JP2969077B2 - Ultrasound diagnostic image interpolation device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an ultrasonic diagnostic image interpolation apparatus, and more particularly, to a clear display of a structure located obliquely to a sound ray of an ultrasonic wave.

[0002]

2. Description of the Related Art In an ultrasonic diagnostic apparatus, an image is displayed based on a received signal obtained by an ultrasonic probe.
FIG. 10 is a schematic diagram illustrating a relationship between an ultrasonic sound ray and a display screen. The squares 2 arranged vertically and horizontally represent pixels of the display. Data along the ultrasonic sound ray 4 is obtained from the received ultrasonic waves. When displaying this on the screen, first, data on the ultrasonic sound ray 4 is sampled at the sampling point 6, and an interpolation pixel value which is a value corresponding to the pixel of the display is obtained by interpolation from the sampling data (real pixel value). Ask. Here, for example, the interpolated pixel value of the pixel a is defined by four sampling points 10, 12, 14,
Sampling data at 16 (points p1, p2, p3, p4)
It is determined from z1, z2, z3, and z4. The data converted into the rectangular coordinate system of the screen by interpolation is sequentially stored in the frame memory, and when the conversion of the data for one screen is completed, the data is read out from the frame memory in the order along the scanning line of the screen and the video signal is read. Is generated. By the way, in linear scan, interpolation between sound rays may be performed, but in other cases such as radial scan, convex scan, etc., the screen is determined from the coordinate system defined by the sound ray direction (R direction) and the sound ray angle direction (θ direction). Needs to be transformed in the rectangular coordinate system. Thereby, in an area where the density of sound rays is high (for example, an area close to the sound source in a sector scan or the like), excess data is arranged, while an area where the density of sound rays is low (for example, in a sector scan or the like) is far from the sound source. In the (region), the lack of data is compensated, and the image quality is improved by making the resolution on the image uniform.

Conventionally, as a technique for interpolating this image, there is a technique disclosed in Japanese Patent Application Laid-Open No. 57-108681. FIG. 11 is a conceptual diagram illustrating an interpolation method according to the related art. In the case of sector scan, the coordinates of a point on the screen can be represented by the coordinates R in the sound ray direction and the angle θ of the sound ray based on the ultrasonic sound ray. According to this coordinate system, the position of an arbitrary point in the rectangular area 20 surrounded by the point sampling points p1 to p4 on the screen is defined by the displacement ΔR in the R direction from the side p1p4 and the θ direction with respect to the side p3p4. And the displacement Δθ. Here, for the sake of simplicity, the description will be made based on the case where the square region 20 is normalized to a square having one side length. In this case, 0 ≦ ΔR
<1, 0 ≦ Δθ <1.

In the conventional method, the pixel value of a point a is set in the R direction, θ
It is determined by linear interpolation in two directions. That is, first, the sides p1p4 and p2p3 are obtained by linear interpolation in one direction, for example, the θ direction.
Are interpolated into (1−Δθ): Δθ, respectively, to obtain interpolation data at points b1 and b2. Next, interpolation in the R direction, which is another direction, is performed on the line segment b1b2, and the line segment b1b2 is converted into ΔR:
Interpolated data at point a which is internally divided into (1−ΔR) is obtained, and this is set as the interpolated pixel value at pixel a. The interpolated pixel value za at the pixel a is expressed by the following equation as a function of ΔR and Δθ.

Za (ΔR, Δθ) = (1−Δθ) · ｛(1−ΔR) · z4 + ΔR · z3｝ + Δθ · ｛(1−ΔR) · z1 + ΔR · z2｝ (1) Equation (1) Interpolated pixel values generally form a curved surface within a rectangular area. The value on the curved surface generally changes parabolically in an oblique direction shifted from the R direction and the θ direction, but changes linearly in a direction along the R direction and the θ direction. That is, this function can accurately reproduce a structure having a strong correlation in the R direction and the θ direction on an image by interpolation between sampling points. FIGS. 12A and 12B are examples in which conventional interpolation is performed based on sampling data obtained from a structure having a strong correlation in the R direction and the θ direction, respectively.
It is the schematic diagram which represented the appearance of the data value in the plane three-dimensionally by taking the axis | shaft of the data value perpendicularly to the plane. For example, in the case of a structure in which points that strongly reflect ultrasonic waves in the R direction are arranged, FIG.
As shown in FIG. 2 (a), a surface shape 30 of a mountain-shaped interpolation pixel value corresponding to the structure and arranged in the R direction is obtained, and the strength of the correlation of the structure in the R direction is favorably displayed on the screen. Is expressed.

[0006]

However, the conventional interpolation method has a problem that, for a structure crossing a rectangular area in an oblique direction (referred to as a ξ direction), the structure is not accurately reproduced on a screen. was there. FIG. 13 is an image diagram showing an example of sampling data on the R-θ plane when a structure is located in a diagonal direction of a rectangular area. In this figure, each intersection point 44 of the grid line 40 in the R direction and the grid line 42 in the θ direction is a sampling point, and only the value of the sampling data 48 on the diagonal 46 in the plane where the value of the sampling data is “0”. Is "1". At this time, the expected interpolation pixel value z of the pixel located on the diagonal line 46 is “1”. FIG.
14 is a graph showing the behavior of the interpolated pixel value z expected in the ξ direction, that is, on the diagonal line in the example shown in FIG. 13. On the other hand, FIG. 15 is a graph showing the behavior of the conventional interpolated pixel value z represented by the equation (1). In the traditional way,
The interpolated pixel value between the sampling points draws a parabola, and its value is smaller than “1”. Incidentally, in this example, z = １ ／ at the middle point of the sampling points. As described above, in the conventional interpolation method, when a structure that reflects ultrasonic waves is positioned obliquely with respect to a pixel, the reproduction accuracy of the structure on an image is poor, and the structure is not clearly displayed. was there. FIG. 16 is an image diagram specifically explaining a conventional problem. This figure shows an image obtained from a conventional ultrasonic diagnostic apparatus when a structure 52 such as a blood vessel exists in a sector scan range 50. In-vivo structures such as blood vessels are often positioned parallel to the body surface, and when these structures 52 are tomographically imaged from the body surface by, for example, sector scan, somewhere in the imaging range (in FIG. ), The structure 52 is inclined with respect to the sound ray. In such an area, a portion having a low luminance value may occur in the image of the structure 52 due to the above problem, and erroneous information may be provided to the observer. For example, in the figure, a portion 54 having a low luminance value is generated in the blood vessel wall due to the interpolation, resulting in a false contour as if the blood vessel wall is torn, which hinders accurate diagnosis.

The present invention provides an ultrasonic diagnostic image interpolation system for clearly displaying not only structures located in the R and θ directions based on sound rays but also structures located in oblique directions deviated from those directions. It is intended to provide a device.

[0008]

The ultrasonic diagnostic image interpolating apparatus determines a main diagonal line having a strong correlation between real pixel values at both ends of the two diagonal lines of a quadrangle composed of four real pixels. Main diagonal determination means, reference value determination means for obtaining an interpolation reference pixel value of a predetermined interpolation reference point located at the center of the rectangle, and four triangles defined by the interpolation reference point and one side of the rectangle, Triangle specifying means for specifying an interpolation triangle including an interpolation pixel; and the interpolation pixel value by two-dimensional linear interpolation based on the interpolation reference pixel value corresponding to three vertices of the interpolation triangle and the two actual pixel values. Means for determining an interpolated pixel value, wherein the reference value determining means determines that an interpolation straight line based on real pixel values at both ends on the main diagonal line is based on the magnitude relationship of the pixel values. To the actual pixel value of Unless it becomes valleys, and wherein the determined based the interpolation reference pixel value in the actual pixel values of the main diagonal of the both ends.

According to the ultrasonic diagnostic image interpolation apparatus, a diagonal line having a strong correlation, for example, a small difference between actual pixel values is set as a main diagonal line. When considering a plane (pixel value plane) formed by pixel values defined by a quadrangular region formed by four real pixels,
Unless the pixel value on the main diagonal is small and the surface forms a valley on the main diagonal, the strong correlation at the main diagonal is attributed to the presence of a structure that reflects ultrasonic waves along the main diagonal. Conceivable. Therefore, when it is considered that there is a structure on the main diagonal, the reference value determining means determines the interpolation reference pixel value of the interpolation reference point based on the actual pixel values at both ends of the main diagonal. The interpolation reference pixel value can be, for example, an arithmetic mean, a weighted average, a geometric mean, a maximum value, a minimum value, and the like of both actual pixel values. The interpolation reference point is, for example, an intersection of line segments connecting the midpoints of the two sides of the two sets of the rectangle,
A point approximately at the center of the rectangle, such as the center of gravity of the rectangle and the intersection of the main diagonal lines, can be selected. An intuitive image of the functions of the triangle specifying means and the interpolated pixel value determining means is that the pixel value plane is approximated by connecting the triangular pixel value planes, and the interpolated pixel value is determined by the pixel value plane. . That is, the quadrangular region is divided into four triangular regions by connecting each vertex thereof to the interpolation reference point, and is defined within each triangular region. At the vertex, the corresponding real pixel value or interpolation reference pixel value is defined. Is defined as the pixel value plane of the triangle. The triangle area to which the interpolation pixel belongs is determined by the triangle specifying means, and the value of the interpolation pixel on the pixel value plane corresponding to the triangle area to which the interpolation pixel belongs is set as the interpolation pixel value.

In the ultrasonic diagnostic image interpolation apparatus,
The reference value determining means, when the average value of the real pixel values at both ends of the main diagonal is equal to or greater than the minimum value of the real pixel values at both ends of another diagonal, sets the average value as the interpolation reference pixel value. It is characterized by doing. Further, in this ultrasonic diagnostic image interpolation device, the reference value determination means may include: when an average value of real pixel values at both ends of the main diagonal is smaller than a minimum value of real pixel values at both ends of another diagonal, The minimum value is used as the interpolation reference pixel value.

According to the ultrasonic diagnostic image interpolation apparatus, the case where the main diagonal forms a valley means that the average value of the real pixel values at both ends of the main diagonal is the sum of the real pixel values at both ends of another diagonal. It is defined as being smaller than the minimum value. If the main diagonal is not a valley, the interpolation reference pixel value is selected as the average of the actual pixel values at both ends of the main diagonal. As a result, the pixel value along the main diagonal can be expected to change between the actual pixel values, and the strength of the correlation at both ends that determine the main diagonal is maintained throughout the main diagonal. On the other hand, when the main diagonal forms a valley, it is unlikely that there is a structure that reflects the ultrasonic waves along the main diagonal. In this case, it is difficult to determine whether a valley is actually formed along the main diagonal or a ridge is formed along another diagonal. Therefore, in this case, the interpolation reference pixel value is not determined based on the real pixel values at both ends of the main diagonal, but as an intermediate value between the two diagonals, of the real pixel values at both ends of the another diagonal, The smaller value is used as the interpolation reference pixel value.

The ultrasonic diagnostic image interpolation apparatus is characterized in that, in the interpolation apparatus, the interpolation reference point is the center of the square. According to the ultrasonic diagnostic image interpolating apparatus, the interpolation reference point is set at the intersection of the line segment connecting the midpoints of the pair of opposite sides of the square and the line segment connecting the midpoints of the other set of opposite sides. . This is a point at which the average value of the actual pixel values at the four vertices of the rectangle is obtained in the conventional method. By using this point as the interpolation reference point, the processing in the triangle specifying means and the interpolation pixel value determining means is simplified.

[0013]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the present invention will be described below with reference to the drawings. First, the principle of the interpolation processing performed by the ultrasonic diagnostic image interpolation apparatus will be described. FIG.
FIG. 3 is a plan view showing a quadrangular region surrounded by four sampling points and subjected to interpolation. If an axis corresponding to the pixel value is taken in a direction perpendicular to this area, an actual pixel value plane corresponding to actual living tissue can be considered on the square area 100. In this case, since the rectangular area is an area sandwiched between two ultrasonic sound rays, the actual pixel value in the area is unknown except on the sound ray, and the interpolation in the rectangular area 100 is based on the actual shape. Is equivalent to determining a plane approximating an unknown actual pixel value plane. The real pixel values corresponding to the four sampling points p1 to p4 are z1 to z4, respectively, and the points on the real pixel value plane corresponding to these are Z1 to Z4.

Each side p1p2, p2p3, p3p of the rectangular area 100
4, the middle points of p4p1 are c1, c2, c3, and c4, respectively, and the center q0 of the rectangular area 100 is defined as the intersection of the line segment c1c3 and the line segment c2c4. This center q0 is set as an interpolation reference point. Interpolation reference point q0
The quadrilateral area 100 becomes four triangular areas Δq0p1p
2, Δq0p2p3, Δq0p3p4, Δq0p4p1 (triangle area 10
2, 104, 106, 108). As an approximate value of the actual pixel value at the interpolation reference point q0, the present apparatus determines an interpolation reference pixel value zm by a method described later. With the interpolation reference pixel value zm, the position of the point on the actual pixel surface corresponding to the point
Approximated by Zm.

In the present apparatus, the real pixel value plane is approximated by a multi-plane with the triangular pixel value planes ΔZmZ1Z2, ΔZmZ2Z3, ΔZmZ3Z4, and ΔZmZ4Z1, which are the pixel value planes defined in the respective triangular areas. The interpolated pixel value za at an arbitrary point a in the rectangular area 100 is represented by a point a
Is a value corresponding to.

FIG. 2 is a schematic diagram for explaining a three-dimensional image of pixel values in a rectangular area. FIG. 2A is a plan view of the three-dimensional image.
00 is shown. The pixel value axis 110 is directed to the front of the drawing. 2 (b), (c), (e) and (f)
FIG. 2D is a front view of the three-dimensional image viewed from the direction of arrow 112 in FIG. 2A, and FIGS.
3 is a side view as viewed from the direction of FIG.

Here, based on the difference between two actual pixel values located on each diagonal of the rectangular area 100, the correlation strength of the pixel values in the diagonal direction is defined. That is, each diagonal
Assuming that the differences between the real pixel values corresponding to p1p3 and p2p4 are d1 and d2, respectively, d1〓 | z1−z3 | (2) d2〓 | z2−z4 | (3) The smaller the value, the stronger the correlation.

FIGS. 2B and 2E show the difference in height between both ends of the lines Z1Z3 and Z2Z4 of the pixel value corresponding to each diagonal line. The diagonal with the smaller difference between the actual pixel values at both ends is called the main diagonal, and the other diagonal is called the sub-diagonal.
Since d1> d2 in (b) and d1 <d2 in (e), p2p4 in (b) and p1p3 in (e) are main diagonals, respectively. (B) shows an example in which the interpolation line Z2Z4 of the actual pixel value on the main diagonal line p2p4 is higher than the interpolation line Z1Z3 of the actual pixel value on the subdiagonal line p1p3.
On the other hand, (e) is an interpolation line Z1 of the actual pixel value on the main diagonal line p1p3.
An example is shown in which Z3 is at a position lower than the interpolation line Z2Z4 of the actual pixel value on the sub-diagonal line p2p4.

In an ultrasonic diagnostic image, as shown in FIG. 2B, there is a high possibility that a structure that reflects the ultrasonic wave exists between two points where the reflection intensity of the ultrasonic wave is large and the correlation is strong. Therefore, in this device, based on the characteristics of this ultrasonic diagnostic image,
The average value of the actual pixel values at both ends of the main diagonal is set as the interpolation reference pixel value zm. The interpolated pixel reference value zm may be another value based on the actual pixel values at both ends of the main diagonal, for example, a weighted average at an appropriate ratio, a geometric mean, or the like.

FIGS. 2C and 2D show interpolation reference pixel values zm.
4 shows three-dimensional images of four triangular pixel value planes when = (z2 + z4) / 2. As shown in these figures,
A structure having a high reflection intensity positioned obliquely in the quadrangular region 100 to be interpolated is displayed clearly on the image without interpolated pixel values dropping halfway in the present apparatus unlike the conventional apparatus.

FIG. 2 (e) shows a case where, contrary to FIG. 2 (b), two points having a small ultrasonic reflection intensity and a strong correlation are arranged on a diagonal line. In such a case, of course, there is a case where there is some structure in which the ultrasonic wave is hardly reflected along the main diagonal, but in addition, in a homogeneous base, for example, in a non-reflective area, the sub-diagonal direction is included. There is a high probability that a high-luminance structure exists. In other words, in this case, the real pixel values at both ends of the main diagonal should be linearly interpolated to form valleys of pixel values on the main diagonal, or the real pixel values at both ends of the sub-diagonal should be linearly interpolated to obtain pixel values on the sub-diagonal. Can't make a clear decision as to whether the mountain should be formed. Therefore, in this case, an intermediate value between the four actual pixel values is determined as the interpolation reference pixel value zm.

In the present apparatus, the case corresponding to (e) is defined as a case where the average value of the real pixel values at both ends of the main diagonal is smaller than the smaller value of the real pixel values at both ends of the sub-diagonal, and For zm, the smaller of the actual pixel values at both ends of the sub-diagonal is given. Incidentally, in the figure, zm = z4. In addition, for example, an average of four actual pixel values can be used as the interpolation reference pixel value zm.

FIGS. 2F and 2G show interpolation reference pixel values zm.
4 shows a three-dimensional image of four triangular pixel value planes when = z4. The shapes of the interpolated pixel value planes shown in these figures have a complete “valley” along the main diagonal line as shown in FIG. 3A and a sub-diagonal line as shown in FIG. It has an intermediate shape with a perfect “mountain”.

As described above, when the pixel value range of the interpolation straight line on the main diagonal is above the pixel value range of the interpolation straight line on the sub-diagonal (FIG. 2B), the interpolation reference pixel When the value zm is the average value of the actual pixel values at both ends of the main diagonal, the pixel value range of the interpolation straight line on the main diagonal is lower than the pixel value range of the interpolation straight line on the sub-diagonal (FIG. 2 (e)). , The interpolation reference pixel value zm is set to the intermediate value. The main feature of the present apparatus is that it can clearly display a high-luminance structure in an oblique direction.

The height difference between the interpolation straight line on the two main diagonal lines and the interpolation straight line on the sub diagonal line is shown in FIG.
In addition to the case where the pixel value ranges of the interpolation straight lines are vertically separated from each other as shown in (e), the ranges may overlap as an intermediate positional relationship.
Where to switch between the above two methods of determining the interpolation reference pixel value zm can be freely determined to some extent by design. In this device, when the average value of the real pixel values at both ends of the main diagonal is smaller than the smaller value of the real pixel values at both ends of the sub-diagonal, the interpolation reference pixel value zm is set to the above-described intermediate value, and In other cases, that is, when the average value of the real pixel values at both ends of the main diagonal is larger than the smaller value of the real pixel values at both ends of the sub-diagonal, zm is the average value of the real pixel values at both ends of the main diagonal.

As described above, the main diagonal line is determined, the interpolation reference pixel value is determined based on the main diagonal line, and a multi-plane approximation of the actual pixel value plane is performed using four triangular pixel value planes. Next, the interpolation pixel a is a triangle area Δq0p1p2, Δq0p2p3, Δq0p3p4,
Δq0p4p1 (triangle areas 102, 104, 106, 10
8). Here, for simplicity of the description, the description will be made based on the case where the rectangular area 100 is normalized to a square having one side as shown in FIG. The position of the interpolation pixel a is determined by the distance ΔR from the side p4p1 and the side
It is represented by the distance Δθ from p3p4, where 0 ≦ ΔR <
1, 0 ≦ Δθ <1. The interpolation pixel value za is the interpolation pixel a
Is determined by the following equation according to which triangular area belongs to.

When the point a belongs to the triangular area Δq0p1p2 (Δθ> ΔR and Δθ> 1-ΔR)

(Equation 1) When the point a belongs to the triangular area Δq0p2p3 (Δθ ≦ ΔR
And Δθ> 1−ΔR)

(Equation 2) When the point a belongs to the triangular area Δq0p3p4 (Δθ ≦ ΔR
And Δθ ≦ 1-ΔR)

(Equation 3) When the point a belongs to the triangular area Δq0p4p1 (Δθ> ΔR
And Δθ ≦ 1-ΔR)

(Equation 4) FIG. 5 is a schematic block diagram of the present apparatus. Ultrasonic reception signal is analog-digital converter (A / D converter)
At 150, it is converted to a digital signal. Line memory 15
Numeral 2 delays the digital signal by a time corresponding to the signal period of one sound ray. The digital signal is branched into a signal that passes through the line memory 152 and a signal that does not pass. Further, the delay time generated by the delay element 154 is a signal time difference between adjacent sampling points in the R direction. The four actual pixel values are input to the A / D converter 150 with a time lag.
2, signal lines 156, 158,
The actual pixel values z4, z3, z1, and z2 are output at the same timing to 160 and 162, respectively. These four real pixel values z
Are the interpolation calculation circuit 164 and the interpolation reference pixel value calculation circuit 166
Is input to The interpolation reference pixel value calculation circuit 166 obtains the interpolation reference pixel value zm from the actual pixel value z by the above-described method, and outputs zm to the interpolation calculation circuit 164.

On the other hand, the address arithmetic circuit 170 obtains the sound ray number 172 and the sampling timing 174 on the sound ray from the control unit for transmitting and receiving the ultrasonic wave, specifies the quadrangular area to be interpolated, and specifies the interpolation area within that area. Pixel coordinates ΔR, Δ
Calculate and output θ. The address operation circuit 170
Generates a clock for controlling the operation of the line memory 152 using the sampling timing 174. The coordinates of the interpolation pixel are input to the area determination circuit 176. The area determination circuit 176 is a triangle specifying unit that specifies which of the four triangle areas the interpolation pixel belongs to, and generates an area determination code. The area determination code and the coordinates of the interpolation pixel are input to the interpolation coefficient generation circuit 178. The interpolation coefficient generation circuit 178 generates an interpolation coefficient of the pixel value on the right side of the equations (4) to (7) using ΔR, Δθ and the area determination code as parameters. The area determination code and the interpolation coefficient are input to the interpolation operation circuit 164.

The interpolation operation circuit 164 selects an expression to be used for interpolation from the expressions (4) to (7) according to the area determination code.
The interpolation pixel value za is calculated using the interpolation coefficient obtained from the interpolation coefficient generation circuit 178. The interpolation pixel value za is stored in the frame memory 180, and when the interpolation for one screen is completed, it is converted into a video signal via a digital-analog converter (D / A converter) 182 and displayed on the display 184.

FIG. 6 is a block diagram of the interpolation reference pixel value calculation circuit 166. The difference between the actual pixel values located on the diagonal of the rectangular area 100 is calculated by subtracter 20 for each diagonal.
It is determined by 0. The difference between the two is calculated by the absolute value converter 20.
After being converted into absolute values by d2, that is, d1 and d2 defined by the equations (2) and (3), they are input to the comparator 204. The comparator 204 determines the magnitude relationship between d1 and d2 and determines the main diagonal (main diagonal determining means). The main diagonal line is a line segment p2p4 when d1 ≧ d2, and a line segment p1p3 when d1 <d2. In each case, the comparator 204 has an H (High) level and an L (Lo) level.
w) Output level signal. The real pixel values at both ends of the main diagonal line correspond to two selectors 20 according to the output of the comparator 204.
6 is selected. The average of these actual pixel values is calculated by the adder 208 and the divider 210,
2 is input.

Each of the two selectors 214 outputs an actual pixel value at each end of the sub-diagonal according to the output of the comparator 204. The comparator 216 compares the average value of the real pixel values at both ends of the main diagonal line with the real pixel value at each end of the sub diagonal line, and when the average value at both ends of the main diagonal line is smaller than the real pixel value to be compared with the sub diagonal line. , H level signals. The H level signal is input to the selector 212 by the AND gate 218 when the average value at both ends of the main diagonal is smaller than the actual pixel value at either end of the sub-diagonal. On the other hand, the smaller one of the real pixel values at both ends of the sub-diagonal is selected by the comparator 220 and the selector 222 and input to the selector 212. The selector 212 determines the interpolation reference pixel value zm according to the output signal of the AND gate 218. That is, when the average value of the real pixel values at both ends of the main diagonal is smaller than the real pixel values at both ends of the sub-diagonal, the smaller of the real pixel values on the sub-diagonal selected by the selector 222 is the interpolation reference pixel value.
zm, otherwise the divider 210
Are selected as interpolation reference pixel values zm (reference value determination means).

FIG. 7 is a block diagram of the area determination circuit 176. The comparator 300 determines the magnitude relationship between the input ΔR and Δθ, and the comparator 302 generates the input Δθ when (1-ΔR) is generated by the subtractor 304 based on the input ΔR.
Is determined, and the area determination code generator 306 determines (4) to (7) based on the outputs of the comparators 300 and 302.
Which of the expression cases corresponds to the case is output as an area determination code.

FIG. 8 is a block diagram of the interpolation operation circuit 164. There are four types of area determination codes corresponding to the four triangular areas. Selectors 400, 402, 404, 4
Numeral 06 selects and outputs one of the four input values according to the area determination code.
It is configured to output the real pixel z value or the interpolation reference pixel value zm included in the first to fourth terms of the equation corresponding to the triangular area including the interpolation pixel among the equations (7) to (7). The pixel value of each term is multiplied by the multiplier 408 with the coefficient of each term obtained from the interpolation coefficient generation circuit 178. The first to fourth terms are added by the adder 410, and an interpolation pixel value za is output (interpolation pixel value determination means).

In this case, the interpolation coefficient generation circuit 178
M. This ROM stores coefficients previously calculated using ΔR and Δθ as parameters for each triangular area. Alternatively, the interpolation coefficient generation circuit 178 may be configured to calculate the coefficients one by one by calculation. FIG. 9 is a schematic diagram showing the contents of a table stored in the ROM used for the interpolation coefficient generation circuit 178. k1, k
2, k3, k4 are each triangular area Δq0p1p2, Δq0p2p3, Δq0p
3p4, Δq0p4p1 (triangle areas 102, 104, 106,
108) are the coefficients of the first to fourth terms of equations (4) to (7), respectively. On the table, 0 ≦ ΔR <
In the range of 1, 0 ≦ Δθ <1, for example, ΔR,
The coefficients in each case in which Δθ is changed are stored, and the interpolation coefficient generation circuit 178 outputs the input ΔR, Δθ
And outputs the coefficient at the address corresponding to.

The above-mentioned hardware components of the apparatus are as follows.
It can be replaced by various configurations that perform the same function, and they may be realized as software on a computer as needed. In addition, the interpolation reference point may take an intersection of a diagonal line, a center of gravity of a rectangle, and the like, in addition to the center of the rectangle area defined above. In the above configuration, the timings of the four sampling data are aligned by the line memory 152 and the delay element 154. Instead, the sampling data arranged in the R direction and the θ direction is temporarily replaced.
After the data is stored in the frame memory, four data required for interpolation may be read. As ΔR and Δθ used in the above configuration, values that are easily obtained by approximation can be used.

[0036]

According to the ultrasonic diagnostic image interpolation apparatus,
Not only structures located in the R and θ directions based on sound rays, but also structures located in oblique directions deviated from those directions can be clearly displayed, and the effect of facilitating accurate diagnosis can be obtained. can get.

[Brief description of the drawings]

FIG. 1 is a plan view showing a quadrangular area where interpolation is performed.

FIG. 2 is a schematic diagram illustrating a three-dimensional image of pixel values in a rectangular area.

FIG. 3 is a schematic diagram illustrating a three-dimensional image of pixel values in a rectangular area.

FIG. 4 is a plan view showing a rectangular area to be interpolated in which the length of one side is normalized to 1;

FIG. 5 is a schematic block diagram of the present apparatus.

FIG. 6 is a block diagram of an interpolation reference pixel value calculation circuit.

FIG. 7 is a block diagram of an area determination circuit.

FIG. 8 is a block diagram of an interpolation operation circuit.

FIG. 9 is a schematic diagram showing the contents of a ROM used for an interpolation coefficient generation circuit.

FIG. 10 is a schematic diagram showing a relationship between an ultrasonic sound ray and a display screen.

FIG. 11 is a conceptual diagram illustrating an interpolation method according to the related art.

FIG. 12 is a schematic diagram showing a three-dimensional image of a conventional interpolation value for a structure having a strong correlation in the R direction and the θ direction.

FIG. 13 is an image diagram of sampling data on an R-θ plane when a structure is located in a diagonal direction of a quadrangular region.

FIG. 14 is a graph showing interpolated pixel values expected in a direction oblique to a sound ray.

FIG. 15 is a graph showing a conventional interpolated pixel value in a direction oblique to a sound ray.

FIG. 16 is an image diagram for explaining a conventional problem.

[Explanation of symbols]

4 Ultrasonic sound ray, 6, 10, 12, 14, 16 sampling points, 46 diagonal lines, 100 square area, 10
2,104,106,108 Triangular area, 150 A
/ D converter, 152 line memory, 154 delay element, 164 interpolation operation circuit, 166 interpolation reference pixel value operation circuit, 170 address operation circuit, 176 area judgment circuit, 178 interpolation coefficient generation circuit, 180 frame memory, 182 digital-analog Transducer, 184 display.

Claims (4)

(57) [Claims] 1. An ultrasonic diagnostic image interpolating apparatus for obtaining an interpolated pixel value of an interpolated pixel based on the actual pixel values of four interpolated pixels existing around the interpolated pixel on an ultrasonic image. Main diagonal determination means for determining a main diagonal line having a strong correlation between real pixel values at both ends of the two diagonal lines of a rectangle formed of pixels; and an interpolation reference of a predetermined interpolation reference point located at the center of the rectangle. Reference value determining means for obtaining a pixel value; triangle specifying means for specifying an interpolation triangle including the interpolation pixel from four triangles defined by the interpolation reference point and one side of the rectangle; three vertices of the interpolation triangle And an interpolated pixel value determining means for obtaining the interpolated pixel value by two-dimensional linear interpolation based on the interpolated reference pixel value and the two actual pixel values. An interpolation straight line based on the real pixel values at both ends on the diagonal line, except for a case where the real pixel value of the other two real pixels becomes a valley based on the magnitude relationship between the pixel values, is defined as the interpolation reference pixel value. An image interpolating apparatus for ultrasonic diagnostics, which is determined based on actual pixel values at both ends of a main diagonal line. 2. The ultrasonic diagnostic image interpolation apparatus according to claim 1, wherein the reference value determination unit calculates an average value of real pixel values at both ends of the main diagonal line as an average value of real pixel values at both ends of another diagonal line. An ultrasonic diagnostic image interpolation apparatus, wherein the average value is used as the interpolation reference pixel value when the average value is equal to or greater than the minimum value. 3. The ultrasonic diagnostic image interpolation apparatus according to claim 2, wherein the reference value determining means calculates an average value of real pixel values at both ends of the main diagonal line as an average value of real pixel values at both ends of another diagonal line. If the value is smaller than the minimum value, the minimum value is used as the interpolation reference pixel value. 4. The ultrasonic diagnostic image interpolating apparatus according to claim 1, wherein the interpolation reference point is a center of the quadrangle. .