustment according to the target criterion Tn of the Laplacian change rate for each region and layer defined in advance and adaptive Laplacian boost according to the edge map.
- the image processing device 150 may assign a weight to Laplacian values of each layer using an edge map.
- X-ray images are captured for each part (eg, chest, hand, pelvis, etc.) and for each frequency unit (layer0, 1, 2... .) There is an optimal range (Laplacian). Accordingly, the image processing device 150 may collect X-ray images taken with the most optimal dose for each region, and measure an appropriate range of the Laplacian for each layer of the collected X-ray images.
- only the value for the anatomy region predicted as a result of the image analysis in FIG. 5(b) may be selectively performed without measuring the entire region.
- the inverse transformation unit 350 may again perform a third transformation process on the X-ray image processed through the contrast processing unit 347 (S113).
- the third transformation process may represent, for example, inverse transformation.
- the inverse transform may be an inverse FFT (IFFT) corresponding to multi-scale transform.
- the image processing device 150 inversely transforms the multi-scale transformed X-ray image, which may include processes such as local contrast coherence, noise reduction, and edge correction.
- the image processing device 150 may, for example, perform histogram-based equalization on an intermediate layer (intermediate frequency) for contrast consistency.
- the image processing device 150 may perform a smoothing operation while adaptively preserving edges of a signal that is not smooth due to boosting in a low layer (high frequency) by utilizing anatomy, brightness, edge map information, and the like.
- the inverse transform unit 350 creates a final result image by sequentially adding the corrected frequency components from the higher layer Ln, for example, as shown in FIG. 10 or 11 .
- the inverse transformation unit 350 solves this problem, since even if the frequency components are boosted so that the X-ray image is added to the brightness for each layer (consistency), if the brightness is too bright or dark, the overall contrast in the final X-ray image may not be flattened. To do this, the brightness and contrast of the image may be flattened based on the histogram in the aforementioned intermediate frequency layer.
- the inverse transform unit 350 maintains the consistency of the output X-ray image regardless of the dose through the flattening and can prevent noise caused by high frequencies from being boosted.
- the noise can be amplified again in the step of combining each layer in the inverse transformation process. Additional noise removal may be separately performed on layers L2, L1, and L0 having a large noise effect.
- the separate noise removal may include smoothing in consideration of a small size noise or unnatural signal directionality due to the above-described boost operation, rather than removing a relatively large size noise.
- the controller 360 may control the output by performing a tone mapping operation on the inversely transformed X-ray image (S115).
- the tone mapping may indicate that, for example, an X-ray image itself is a 16-bit image and thus has a wide range, whereas the display 180 is 8-bit and thus has a mismatched range.
- FIG. 13 shows original X-ray images and post-processed X-ray images before processing according to the present invention. represents the post-processing X-ray image according to the present invention described above.
- the image processing apparatus has an effect of minimizing deterioration in image quality of a low-dose X-ray image and guaranteeing or providing consistent image quality of an X-ray image regardless of dose, so that industrial applicability is remarkable. do.
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Physics & Mathematics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Medical Informatics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Radiology & Medical Imaging (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Nuclear Medicine, Radiotherapy & Molecular Imaging (AREA)
- General Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Public Health (AREA)
- Heart & Thoracic Surgery (AREA)
- Molecular Biology (AREA)
- Surgery (AREA)
- Animal Behavior & Ethology (AREA)
- Optics & Photonics (AREA)
- Pathology (AREA)
- Veterinary Medicine (AREA)
- Biophysics (AREA)
- High Energy & Nuclear Physics (AREA)
- Image Processing (AREA)
- Apparatus For Radiation Diagnosis (AREA)
- Physiology (AREA)
- Quality & Reliability (AREA)
Abstract
Description
Claims (15)
- 엑스레이 영상 처리 장치에 있어서,수신된 원본 엑스레이 영상을 분석하는 영상 분석부;상기 분석 결과에 기초하여 원본 엑스레이 영상을 멀티-스케일로 변환하여 주파수 단위로 분리하는 변환부;특정 주파수 부분을 인핸스먼트하고 컴바인하여 노이즈 예측 맵을 생성하는 노이즈 예측부;상기 생성된 노이즈 예측 맵에 기초하여 에지 맵을 생성하는 에지 맵 처리부;상기 생성된 에지 맵에 기초하여 상기 엑스레이 영상의 콘트라스트를 인핸스먼트하는 콘트라스트 처리부;상기 콘트라스트가 인핸스먼트된 엑스레이 영상에 대해 평탄화 과정 및 에지 보정 과정을 수행하고 역변환하는 역변환부; 및상기 역변환된 엑스레이 영상을 톤 맵핑하고 출력을 제어하는 제어부 포함하는,엑스레이 영상 처리 장치.
- 제1항에 있어서,상기 영상 분석부는,상기 수신되는 엑스레이 영상을아나토미 영역, 다이렉트 익스포저 영역 및 콜리메이션 영역으로 세그멘테이션 하고,상기 세그멘테이션된 아나토미 영역에 대해 노이즈 영역과 에지 영역을 분리하고, 분리된 노이즈 영역은 억제하고 에지 영역은 부스트하는,엑스레이 영상 처리 장치.
- 제2항에 있어서,상기 변환부는,가우시안-라플라시안 피라미드 구조를 이용하고,엑스레이 영상 처리 장치.
- 제3항에 있어서,상기 노이즈 예측부는,고주파수 부분에 해당하는 복수의 레이어들에 대한 정보에 기초하여 각 레이어 단위로 노이즈 예측 정보를 생성하되,상기 복수의 레이어들에 대한 정보는 상기 원본 영상와 복수의 특정 주파수 부분의 밝기에 해당하는 가우시안 정보를 이용하여 산출되는 라플라시안값에 기초하고,상기 복수의 레이어 중 최상위 레이어에 대해 예측된 노이즈 예측 정보를 다음 상위 레이어에 대한 노이즈 예측 정보에 반영하여 보정하는,엑스레이 영상 처리 장치.
- 제4항에 있어서,상기 에지 맵 처리부는,상기 고주파수 부분에 해당하는 각 레이어 단위로 에지 맵을 생성하되,콘트라스트 맵을 생성하고,상기 고주파수 부분에 해당하는 각 레이어를 코릴레이션하고,상기 코릴레이션을 통해 각 레이어의 에지 맵에 대한 억제 여부 또는 억제 강도를 결정하여 적용하되,상기 아나토미 영역 정보에 기초하여, 상기 생성된 콘트라스트 맵과 에지 맵을 합친 값을 해당 레이어에 대한 상기 노이즈 예측 정보로 나누어 상기 각 레이어 단위로 에지 맵에 대한 억제 여부 또는 억제 강도를 결정하는,엑스레이 영상 처리 장치.
- 제5항에 있어서,미리 정의된 부위별, 레이어별 라플라시안 변화율의 타겟 값에 기초하여, 상기 고주파수 부분에 해당하는 각 레이어에 대해 노멀라이제이션을 수행하는 레인지 제어부를 더 포함하는,엑스레이 영상 처리 장치.
- 제1항에 있어서,상기 역변환부는,중간 주파수 부분에 해당하는 레이어에 대해 히스토그램 기반 평탄화한 값, 상기 아나토미 영역 정보와 상기 각 레이어에 대한 에지 맵 정보에 기초하여, 각 레이어에 대해 스무딩 동작을 수행하는,엑스레이 영상 처리 장치.
- 엑스레이 영상 처리 방법에 있어서,원본 엑스레이 영상을 수신하는 단계;상기 수신한 원본 엑스레이 영상을 분석하는 단계;상기 분석 결과에 기초하여 원본 엑스레이 영상을 멀티-스케일로 변환하여 주파수 단위로 분리하는 단계;특정 주파수 부분을 인핸스먼트하고 컴바인하여 노이즈 예측 맵을 생성하는 단계;상기 생성된 노이즈 예측 맵에 기초하여 에지 맵을 생성하는 단계;상기 생성된 에지 맵에 기초하여 상기 엑스레이 영상의 콘트라스트를 인핸스먼트하는 단계;상기 콘트라스트가 인핸스먼트된 엑스레이 영상에 대해 평탄화 과정 및 에지 보정 과정을 수행하고 역변환하는 단계; 및상기 역변환된 엑스레이 영상을 톤 맵핑하고 출력하는 단계를 포함하는,엑스레이 영상 처리 방법.
- 제8항에 있어서,상기 엑스레이 영상을 분석하는 단계는,아나토미 영역, 다이렉트 익스포저 영역 및 콜리메이션 영역으로 세그멘테이션 하는 단계; 및상기 세그멘테이션된 아나토미 영역에 대해 노이즈 영역과 에지 영역을 분리하고, 분리된 노이즈 영역은 억제하고 에지 영역은 부스트하는 단계를 포함하는,엑스레이 영상 처리 방법.
- 제9항에 있어서,상기 원본 엑스레이 영상을 멀티-스케일로의 변환은,가우시안-라플라시안 피라미드 구조를 이용하는,엑스레이 영상 처리 방법.
- 제10항에 있어서,상기 노이즈 예측 맵 생성 단계는,고주파수 부분에 해당하는 복수의 레이어들에 대한 정보에 기초하여 각 레이어 단위로 이루어지되,상기 복수의 레이어들에 대한 정보는 상기 원본 영상와 복수의 특정 주파수 부분의 밝기에 해당하는 가우시안 정보를 이용하여 산출되는 라플라시안값에 기초하는,엑스레이 영상 처리 방법.
- 제11항에 있어서,상기 노이즈 예측 맵 생성 단계는,상기 복수의 레이어 중 최상위 레이어에 대해 예측된 노이즈 예측 정보를 다음 상위 레이어에 대한 노이즈 예측 정보에 반영하여 보정하는,엑스레이 영상 처리 방법.
- 제12항에 있어서,상기 에지 맵 생성 단계는,상기 고주파수 부분에 해당하는 각 레이어 단위로 이루어지되,콘트라스트 맵을 생성하는 단계;상기 고주파수 부분에 해당하는 각 레이어를 코릴레이션하여 각 레이어의 에지 맵에 대한 억제 여부 또는 억제 강도를 결정하여 적용하는 단계를 포함하되,상기 고주파수 부분에 해당하는 각 레이어를 코릴레이션하여 각 레이어의 에지 맵에 대한 억제 여부 또는 억제 강도를 결정하여 적용하는 단계는, 상기 아나토미 영역 정보에 기초하여, 상기 생성된 콘트라스트 맵과 에지 맵을 합친 값을 해당 레이어에 대한 상기 노이즈 예측 정보로 나누는 단계를 포함하는,엑스레이 영상 처리 방법.
- 제13항에 있어서,미리 정의된 부위별, 레이어별 라플라시안 변화율의 타겟 값에 기초하여, 상기 고주파수 부분에 해당하는 각 레이어에 대해 노멀라이제이션을 수행하는 레인지 컨트롤 단계를 더 포함하는,엑스레이 영상 처리 방법.
- 제8항에 있어서,상기 역변환 단계는,중간 주파수 부분에 해당하는 레이어에 대해 히스토그램 기반 평탄화한 값, 상기 아나토미 영역 정보와 상기 각 레이어에 대한 에지 맵 정보에 기초하여, 각 레이어에 대해 스무딩 동작을 수행하는,엑스레이 영상 처리 방법.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR1020247012512A KR20240060662A (ko) | 2021-09-14 | 2022-07-20 | 영상 처리 방법 및 장치 |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
KR10-2021-0122696 | 2021-09-14 | ||
KR20210122696 | 2021-09-14 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2023043036A1 true WO2023043036A1 (ko) | 2023-03-23 |
Family
ID=85603069
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/KR2022/010640 WO2023043036A1 (ko) | 2021-09-14 | 2022-07-20 | 영상 처리 방법 및 장치 |
Country Status (2)
Country | Link |
---|---|
KR (1) | KR20240060662A (ko) |
WO (1) | WO2023043036A1 (ko) |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070058883A1 (en) * | 2005-09-15 | 2007-03-15 | Zhanfeng Xing | Image processing method and x-ray ct system |
JP2013119021A (ja) * | 2011-12-09 | 2013-06-17 | Hitachi Medical Corp | X線ct装置及び画像処理方法 |
JP2013240696A (ja) * | 2010-06-08 | 2013-12-05 | Zakrytoe Akcionernoe Obshchestvo Impul's | デジタル画像の補正方法 |
US20170337686A1 (en) * | 2016-05-19 | 2017-11-23 | Sichuan University | Kind of x-ray chest image rib suppression method based on poisson model |
KR102165610B1 (ko) * | 2014-04-23 | 2020-10-14 | 삼성전자주식회사 | 엑스선 영상 장치 및 엑스선 영상 장치의 영상 처리 방법 |
-
2022
- 2022-07-20 KR KR1020247012512A patent/KR20240060662A/ko unknown
- 2022-07-20 WO PCT/KR2022/010640 patent/WO2023043036A1/ko active Application Filing
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20070058883A1 (en) * | 2005-09-15 | 2007-03-15 | Zhanfeng Xing | Image processing method and x-ray ct system |
JP2013240696A (ja) * | 2010-06-08 | 2013-12-05 | Zakrytoe Akcionernoe Obshchestvo Impul's | デジタル画像の補正方法 |
JP2013119021A (ja) * | 2011-12-09 | 2013-06-17 | Hitachi Medical Corp | X線ct装置及び画像処理方法 |
KR102165610B1 (ko) * | 2014-04-23 | 2020-10-14 | 삼성전자주식회사 | 엑스선 영상 장치 및 엑스선 영상 장치의 영상 처리 방법 |
US20170337686A1 (en) * | 2016-05-19 | 2017-11-23 | Sichuan University | Kind of x-ray chest image rib suppression method based on poisson model |
Also Published As
Publication number | Publication date |
---|---|
KR20240060662A (ko) | 2024-05-08 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US9251573B2 (en) | Device, method, and storage medium for high dynamic range imaging of a combined image having a moving object | |
US8675091B2 (en) | Image data processing with multiple cameras | |
CN108476326B (zh) | 视频系统及方法、介质、摄像机系统和视频转换器 | |
EP3479346A1 (en) | Method and electronic device for producing composite image | |
WO2015137759A1 (ko) | 디지털 엑스레이 영상 시스템, 엑스레이 조사 조절 장치 및 그 방법 | |
KR20100053287A (ko) | 다수 영상의 생성 및 합성을 통한 영상 보정 장치 및 그 방법 | |
WO2020171305A1 (en) | Apparatus and method for capturing and blending multiple images for high-quality flash photography using mobile electronic device | |
US20200374444A1 (en) | Evaluation device, evaluation method, and camera system | |
WO2017034075A1 (ko) | 방사선 영상의 처리방법 및 방사선 촬영시스템 | |
US20150359413A1 (en) | Color reduction in images of an interior of a human body | |
US10067327B2 (en) | Image acquisition apparatus | |
WO2023043036A1 (ko) | 영상 처리 방법 및 장치 | |
US20220277490A1 (en) | Image processing method, image processing apparatus and program | |
WO2017052215A1 (en) | Method and apparatus for enhancing images via white pop-out | |
JP2018201156A (ja) | 画像処理装置および画像処理方法 | |
WO2019135625A1 (ko) | 영상 디스플레이 장치 및 그 제어 방법 | |
CN112566539A (zh) | 医学控制装置和医学观察装置 | |
US11672495B2 (en) | Radiographic imaging system, computer readable storage medium, optical imaging condition setting method, and optical imaging apparatus | |
CN114005026A (zh) | 机器人的图像识别方法、装置、电子设备及存储介质 | |
WO2023058858A1 (ko) | 영상 처리 방법 및 장치 | |
CN112243118B (zh) | 白平衡校正方法、装置、设备及存储介质 | |
JP2005284534A (ja) | 高ダイナミックレンジ画像を低ダイナミックレンジ画像に変換する方法及び関連装置 | |
CN111988577A (zh) | 一种基于图像增强的视频监控方法 | |
US20240056691A1 (en) | Electronic device and control method | |
WO2018131733A1 (ko) | Ct 이미지의 잡음 저감 방법 및 장치 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 22870123 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 20247012512 Country of ref document: KR Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 2022870123 Country of ref document: EP |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
ENP | Entry into the national phase |
Ref document number: 2022870123 Country of ref document: EP Effective date: 20240415 |