JP4530834B2 - Ultrasonic image processing method, ultrasonic image processing apparatus, and ultrasonic image processing program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing method and the like scarcely generating artifact when processing to reduce speckles on an ultrasonic image. <P>SOLUTION: This ultrasonic image processing method includes a step (a) of multiplying values of a plurality of pixels in the mask area determined based on structural elements in a morphology processing on a plurality of signals expressing the values of the plurality of pixels constituting an ultrasonic image respectively by a first group weighting factor set according to positions in a mask area; and a step (b) of multiplying the values of the plurality of pigments in the mask area determined based on the structural elements in the morphology processing on the plurality of signals processed in the step (a) by a second group weighting factor set according to the positions in the mask area, and converting the pixel value in the mask center to the minimum value in the plurality of pixel values multiplied by the second group weighting factor, while moving the mask area. <P>COPYRIGHT: (C)2005,JPO&amp;NCIPI

Description

  The present invention relates to an ultrasonic image processing method for processing an ultrasonic image signal obtained by transmitting ultrasonic waves and receiving ultrasonic echoes, an ultrasonic image processing apparatus using the same, and such The present invention relates to an ultrasonic image processing program for causing a CPU to execute an ultrasonic image process.

  In the medical field, various imaging techniques have been developed in order to observe and diagnose the inside of a subject. In particular, ultrasonic imaging that acquires internal information of a subject by transmitting and receiving ultrasonic waves enables real-time image observation, and other medical applications such as X-ray photographs and RI (radio isotope) scintillation cameras. Unlike imaging technology, there is no radiation exposure. Therefore, ultrasonic imaging is used as a highly safe imaging technique in a wide range of areas including gynecological system, circulatory system, digestive system, etc. in addition to fetal diagnosis in the obstetrics field.

  Ultrasonic imaging is an imaging technique that generates an image based on the following principle. Ultrasonic waves are reflected at boundaries between regions having different acoustic impedances, such as boundaries between structures. Therefore, by transmitting an ultrasonic beam into a subject such as a human body, receiving an ultrasonic echo generated in the subject, and determining the reflection point and reflection intensity at which the ultrasonic echo was generated, It is possible to extract the contour of the structure (eg, internal organs or lesion tissue).

  By the way, when a structurally non-uniform subject such as a living body is imaged using such a principle, a pattern in which bright portions and / or dark portions are scattered appears in the generated ultrasonic image. Such a pattern is called a speckle pattern, and is generated, for example, by interference of ultrasonic echoes reflected by a non-uniform tissue existing inside an internal organ or the like. Since the speckle pattern acts as a kind of noise, the contour of the drawn structure is often unclear. Therefore, in order to generate an image suitable for medical diagnosis from such an original image, an image including sharpness enhancement processing (contour enhancement processing) and grain suppression processing (smoothing processing) is performed on the acquired original image data. It is necessary to perform processing. As such processing, specifically, averaging processing, median filter processing, hysteresis smoothing processing, morphology processing (also referred to as “morphology” or “morphology”) processing, and the like are known.

Morphological processing is image processing that uses an element related to image movement called a structural element and an operation called Minkowski sum and Minkowski difference. The Minkowski sum and Minkowski difference of the functions f and g are defined by equations (1) and (2), respectively. In the following formulas (1) to (4), F and G are defined regions of f and g, respectively.

The morphological process includes four basic processes called dilation, erosion, opening, and closing. The basic processing for the image function f by the structural element g ^S is defined by the following equations (3) to (6). In Expressions (3) to (6), the function g is symmetric about the origin.

As shown in the equation (3), dilation is a process for obtaining the Minkowski sum of the image function f moved by the structural element g ^S , and intuitively, in the mask region defined based on the structural element. The original value is expanded by searching for the maximum value and replacing the pixel value at the center of the mask with the maximum value. Further, as shown in Expression (4), erosion is a process for obtaining the Minkowski difference of the image function f moved by the structural element g ^S. Intuitively, the minimum value is searched in the mask area. The original image is contracted by replacing the pixel value at the center of the mask with the minimum value. Furthermore, the opening is a process of performing dilation after erosion and has, for example, a function of removing convex portions. Closing is a process of performing erosion after dilation, and has a function of filling a concave portion, for example.

It has been studied to extract a structure from an ultrasonic image and improve image quality by applying such morphology processing to ultrasonic image processing. For example, Patent Document 1 discloses that a contour line is extracted by performing morphological processing as preprocessing in an apparatus that encodes and decodes time-series data in order to perform diagnosis efficiently. Non-Patent Document 1 discloses that boundary enhancement and speckle reduction are simultaneously performed by controlling structural elements in morphology processing.
Japanese Patent Laid-Open No. 10-84286 (5th page, FIG. 18) Masayoshi Sakurai, Masayasu Ito, “Control of Variable Structural Elements in Adaptive Morphology for Boundary Enhancement of Ultrasound Images”, IEICE Transactions, June 2003, J86-D-II, Vol. 6 No., p. 895-907

  However, such a morphological process has a function of smoothing a smaller pattern on the basis of the mask size, so that an artifact (virtual image) due to the mask size, that is, a structural element is generated. There is.

  Therefore, in view of the above points, an object of the present invention is to provide an image processing method and the like that hardly cause artifacts when performing processing for reducing speckles on an ultrasonic image obtained by transmitting and receiving ultrasonic waves. And

  In order to solve the above-described problems, an ultrasonic image processing method according to the present invention is generated based on a signal obtained by scanning an inside of a subject using an ultrasonic beam, and forms a plurality of ultrasonic images. A method of processing a plurality of signals each representing a pixel value of a pixel of the pixel, wherein the pixel value of the plurality of pixels in the mask region determined based on a structural element in the morphology processing is processed for the plurality of signals. A process of converting the pixel value at the center of the mask to a maximum value among a plurality of pixel values multiplied by the weighting coefficient of the first group, respectively by multiplying the weighting coefficient of the first group set according to the position within Step (a) applied while moving the mask region, and a mask determined based on the structural elements in the morphology processing for the plurality of signals processed in step (a) The pixel values of a plurality of pixels in the area are respectively multiplied by a second group of weighting factors set in accordance with the position in the mask region, and the pixel values at the mask center are multiplied by a plurality of weighting factors of the second group. And (b) performing a process of converting the pixel value to the minimum value while moving the mask area.

  In addition, the ultrasonic image processing apparatus according to the present invention generates pixel values of a plurality of pixels that are generated based on a signal obtained by scanning the inside of a subject using an ultrasonic beam and constitute an ultrasonic image. An apparatus for processing a plurality of signals respectively representing a storage means for storing the plurality of signals, and a plurality of signals stored by the storage means in a mask region determined based on a structural element in morphological processing A plurality of pixels obtained by multiplying pixel values of a plurality of pixels by a first group of weighting coefficients set according to positions in the mask area, and multiplying a pixel value at the mask center by a weighting coefficient of the first group The first signal processing for converting to the maximum value among the values is performed while moving the mask area, and the signal subjected to the first signal processing is based on the structural elements in the morphology processing. The pixel values of a plurality of pixels in the determined mask area are respectively multiplied by a second group of weighting coefficients set according to the position in the mask area, and the mask center pixel value is multiplied by the second group of weighting coefficients. And image signal processing means for performing second signal processing for converting to a minimum value among the multiplied pixel values while moving the mask area.

  Furthermore, the ultrasonic image processing program according to the present invention generates pixel values of a plurality of pixels that are generated based on a signal obtained by scanning the inside of a subject using an ultrasonic beam and that constitute an ultrasonic image. A program for processing a plurality of signals each representing a pixel value of a plurality of pixels in a mask region determined based on a structural element in morphology processing for the plurality of signals according to a position in the mask region A process of converting the pixel value at the center of the mask to the maximum value among a plurality of pixel values multiplied by the weighting coefficient of the first group is performed by moving the mask area. The procedure (a) to be performed while the signals processed in the procedure (a) are determined for a plurality of pixels in the mask area determined based on the structural elements in the morphology processing. The prime value is multiplied by a second group of weighting coefficients set in accordance with the position in the mask area, the mask center pixel value is multiplied by the second group of weighting coefficients, and the minimum of the plurality of pixel values is multiplied. The CPU executes the procedure (b) of performing the process of converting to a value while moving the mask area.

  According to the present invention, since the morphology processing is performed on the acquired ultrasonic image using the weighting coefficient set according to the position in the mask region, speckle is reduced while suppressing artifacts due to the structural elements. can do. Therefore, it is possible to obtain an ultrasonic image with good image quality with reduced noise.

Hereinafter, the best mode for carrying out the present invention will be described in detail with reference to the drawings. The same constituent elements are denoted by the same reference numerals, and the description thereof is omitted.
FIG. 1 is a flowchart showing an ultrasonic image processing method according to an embodiment of the present invention. The ultrasonic image processing method according to the present embodiment reduces the influence of speckle in an ultrasonic image by using morphology processing.

First, a general speckle reduction process using a morphology process will be described with reference to FIG. 2A to 2C schematically show an ultrasonic image constituted by a plurality of pixels 10. FIG. 2A shows a high luminance region 11 a divided by the low luminance region 12. In order to connect the divided high-intensity regions 11a, dilation that is a basic operation of the morphology processing shown in the following equation (7) is performed on the ultrasonic image. That is, the process of extracting the maximum value from the pixels in the mask area 13 centered on the target pixel 14 and converting the pixel value of the target pixel 14 to the maximum value is performed for each of the plurality of pixels 10.
Here, g is a set of pixel values representing an image obtained by dilation, Y is a pixel value of the original image, G is a mask determined based on a structural element, and (x, y) Are coordinates in the mask area. As a result, as shown in FIG. 2B, the high luminance area expands, and the low luminance area that has divided it is deleted.

Next, the erosion shown in the following equation (8) is performed on the ultrasonic image shown in FIG. That is, a process of extracting the minimum value from the pixels in the mask area 13 centered on the target pixel 14 and converting the pixel value of the target pixel 14 to the minimum value is performed by a plurality of processes shown in FIG. This is performed for each of the pixels 10.
Here, f is a set of pixel values representing an image obtained by erosion. As a result, as shown in FIG. 2C, the high brightness region 11b contracts and returns to its original size. In this way, the high luminance region 11a divided by the influence of speckle can be connected like the high luminance region 11c.

  Next, a case where such a general morphology process is applied to the ultrasonic image (original image) shown in FIG. First, dilation is performed on the original image shown in FIG. 3A with a mask size of 7 × 7 (mask region 15). As a result, as shown in FIG. 3B, the high brightness area expands and the relatively low brightness areas gather at the center.

Further, erosion is applied to the image shown in FIG. At this time, when a part of the mask area 15 covers the central low-luminance area 16, that is, when the mask is located in the areas 15a to 15d or inside thereof, the pixel value of the target pixel is the low-luminance area. It is converted to 16 pixel values. Therefore, the central pixel of the areas 15a to 15d or all the pixels existing inside thereof are converted into low-brightness pixel values, and a rectangular low-brightness area is displayed at the center of the image as shown in FIG. 17 will occur.
The low luminance region 17 is an artifact that appears due to the mask size or mask shape. The presence of such artifacts is undesirable when using ultrasound images for medical diagnosis. Therefore, in the present embodiment, the morphology process is performed using a weighting coefficient set according to the distance from the target pixel.

Hereinafter, the morphology processing in the present embodiment will be described with reference to FIG. 1, FIG. 4, and FIG. FIG. 4 is a diagram for explaining the morphology processing in the present embodiment, and FIG. 5 is a diagram showing weighting coefficients used in the present embodiment.
In step S1 of FIG. 1, an ultrasonic detection signal is acquired by transmitting an ultrasonic wave from the ultrasonic probe toward the subject and receiving an ultrasonic echo reflected by the subject. Ultrasonic image data is acquired by performing signal processing such as amplification, A / D conversion, and detection on the ultrasonic detection signal. FIG. 4A is an ultrasonic image (original image) represented by the ultrasonic image data acquired in this way.

Next, in steps S2 and S3, dilation using a weighting coefficient is performed on the original image shown in FIG.
That is, in step S2, for example, pixels in a mask area having a mask size of 7 × 7 with a certain target pixel (x, y) = (0, 0) constituting the original image shown in FIG. The pixel value (x, y) is multiplied by a weighting coefficient (hereinafter also referred to as “dilation coefficient”) dk (x, y) as shown in FIG. Here, (x, y) is a coordinate in the mask area. As shown in FIG. 5A, the dilation coefficient dk (x, y) has a value of 1 at the target pixel (dk (0, 0) = 1), and the value decreases as the distance from the target pixel increases. Is set to

Further, in step S3, as shown in the following equation (9), the maximum value is extracted from the pixel values multiplied by the weighting coefficient dk (x, y), and the pixel value of the target pixel is converted to the maximum value. To do.
Here, g ′ is a set of pixel values representing an image obtained by dilation, Y is a pixel value of the original image, and G is a mask determined based on a structural element. By performing such processing for each of the plurality of pixels constituting the ultrasonic image, an image in which the high luminance region is expanded is obtained as shown in FIG.

Next, in steps S4 and S5, erosion using a weighting coefficient is performed on the ultrasonic image subjected to dilation.
That is, in step S4, the pixel value of the pixel (x, y) in the mask area of the mask size 7 × 7 with the target pixel (x, y) = (0, 0) as the center is changed to (b) in FIG. And ek (x, y) as shown in FIG. As shown in FIG. 5B, the erosion coefficient is set such that the value at the pixel of interest is 1 (ek (0,0) = 1), and the value increases as the distance from the pixel of interest increases.

Further, in step S5, as shown in the following equation (10), the minimum value is extracted from the pixel values multiplied by the weighting coefficient ek (x, y), and the pixel value of the target pixel is converted into the minimum value. To do.
Here, f ′ is a set of pixel values representing an image obtained by erosion. By performing such processing for each of the plurality of pixels constituting the ultrasonic image, an image in which the high luminance region is contracted is obtained as shown in FIG.

  As apparent from a comparison between FIG. 3C and FIG. 4C, almost no artifacts due to the mask size or mask shape appear in FIG. 4C. As shown in FIGS. 5A and 5B, in the dilation that takes the maximum value in the mask area, the value of the coefficient multiplied by the pixel far from the mask center is reduced to reduce the value in the mask area. This is because in the erosion that takes the minimum value of, the influence of the mask peripheral portion on the mask center can be reduced by increasing the value of the coefficient multiplied to the pixel far from the mask center.

Next, weighting coefficients (dilation coefficients and erosion coefficients) used in the present embodiment will be described in detail.
In order to increase the effect of reducing the influence of speckle without causing artifacts due to mask size or mask shape, the dilation coefficient depends on the speckle feature size including the size and shape of the speckle pattern. It is desirable to set the erosion coefficient.

  Here, the resolution of the ultrasonic image increases as the frequency of the transmitted ultrasonic wave increases, and decreases as the frequency decreases. In addition, since the ultrasonic wave is more easily attenuated as the frequency becomes higher, a component having a relatively low frequency remains in the ultrasonic echo reflected from the deep part of the subject. Therefore, in the depth direction (distance direction) of the subject, the resolution of the ultrasonic image decreases as the depth increases. For these reasons, the size of the speckle pattern appearing in the ultrasonic image tends to decrease as the frequency of the ultrasonic wave transmitted from the ultrasonic probe increases and increase as the frequency of the ultrasonic wave decreases. In addition, the speckle pattern tends to be small in the ultrasonic image representing the shallow part of the subject, and the speckle pattern tends to be large in the ultrasonic image representing the deep part of the subject.

  On the other hand, the shape of the speckle pattern is applied to the depth of the subject represented in the ultrasound image (hereinafter also referred to as “depth of the ultrasound image”), the ultrasound scanning method, and the image data. Influenced by image processing. For example, in the original image data before coordinate conversion corresponding to the scanning method, the speckle pattern has an elliptical shape that is long in the distance direction and short in the azimuth direction. Further, when interpolation processing is performed on original image data obtained by linear scanning, the speckle pattern becomes an elliptical shape that is short in the distance direction and long in the azimuth direction.

  Therefore, the mask size, mask shape, and coefficient value of the dilation coefficient and erosion coefficient are set by using the frequency of the transmission ultrasonic wave and the depth of the ultrasonic image as parameters. Alternatively, the type or part of the organ to be observed, such as the heart or liver, may be used as the parameter. The frequency of transmission ultrasonic waves used when imaging a certain organ, the depth range in which the organ exists, the contents of the scanning method, coordinate conversion, and the like are generally determined. Therefore, the frequency of the transmission ultrasonic wave, the depth of the ultrasonic image, the scanning method, and the like may be associated with the observation region in advance, and the dilation coefficient and the erosion coefficient may be set for each observation region. Furthermore, parameters such as the frequency of the transmitted ultrasound and the depth of the ultrasound image, the frequency of the observed region and the transmitted ultrasound, the depth of the observed region and the ultrasound image, or the frequency of the transmitted ultrasound and the depth of the ultrasound image and the observed region A combination of these may be used.

  Specific examples of the mask size and mask shape of the weighting coefficient set based on the above parameters will be given. For example, consider a case in which a linear image of the neck is generated using an ultrasonic probe that transmits ultrasonic waves having a center frequency of 12 MHz. In this case, since the frequency of the transmission ultrasonic wave is relatively high and the depth of the ultrasonic image is small, the speckle pattern is relatively small, but the speckle pattern is elongated in the azimuth direction by the interpolation process. Therefore, for example, an anisotropic mask of azimuth direction × distance direction = 9 × 5 is used. Also, consider a case where an abdominal ultrasound image is generated using an ultrasound probe that transmits ultrasound with a center frequency of 3.5 MHz. In this case, the speckle pattern is relatively large because the frequency of the transmitted ultrasound is relatively low and the depth of the ultrasound image is large. Therefore, the mask size is increased to, for example, about azimuth direction × distance direction = 9 × 13.

  Next, the value of the weighting coefficient will be described. In dilation and erosion, the smaller the difference between the weighting coefficients at the mask center and the mask peripheral edge, the greater the influence of the mask peripheral edge on the mask center, and the higher the speckle reduction effect. On the other hand, the speckle reduction effect becomes more conspicuous as the difference between the weighting factors increases. Therefore, when the frequency of the transmission ultrasonic wave is low, or when imaging the deep part of the subject (that is, when the size of the speckle pattern is large), the speckle reduction effect is reduced by reducing the difference between the weighting coefficients. Can be raised. Specifically, the value of the dilation coefficient is increased in a range smaller than 1, and the value of the erosion coefficient is decreased in a range larger than 1. On the contrary, when the frequency of the transmission ultrasonic wave is high or when a shallow part of the subject is imaged (that is, when the size of the speckle pattern is small), the difference between the weighting coefficients may be increased. Specifically, the value of the dilation coefficient is decreased and the value of the erosion coefficient is increased. In this case, the effect of suppressing artifacts becomes higher.

  Further, the dilation coefficient and the erosion coefficient may be arbitrarily set by the user. The user can obtain desired speckle reduction and artifact suppression effects by setting these coefficients while observing the ultrasonic image displayed on the screen.

  Such a dilation coefficient and an erosion coefficient have a relationship of 1 when multiplied by coefficients at corresponding positions in the mask region, that is, a relationship of dk (x, y) × ek (x, y) = 1. It is desirable to be. Thereby, the brightness of the whole ultrasound image, that is, the fluctuation of the average level of the pixel values of all the pixels constituting the ultrasound image can be suppressed before and after the morphology processing.

  As described above, according to the present embodiment, speckle can be reduced while suppressing artifacts due to the mask size and mask shape when performing the smoothing process on the ultrasonic image. Therefore, an ultrasonic image with good image quality can be obtained.

  The ultrasonic image processing method according to the present embodiment smoothes an ultrasonic image by combining known linear or non-linear luminance conversion, enhancement processing such as an unsharp mask, frequency processing, etc., and contour enhancement. It is also possible to obtain an effect. For example, it is possible to improve the accuracy of the edge-enhanced image by performing frequency enhancement processing after performing image processing according to the present embodiment on the original image data.

  FIG. 6 is a block diagram showing the ultrasonic imaging apparatus according to the first embodiment of the present invention. In this ultrasonic imaging apparatus, the ultrasonic image processing method according to an embodiment of the present invention is used. As shown in FIG. 6, the ultrasonic imaging apparatus according to the present embodiment controls the ultrasonic probe 30 that transmits and receives ultrasonic waves, and the transmission and reception of ultrasonic waves, and is based on the acquired ultrasonic detection signals. And an ultrasonic imaging apparatus main body for generating an ultrasonic image.

  The ultrasonic probe 30 includes an ultrasonic transducer array in which a plurality of ultrasonic transducers are arranged. Each ultrasonic transducer is, for example, a piezoelectric ceramic represented by PZT (Pb (lead) zirconate titanate) or a polymer piezoelectric element represented by PVDF (polyvinylidene difluoride). It is manufactured by forming electrodes on both ends of a piezoelectric material (piezoelectric element). When a voltage is applied by sending a pulsed electric signal or a continuous wave electric signal to the electrodes of such an ultrasonic transducer, the piezoelectric element expands and contracts to generate ultrasonic waves. Therefore, a plurality of ultrasonic transducers are electronically controlled to generate pulsed or continuous ultrasonic waves from the respective ultrasonic transducers. Thereby, an ultrasonic beam is formed by synthesizing those ultrasonic waves, and the subject is electronically scanned. The plurality of ultrasonic transducers expand and contract by receiving propagating ultrasonic waves and generate electrical signals. These electric signals are output as ultrasonic detection signals.

Such an ultrasonic probe 30 is connected to the ultrasonic imaging apparatus main body via a cable.
As the ultrasonic probe 30, a linear array probe in which a plurality of ultrasonic transducers are arranged one-dimensionally, a convex array probe in which a plurality of ultrasonic transducers are arranged on a convex surface, or the like is used. A two-dimensional array probe in which a plurality of ultrasonic transducers are two-dimensionally arranged may be used. In this case, ultrasonic images relating to a plurality of different cross sections can be obtained without mechanically moving the ultrasonic probe.

  Alternatively, an intracavity probe that is inserted into a subject and performs ultrasonic imaging may be used as the ultrasonic probe 30. As the body cavity probe, there are known an ultrasonic probe that is inserted into a treatment instrument insertion hole of an endoscope and an ultrasonic endoscope integrated with an endoscope as shown in FIG. ing.

  The ultrasonic endoscope shown in FIG. 7 includes an insertion portion 31, an operation portion 32, a connection cord 33, and a universal cord 34. The insertion portion 31 has a flexible and long tubular shape so that it can be inserted into the patient's body. The operation unit 32 provided at the proximal end of the insertion unit 31 is connected to the ultrasonic observation device via the connection cord 33 and is connected to the light source device via the universal cord 34. The operation unit 32 is provided with a treatment instrument insertion port 35 through which various treatment instruments are inserted toward the distal end of the insertion section 31.

  The insertion portion 31 of the ultrasonic endoscope is provided with an illumination window and an observation window. The illumination window is equipped with an illumination lens for emitting illumination light supplied from the light source device via the light guide. These constitute an illumination optical system. In addition, an objective lens is attached to the observation window, and an image guide input end or a solid-state imaging device such as a CCD camera is disposed at the imaging position of the objective lens. These constitute an observation optical system. Furthermore, an ultrasonic transducer that transmits ultrasonic waves into the subject and receives ultrasonic echoes generated in the subject is disposed at the distal end of the insertion portion 31.

  In such a body cavity probe, ultrasonic imaging is performed by a radial scanning method. In the radial scanning method, an ultrasonic wave is transmitted and received while rotating the probe, and an ultrasonic signal is imaged in synchronization with the rotation, and a plurality of transducers arranged in a circle are electrically connected. There is an electronic radial scanning method in which scanning is performed by controlling automatically. According to such a scanning method, 360 ° around the probe can be displayed at a time. Alternatively, an intracavity probe using a scanning method other than radial scanning is known in which a convex array is arranged at the tip. When using a convex array, a wide viewing angle can be obtained.

  Referring to FIG. 6 again, the main body of the ultrasonic imaging apparatus includes a control unit 40, a recording unit 41, a drive signal generation unit 42, a transmission / reception switching unit 43, a signal processing unit 51, and an A / D converter 52. A digital processing unit 50, an image memory 58, a D / A converter 59, a display unit 60, and an input unit 61. The control unit 40 includes a CPU and software, and controls each unit of the ultrasonic imaging apparatus.

  The recording unit 41 records a basic program for causing a CPU included in the ultrasonic imaging apparatus to execute an operation, a program (software) used for various processes, information used for the processes, and the like. Control the recording medium. For example, in the weighting coefficient recording unit 41a, a plurality of types of dilation coefficients and erosion coefficients used for speckle reduction processing are used as parameters such as transmission ultrasonic frequency, ultrasonic image depth, observation region, and the like. Correspondingly recorded. In addition to the built-in hard disk, an external hard disk, flexible disk, MO, MT, RAM, CD-ROM, or DVD-ROM may be used as the recording medium.

  The drive signal generation unit 42 includes a plurality of pulsers respectively corresponding to the plurality of ultrasonic transducers included in the ultrasonic probe 30. Each pulser generates a drive signal at a predetermined timing under the control of the control unit 40. Thereby, ultrasonic waves are respectively generated from the plural ultrasonic transducers with a predetermined time difference.

  The transmission / reception switching unit 43 inputs the drive signal generated by the drive signal generation unit 42 to the ultrasonic probe 30 and captures a detection signal in a signal processing unit 51 described later according to the control of the control unit 40. Switch at the timing. By limiting the reading time zone of the detection signal in this way, an ultrasonic echo signal reflected from a specific depth of the subject is detected.

The signal processing unit 51 includes a plurality of channels respectively corresponding to the plurality of ultrasonic transducers. Each of these channels captures a detection signal output from the corresponding ultrasonic transducer at a predetermined timing, and performs signal processing such as amplification and Nyquist filter processing.
The A / D converter 52 generates detection data by digitally converting the analog signal processed in the signal processing unit 51.

  The digital processing unit 50 includes a memory 53, a reception focus processing unit 54, a digital scan converter (DSC) 55, a speckle reduction processing unit 56, and an image processing unit 57. The memory 53 includes line memories respectively corresponding to the plurality of channels of the signal processing unit 51, and stores the generated detection data in time series for each line. Or you may comprise the memory 53 by the cine memory which memorize | stores the moving image data for a fixed time.

  The reception focus processing unit 54 performs reception focus processing by delaying a plurality of detection data stored in the memory 53 and adding them. As a result, sound ray data representing a reception beam focused in a predetermined sound ray direction is generated. Furthermore, image data is obtained by detecting the waveform represented by the sound ray data. The value of this image data represents the pixel values of a plurality of pixels constituting the ultrasonic image.

  The DSC 55 generates image data for display by converting the scanning format from image data in the scanning space of the ultrasonic beam to image data in the physical space. That is, the DSC 55 performs resampling corresponding to the image display range, and coordinate conversion and interpolation to a display form corresponding to the ultrasonic scanning method. For example, image data obtained by linear scanning is subjected to an interpolation process for generating a linear image. Also, polar coordinate conversion and interpolation processing are performed on image data obtained by sector scanning, convex scanning, or radial scanning.

The speckle reduction processing unit 56 performs a process of reducing speckles in the ultrasonic image by performing a morphology process using a weighting coefficient on the input image data. Note that the details of the speckle reduction processing are the same as those described with reference to FIG.
The image processing unit 57 performs, for input image data, STC (sensitivity time control) for correcting distance attenuation, linear gradation processing including gain adjustment and contrast adjustment, and non-linear including γ correction. Image processing such as gradation processing.

  The image memory 58 stores image data for display in a format capable of raster scanning, for example. The D / A converter 59 converts the image data read from the image memory 58 into an analog signal and outputs it.

The display unit 60 is, for example, a raster scan CRT display or LED display, and displays an ultrasonic image based on the D / A converted image signal.
The input unit 61 is an input device used when inputting various commands and information to the ultrasonic imaging apparatus main body. The input unit 61 may be configured by an adjustment knob, an input button, a keyboard, an adjustment console including a touch panel, or an external keyboard or a pointing device such as a mouse. The input unit 61 includes a transmission frequency input unit 61a used for setting a transmission frequency, an observation part (body mark) input part 61b used for inputting a part that the user wants to observe, and the like.

  In the present embodiment, the reception focus processing unit 54, the DSC 55, the speckle reduction processing unit 56, and the image processing unit 57 are configured by a CPU and software. However, these units 54 to 57 may be configured using an analog circuit or a digital circuit. For example, STC, gain adjustment, and contrast adjustment performed in the image processing unit 57 may be performed on an analog signal by providing an analog circuit in the signal processing unit.

  The digital processing unit 50 including the memory 53 to the image processing unit 57 may be configured using a personal computer (PC). In that case, data processed by the digital processing unit 50 may be directly input via the ultrasonic probe 30, the signal processing unit 51, and the A / D converter 52, or may be a network or recording. You may input via a medium.

Next, the operation of the ultrasonic imaging apparatus shown in FIG. 6 will be described.
Before starting the ultrasonic imaging, the user inputs predetermined setting items using the input unit 61. For example, the user inputs information such as the frequency of the transmission ultrasonic wave and the observation site (for example, the names of internal organs such as the heart and liver) using the transmission frequency input unit 61a and the observation site input unit 61b. Alternatively, when the user connects the ultrasonic probe 30 to the ultrasonic imaging apparatus main body, the control unit 40 may recognize the type of the ultrasonic probe 30 and the frequency of the transmission ultrasonic wave. . As a result, the control unit 40 sets predetermined items for transmitting and receiving ultrasonic waves, and based on these parameters, a morphology is selected from a plurality of weighting factors recorded in the weighting factor recording unit 41a. A weighting factor used in the process is selected.

  Next, when the user starts ultrasonic imaging, an ultrasonic beam having a set frequency is transmitted from the ultrasonic probe 30 under the control of the control unit 40, and linear scanning, sector scanning, The subject is scanned by a scanning method such as convex scanning or radial scanning. The ultrasonic beam is reflected by a reflector present in the subject, and a plurality of ultrasonic echoes are received by the ultrasonic probe 30. The received ultrasonic echo is converted into an electric signal by the ultrasonic probe 30 and input to the ultrasonic imaging apparatus main body as a detection signal.

  The plurality of detection signals input to the ultrasonic imaging apparatus main body are subjected to predetermined signal processing in the signal processing unit 51. Thereby, a detection signal from which a wide band and unnecessary frequency components are removed is obtained. These detection signals are A / D converted, temporarily stored in the memory 53, and then subjected to reception focus processing in the reception focus processing unit 54 to generate sound ray data. Image data obtained based on the sound ray data generated in this way is input to the DSC 55. In the DSC 55, scanning format conversion corresponding to the scanning method is performed on the input sound ray data.

The image data generated in this way is subjected to a morphology process using a weighting coefficient selected in advance by the control unit 40 in the speckle reduction processing unit 56. Alternatively, at that time, dilation coefficients and erosion coefficients arbitrarily input by the user may be used.
The image data subjected to the speckle reduction processing is subjected to predetermined image processing such as gradation processing in the image processing unit 57, temporarily stored in the image memory 58, and then subjected to D / A conversion and output. Thereby, an ultrasonic image is displayed on the screen of the display unit 60.

  Next, an ultrasonic imaging apparatus according to the second embodiment of the present invention will be described with reference to FIGS. FIG. 8 is a block diagram showing the ultrasonic imaging apparatus according to this embodiment. This ultrasonic imaging apparatus includes a digital processing unit 70 shown in FIG. 7 instead of the digital processing unit 50 shown in FIG. About another structure, it is the same as that of the ultrasonic imaging device shown in FIG.

  As apparent from the figure, the digital processing unit 50 shown in FIG. 6 and the digital processing unit 70 shown in FIG. 8 are different in the order of data processing after the reception focus processing. That is, in the digital processing unit 70 shown in FIG. 8, the image data obtained based on the sound ray data generated by the reception focus processing is the speckle reduction processing in the speckle reduction processing unit 56 and the image processing unit 57. After image processing such as gradation processing is performed, the scanning format is converted in the DSC 55.

As shown in FIG. 8, when the speckle reduction processing unit 56 is arranged in the front stage of the DSC 55, there are the following advantages. As shown in FIG. 9A, in the image data before the scan conversion is performed in the DSC 55, the shape of the speckle pattern is almost uniform over the entire ultrasonic image. For such image data, for example, polar coordinate conversion (scan conversion) is performed in order to obtain a sector image. Here, in a general ultrasonic imaging apparatus, polar coordinate conversion with a display angle of 60 ° or more is often performed. As a result, as shown in FIG. 9B, the ultrasonic image becomes an image that extends horizontally as the depth L increases, and extends in the circumferential direction as the azimuth angle θ increases (for example, θ = θ ₁ ). Converted. Therefore, the speckle pattern appearing in the ultrasonic image is also deformed into such a shape. That is, the shape of the speckle pattern becomes non-uniform due to the influence of image processing.

  When speckle reduction processing is performed on such an ultrasonic image, the speckle reduction effect becomes uneven depending on the position in the ultrasonic image. Moreover, in order to make the speckle reduction effect uniform, it is conceivable to change the mask shape according to the position in the ultrasonic image, but the calculation processing becomes complicated. Therefore, by performing speckle reduction processing before speckle pattern deformation occurs, that is, before scan conversion, it is possible to obtain a uniform speckle reduction effect over the entire ultrasound image by simple arithmetic processing. Can do.

  Further, in order to obtain a linear image, interpolation processing is performed on the image data in the DSC 55, which increases the amount of data to be calculated. However, it is possible to perform high-speed processing by performing speckle reduction processing at a stage where the amount of data before performing interpolation processing is small. In particular, in the case of a radial image acquired by an ultrasonic endoscope or the like, a great effect can be obtained.

An example of the mask shape of the weighting coefficient used in this embodiment will be described. In general, before scan conversion, the speckle pattern has a shape extending long in the distance direction. Therefore, in order to reduce such speckle, it is effective to use an anisotropic mask that is long in the distance direction. For example, when generating an abdominal ultrasound image, the frequency of the transmitted ultrasound is relatively low (for example, the center frequency is about 3.5 MHz) and the depth of the ultrasound image is large, so the speckle pattern is relatively growing. Therefore, in this case, for example, an anisotropic mask of azimuth direction × distance direction = 9 × 13 is used. An abdominal convex image is obtained by performing scan conversion on the image after the morphology processing using such a mask.
In the present embodiment, among the image processing in the image processing unit 57, the γ correction may be performed after the scan format conversion in the DSC 55.

  Next, an ultrasonic imaging apparatus according to the third embodiment of the present invention will be described. FIG. 10 is a block diagram showing the ultrasonic imaging apparatus according to this embodiment. This ultrasonic imaging apparatus includes a digital processing unit 80 shown in FIG. 10 instead of the digital processing unit 50 shown in FIG. About another structure, it is the same as that of the ultrasonic imaging device shown in FIG.

  The digital processing unit 80 further includes a frequency band division processing unit 81 provided in the subsequent stage of the speckle reduction processing unit 56 with respect to the digital processing unit 50 shown in FIG. The frequency band division processing unit 81 performs frequency enhancement processing on the image data that has been subjected to the speckle reduction processing by the speckle reduction processing unit 56 by dividing the spatial frequency component into a plurality of frequency bands.

The operation of the frequency band division processing unit 81 will be described in detail with reference to FIG. FIG. 11 is a diagram for explaining the frequency band division processing.
As shown in FIG. 11, when the image data DT (0) processed in the speckle reduction processing unit is input to the frequency band division processing unit 81, the image data DT (0) is thinned out in the downsampling unit 801. At the same time, the thinned data is subjected to filter processing such as Nyquist filter processing. By repeating such processing, downsampling data DT (1), DT (2),..., DT (N) having low spatial frequency components are sequentially generated.

  Next, in the upsampling unit 802, 0-value data is inserted into the nth downsampled data DT (n) (n = 1 to N), and filter processing such as smoothing filter processing is performed. As a result, upsampling data DT (n) ′ having the same size as the adjacent (n−1) th data is obtained.

Next, the subtraction unit 803 performs a subtraction process between the (n−1) th downsampling data DT (n−1) and the adjacent nth upsampling data DT (n) ′. Thereby, the subtraction data DS (0) to DS (N-1) are obtained. These subtraction data DS (0) ~DS (N- 1) , the data group including image data DT (0) frequency components of the spatial frequency components _f 0 ~f _N is divided into N frequency bands included in each It is. For example, the subtraction data DS (n) (n = 0 to N−1) includes frequency components f _{n to} f _{n + 1} .

Then, the multiplication section 804, subtraction data DS (0), DS (1 ), ..., the DS (N-1), the weighting coefficients _{k 0, k 1, ...,} k N-1 are multiplied respectively. Further, the data DS (n) ′ (n = 1 to N−1) multiplied by the weighting coefficient is upsampled by the upsampling unit 805 so that the data size is equal to that of the original image data DT (0). The

The data DS (0) and DS (1) ′, DS (2) ′,..., DS (N−1) ′ having the same data size are added in the adding unit 806. Thus, the data DT _EN weighted for each spatial frequency band is generated. Furthermore, the weighted data _{DT EN} and the original image data DT (0), in the multiplication section 807, is multiplied predetermined weighting coefficients K and (1-K) respectively, are added in the adding unit 808. In this manner, the frequency-enhanced image data DT _OUT is generated and output.

The weighting coefficients k _{0 to} k _N−1 used in the multiplication unit 804 are set according to the characteristics of the image data to be processed. As in the present embodiment, when frequency enhancement processing is performed on image data that has been subjected to speckle reduction processing, it is preferable to emphasize components having a relatively high spatial frequency. For example, the image data obtained by the ultrasonic frequency to transmit and receive ultrasonic beams of approximately 12 MHz, the case of dividing into six spatial frequency band by the down-sampling rate 1/2, when setting a large weighting coefficient k ₀ Good results have been obtained. The weighting coefficients k _{0 to} k _N−1 may be recorded in advance in the recording unit 41 shown in FIG. 9 in association with parameters such as the ultrasonic frequency, the depth of the subject, and the observation site. The user may input an arbitrary value. In the former case, an appropriate weighting coefficient is set according to those parameters, and in the latter case, a user-desired frequency enhancement effect can be obtained.

  As described above, according to the present embodiment, the image data processed in the speckle reduction processing unit 56 is subjected to frequency enhancement processing by frequency band division, so that an image in which speckle is reduced while suppressing artifacts. The contour can be emphasized. Accordingly, it is possible to obtain a highly accurate ultrasonic image.

  Next, an ultrasonic imaging apparatus according to the fourth embodiment of the present invention will be described. FIG. 12 is a block diagram showing the ultrasonic imaging apparatus according to this embodiment. This ultrasonic imaging apparatus includes a digital processing unit 90 shown in FIG. 12 instead of the digital processing unit 80 shown in FIG. About another structure, it is the same as that of the ultrasonic imaging device shown in FIG.

  The digital processing unit 80 shown in FIG. 10 and the digital processing unit 90 shown in FIG. 12 are different in the order of data processing after the reception focus processing. That is, the image data obtained based on the sound ray data generated by the reception focus processing in the digital processing unit shown in FIG. 12 is processed by the speckle reduction processing in the speckle reduction processing unit 56 and the frequency band division processing unit 81. After performing frequency band division processing and image processing such as gradation processing in the image processing unit 57, the scanning format is converted in the DSC 55.

  Here, an advantage of performing the frequency band division processing and the like before the conversion of the scanning format will be described. The image data before being subjected to processing such as scan conversion includes a wider band and an abundant amount of image information. Therefore, when performing frequency band division processing on image data before scan conversion, the frequency bandwidth can be narrowed and the number of adjustable frequency band divisions can be increased. That is, since the frequency enhancement process can be controlled more finely, an accurate ultrasonic image can be obtained by an accurate contour enhancement process, or an easy-to-see ultrasonic image can be generated by enhancing a desired frequency component. It becomes possible.

  INDUSTRIAL APPLICABILITY The present invention can be used in an ultrasonic imaging apparatus used for medical treatment and nondestructive inspection of structures.

It is a flowchart which shows the ultrasonic image processing method which concerns on one Embodiment of this invention. It is a figure for demonstrating the dilation and erosion in a general morphology process. It is a figure for demonstrating the speckle reduction by a general morphology process, and the artifact resulting from a mask size. It is a figure for demonstrating the morphology process performed in the ultrasonic image processing method which concerns on one Embodiment of this invention. It is a figure which shows the example of the dilation coefficient and erosion coefficient which are used in one Embodiment of this invention. 1 is a block diagram illustrating an ultrasonic imaging apparatus according to a first embodiment of the present invention. It is a schematic diagram which shows the structure of an ultrasonic endoscope. It is a block diagram which shows the ultrasonic imaging apparatus which concerns on the 2nd Embodiment of this invention. It is a figure for demonstrating the influence on the speckle pattern by polar coordinate transformation. It is a block diagram which shows the ultrasonic imaging apparatus which concerns on the 3rd Embodiment of this invention. It is a figure for demonstrating a frequency band division process. It is a block diagram which shows the ultrasonic imaging apparatus which concerns on the 4th Embodiment of this invention.

Explanation of symbols

10 pixels 11a to 11c high luminance regions 12, 16, 17 low luminance regions 13, 15, 15a to 15d mask region 14 target pixels 15a to 15d region 30 ultrasonic probe 31 insertion unit 32 operation unit 33 connection code 34 universal Code 35 Treatment instrument insertion port 40 Control unit 41 Recording unit 41a Weighting coefficient recording unit 42 Drive signal generating unit 43 Transmission / reception switching units 50, 70, 80, 90 Digital processing unit 51 Signal processing unit 52 A / D converter 53 Memory 54 Reception Focus control unit 55 Digital scan converter (DSC)
56 speckle reduction processing unit 57 image processing unit 58 image memory 59 D / A converter 60 display unit 61 input unit 61a transmission frequency input unit 61b observation region input unit 81 frequency band division processing unit 801 downsampling units 802 and 805 upsampling Section 803 Subtraction section 804, 807 Multiplication section 806, 808 Addition section

Claims (25)

A method of processing a plurality of signals generated based on a signal obtained by scanning an inside of a subject using an ultrasonic beam, each representing a pixel value of a plurality of pixels constituting an ultrasonic image,
For the plurality of signals, the pixel values of the plurality of pixels in the mask region determined based on the structural elements in the morphology processing are respectively multiplied by the first group of weighting factors set according to the positions in the mask region, (A) performing a process of converting the pixel value at the center of the mask into a maximum value among a plurality of pixel values multiplied by the weighting coefficient of the first group while moving the mask area;
For the plurality of signals processed in step (a), the pixel values of the plurality of pixels in the mask region determined based on the structural elements in the morphology processing are set according to the positions in the mask region. A step of multiplying the two groups of weighting coefficients and converting the pixel value at the center of the mask into a minimum value among a plurality of pixel values multiplied by the weighting coefficient of the second group while moving the mask area ( b) and
An ultrasonic image processing method comprising: The weighting coefficient of the first group is set to decrease as the distance from the mask center increases.
The weighting coefficient of the second group is set to increase as the distance from the mask center increases.
The ultrasonic image processing method according to claim 1.   The ultrasonic image processing according to claim 1 or 2, wherein the weighting coefficient of the first group and the weighting coefficient of the second group are set to be 1 when they are multiplied by corresponding coefficients in the mask region. Method.   The ultrasonic image processing method according to claim 1, wherein the weighting coefficients of the first group and the second group are set based on a frequency of ultrasonic waves to be transmitted.   The ultrasonic image processing method according to any one of claims 1 to 4, wherein the weighting coefficients of the first group and the second group are set based on a depth of the subject represented in the ultrasonic image. .   The ultrasound image processing method according to claim 1, wherein the weighting coefficients of the first group and the second group are set based on a site to be observed.   The ultrasonic image represented by the plurality of signals processed in step (b) is divided into a plurality of frequency components, each of the plurality of frequency components is multiplied by a predetermined weighting factor, and the weighting factor is multiplied. The ultrasonic image processing method according to claim 1, further comprising a step (c) of performing a process of adding a plurality of frequency components.   Prior to step (a), the method further comprises a step of converting a scan format for a plurality of signals respectively representing pixel values of a plurality of pixels constituting the ultrasonic image. The ultrasonic image processing method as described.   The ultrasonic image processing method according to claim 1, further comprising a step of converting a scan format for the signal processed in step (b) or step (c). An apparatus for processing a plurality of signals generated based on a signal obtained by scanning an inside of a subject using an ultrasonic beam, each representing a pixel value of a plurality of pixels constituting an ultrasonic image,
Storage means for storing the plurality of signals;
For the plurality of signals stored by the storage means, the pixel values of the plurality of pixels in the mask region determined based on the structural elements in the morphology processing are set according to the positions in the mask region. Each of the weighting coefficients is multiplied, and the first signal processing for converting the pixel value at the center of the mask to the maximum value among the plurality of pixel values multiplied by the weighting coefficient of the first group is performed while moving the mask area, For the signal subjected to the first signal processing, the pixel values of the plurality of pixels in the mask region determined based on the structural element in the morphology processing are set according to the position in the mask region. The second signal processing for converting the pixel value at the center of the mask to the minimum value among the plurality of pixel values multiplied by the weighting coefficient of the second group is performed by shifting the mask area. An image signal processing means for performing while,
An ultrasonic image processing apparatus comprising: The weighting coefficient of the first group is set to decrease as the distance from the mask center increases.
The weighting coefficient of the second group is set to increase as the distance from the mask center increases.
The ultrasonic image processing apparatus according to claim 10.   The ultrasonic image processing according to claim 10 or 11, wherein the weighting coefficient of the first group and the weighting coefficient of the second group are set to be 1 when they are multiplied by corresponding coefficients in the mask region. apparatus.   The ultrasonic image processing apparatus according to claim 10, wherein the weighting coefficients of the first group and the second group are set based on a frequency of ultrasonic waves to be transmitted.   The ultrasound image processing apparatus according to claim 10, wherein the weighting coefficients of the first group and the second group are set based on a depth of the subject represented in the ultrasound image. .   The ultrasonic image processing apparatus according to claim 10, wherein the weighting coefficients of the first group and the second group are set based on a site to be observed.   An ultrasound image represented by a plurality of signals processed by the image processing means is divided into a plurality of frequency components, each of the plurality of frequency components is multiplied by a predetermined weighting coefficient, and a plurality of weighted coefficients are multiplied. The ultrasonic image processing apparatus according to claim 10, further comprising a second image processing unit that performs a process of adding frequency components.   The plurality of signals according to any one of claims 10 to 16, wherein the plurality of signals are a plurality of signals obtained by scanning the inside of the subject with an ultrasonic endoscope that is inserted into the subject and used. Ultrasonic image processing device.   The ultrasonic image processing apparatus according to claim 10, further comprising a switching unit that switches whether to perform processing by the image processing unit based on information input by a user. A program that processes a plurality of signals that are generated based on signals obtained by scanning the inside of a subject using an ultrasonic beam and that respectively represent pixel values of a plurality of pixels that form an ultrasonic image,
For the plurality of signals, the pixel values of the plurality of pixels in the mask region determined based on the structural elements in the morphology processing are respectively multiplied by the first group of weighting factors set according to the positions in the mask region, A step (a) of performing a process of converting the pixel value at the center of the mask into a maximum value among a plurality of pixel values multiplied by the weighting coefficient of the first group while moving the mask area;
For the signal processed in step (a), the second group is set to pixel values of a plurality of pixels in the mask area determined based on the structural element in the morphology process according to the position in the mask area (B) A process of performing the process of converting the pixel value at the center of the mask to the minimum value among the plurality of pixel values multiplied by the weighting coefficient of the second group while moving the mask area. When,
Is an ultrasonic image processing program for causing the CPU to execute. The weighting coefficient of the first group is set to decrease as the distance from the mask center increases.
The weighting coefficient of the second group is set to increase as the distance from the mask center increases.
The ultrasonic image processing program according to claim 19.   21. The ultrasonic image processing according to claim 19 or 20, wherein the weighting coefficient of the first group and the weighting coefficient of the second group are set to be 1 when multiplied by corresponding coefficients in the mask region. program.   The ultrasound image processing program according to any one of claims 19 to 21, wherein the weighting coefficients of the first group and the second group are set based on a frequency of transmitted ultrasound.   The ultrasound image processing program according to any one of claims 19 to 22, wherein the weighting coefficients of the first group and the second group are set based on a depth of the subject represented in the ultrasound image. .   The ultrasound image processing program according to any one of claims 19 to 23, wherein the weighting coefficients of the first group and the second group are set based on a site to be observed.   The ultrasound image represented by the plurality of signals processed in step (b) is divided into a plurality of frequency components, each of the plurality of frequency components is multiplied by a predetermined weighting factor, and the weighting factor is multiplied. The ultrasonic image processing program according to any one of claims 19 to 24, further causing a CPU to execute a procedure for performing a process of adding a plurality of frequency components.