PHASE IMAGE PROCESSING APPARATUS AND PHASE IMAGE PROCESSING METHOD

Abstract

There is provided a phase-image processing apparatus and a phase-image processing method capable of highly accurately correcting a phase singularity included in a phase image. A phase-image processing apparatus that applies image processing to a phase image includes: a fringe pattern-creating unit that creates a plurality of fringe patterns based on a first interference fringe image corresponding to a first phase image including a phase singularity; a patch image-creating unit that creates a patch image based on the fringe pattern; an interference fringe image correcting unit that pastes the patch image to an area of the first interference fringe image corresponding to the phase singularity, corrects the first interference fringe image, and creates a second interference fringe image; and a phase image correcting unit that creates a second phase image from the second interference fringe image.

Claims

1. A phase-image processing apparatus that applies image processing to a phase image, the apparatus comprising: a fringe pattern-creating unit that creates a plurality of fringe patterns based on a first interference fringe image corresponding to a first phase image including a phase singularity; a patch image-creating unit that creates a patch image based on the fringe pattern; an interference fringe image correcting unit that pastes the patch image to an area of the first interference fringe image corresponding to the phase singularity, corrects the first interference fringe image, and creates a second interference fringe image; and a phase image correcting unit that creates a second phase image from the second interference fringe image.

2. The phase-image processing apparatus according to claim 1, wherein the patch image-creating unit creates the patch image by weighted addition of the plurality of fringe patterns.

3. The phase-image processing apparatus according to claim 2, wherein the patch image-creating unit creates the patch image using a sparse coding algorithm.

4. The phase-image processing apparatus according to claim 1, wherein the fringe pattern-creating unit displays a screen to which a parameter used for creating the fringe pattern is input, and creates the fringe pattern using a parameter input to the screen.

5. The phase-image processing apparatus according to claim 1, wherein the first interference fringe image is created based on the first phase image.

6. The phase-image processing apparatus according to claim 1, wherein after the second phase image is created to the first phase image, a process in the fringe pattern-creating unit, a process in the patch image-creating unit, a process in the interference fringe image correcting unit, and a process in the phase image correcting unit are repeated until the number of phase singularities included in the second phase image reaches a predetermined number or less, or until a number of iterations reaches a predetermined number of times.

7. A phase-image processing method that applies image processing to a phase image, the method comprising the steps of: creating a plurality of fringe patterns based on a first interference fringe image corresponding to a first phase image including a phase singularity; creating a patch image based on the fringe pattern; pasting the patch image to an area of the first interference fringe image corresponding to the phase singularity, correcting the first interference fringe image, and creating a second interference fringe image; and creating a second phase image from the second interference fringe image.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0011] FIG. 1 is a diagram of the overall structure of a phase-image processing apparatus;

[0012] FIG. 2 is a diagram showing examples of an interference fringe image, a two-dimensional spatial frequency spectrum, and a phase image;

[0013] FIG. 3 is a diagram showing an example flow of a process according to a first embodiment;

[0014] FIG. 4 is a diagram showing examples of areas of an interference fringe image corresponding to the defect area of a phase image;

[0015] FIG. 5 is a diagram showing an example of an interference fringe image prepared from a phase image;

[0016] FIG. 6 is a diagram showing an example of procedures for creating a plurality of fringe patterns;

[0017] FIG. 7 is a diagram showing an example of a corrected interference fringe image;

[0018] FIG. 8 is a diagram showing an example of a phase image created from a corrected interference fringe image; and

[0019] FIG. 9 is a diagram showing an example of a parameter input screen.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0020] In the following, an embodiment of a phase-image processing apparatus and a phase-image processing method according to the present invention will be described with reference to the accompanying drawings. Note that in the following description and the accompanying drawings, components having the same functional configuration are designated with the same reference signs, and the redundant description is omitted.

First Embodiment

[0021] FIG. 1 is a diagram showing the hardware configuration of a phase-image processing apparatus 101. The phase-image processing apparatus 101 is configured such that an operational unit 102, a memory 103, a storage unit 104, and a network adapter 105 are connected through a system bus 106 to make signals receivable and transmutable. The phase-image processing apparatus 101 is connected to an interference fringe image imaging device 110 and an image database 111 via a network 109 to make signals receivable and transmutable. To the phase-image processing apparatus 101, a display device 107 and an input device 108 are further connected. Here, the phrase to make signals receivable and transmutable expresses a state where signals are electrically and optically receivable and transmutable mutually or from one to the other, regardless of cable and wireless manners.

[0022] The operational unit 102 is a device that controls the operation of the components, specifically a Central Processing Unit (CPU), a Micro Processor Unit), or the like. The operational unit 102 loads a program stored in the storage unit 104 or data necessary to execute the program to the memory 103 for execution, and applies various image processes to a phase image. The memory 103 is a device that stores the program executed by the operational unit 102 and the midway point of a lapse of the arithmetic operation process. The storage unit 104 is a device that stores the program executed by the operational unit 102 and data necessary to execute the program, specifically a Hard Disk Drive (HDD), a Solid State Drive (SSD), and the like. The network adapter 105 is a device that connects the phase-image processing apparatus 101 to the network 109 such as a Local Area Network (LAN), a telephone circuit, the Internet, and the like. Various items of data handled by the operational unit 102 may be received from and transmitted to the outside of the phase-image processing apparatus 101 via the network 109 such as a LAN.

[0023] The display device 107 is a device that displays the processed result and the like of the phase-image processing apparatus 101, specifically a liquid crystal display and the like. The input device 108 is an operating device used by an operator to operate and instruct the phase-image processing apparatus 101, specifically a keyboard, a mouse, a touch panel, and the like. The mouse may be a pointing device such as a trackpad and a trackball.

[0024] The interference fringe image imaging device 110 is a device that images an interference fringe image obtained by interference between an object wave which is a wave created by reflection or transmutation of a wave applied to an observation target and a reference wave which is a reference wave. The wave applied to the observation target is a coherent wave such as light, laser beams, radio waves, and electron waves. The interference fringe image imaging device 110 is, for example, a transmission electron microscope (TEM) including an electron biprism, and the TEM including the electron biprism images an electron beam hologram as an interference fringe image.

[0025] The image database 111 is a database system that stores an interference fringe image acquired by the interference fringe image imaging device 110, a phase image prepared from the interference fringe image, and the like. The phase image is prepared by subjecting the interference fringe image to the arithmetic operation process using a Fourier transform method, a fringe scanning method, and the like.

[0026] Referring to FIG. 2, an example of procedures for preparing a phase image from an interference fringe image using a Fourier transform method will be described. The interference fringe image exemplified in FIG. 2 is an image imaged by the TEM including an electron biprism, and the image includes interference fringes which are fine fringe patterns going from the upper right to the lower left.

[0027] The interference fringe image in FIG. 2 is discrete Fourier transformed, and then a two-dimensional spatial frequency spectrum exemplified in FIG. 2 is obtained. The two-dimensional spatial frequency spectrum is a spectrum in which the intensity for every spatial frequency is plotted on a plane whose horizontal axis is the horizontal frequency and whose vertical axis is the vertical frequency. At the intersection point of these axes, the intensity of zero spatial frequency, i.e., a direct current component is plotted. Since the interference fringe image includes interference fringes going from the upper right to the lower left, a high-brightness area referred to as a sideband appears in the second quadrant and the fourth quadrant of the two-dimensional spatial frequency spectrum.

[0028] Subsequently, the sideband is cut from the two-dimensional spatial frequency spectrum, the sideband is pasted to the center of a new two-dimensional spatial frequency spectrum, i.e., at the position of the zero spatial frequency to conduct inverse discrete Fourier transform, and then a complex image R(x, y)+iI(x, y) is obtained, where (x, y) are the coordinates of the 2-dimensional image and i is an imaginary unit. The phase image as illustrated in FIG. 2 is obtained by the following formula.

atan 2(R(x,y),I(x,y))(Formula 1),

where atan 2( ) is the inverse function of tan( ) with two variables. Since the phase values obtained by (Formula 1) are wrapped into the range of [, ) [rad], it is necessary to unwrap the wrapped phase values.

[0029] However, in the case where the phase image includes a phase singularity, the unwrapping process is hindered. Therefore, in the first embodiment, the process flow for correcting the phase singularity included in the phase image with high accuracy is performed.

[0030] Referring to FIG. 3, an example of a process executed in the first embodiment will be described step by step.

[0031] (S300)

[0032] The operational unit 102 acquires a phase image. Specifically, a phase image is prepared from the interference fringe image imaged by the interference fringe image imaging device 110 using (Formula 1), or a phase image stored in advance in the storage unit 104 or the image database 111 is read.

[0033] (S301)

[0034] The operational unit 102 determines whether there is a phase singularity in the phase image acquired in S300, or a phase image to be created in S305 described later. In the case where there is a phase singularity in the phase image, the process proceeds to S302, and if not, the process is terminated.

[0035] The presence or absence of a phase singularity is determined based on, for example, the local circumferential integral of the phase gradient, and specifically the following formula is used.

W(upper rightupper left)+W(upper leftlower left)+W(lower leftlower right)+W(lower rightupper right)(Formula 2)

where W(upper rightupper left) is a function that returns 1 when the value obtained by subtracting the upper left pixel value from the upper right pixel value in the four adjacent 22 pixels is [rad] or more, 1 when the value is [rad] or less, and 0 when the value is within [, ) [rad]. When the value of (Formula 2) is not 0, it is determined that there is a phase singularity at the center point of the four pixels, and the four pixels around the phase singularity are extracted as the defect area, which is the area containing the phase singularity.

[0036] Note that the determination of the presence or absence of a phase singularity is not limited to the use of (Formula 2). For example, a boundary line may be formed by connecting pixel boundaries where the difference in pixel values between adjacent pixels in the phase image is [rad] or more, the end point where the line is broken may be determined to be a phase singularity, and the four pixels around the end point may be extracted as the defective area.

[0037] (S302)

[0038] The operational unit 102 extracts the area corresponding to the defect area extracted in S301 from the interference fringe image corresponding to the phase image. Specifically, an area with the same coordinates as the defect area is extracted from the interference fringe image. The area of the interference fringe image corresponding to the defect area has a blurred stripe pattern, as illustrated in FIG. 4.

[0039] Note that in the case where there is no interference fringe image corresponding to the phase image, a new interference fringe image corresponding to the phase image may be prepared by the following formula.

|exp[ikx]+exp[i(x,y)]|{circumflex over ()}2(Formula 3)

where (x, y) is the phase image, 2/k is the fringe period of the interference fringe to be created, and exp[ ] is an exponential function with the base of Napier number. By using (Formula 3), a periodic interference fringe image in the horizontal direction, as exemplified in FIG. 5, is prepared.

[0040] (S303)

[0041] The operational unit 102 creates a plurality of fringe patterns based on the interference fringe image corresponding to the phase image acquired in S300 or the phase image created in S305, which will be described later.

[0042] Referring to FIG. 6, an example of procedures for creating a plurality of fringe patterns will be described. First, a two-dimensional spatial frequency spectrum is obtained by discrete Fourier transforming the interference fringe image corresponding to the phase image. Subsequently, the position of the peak intensity is searched from the two-dimensional spatial frequency spectrum, and sidebands, which are areas around the position of the peak intensity, are extracted. Since the coordinates of each pixel in the sidebands indicate the direction and pitch of the stripes, a stripe pattern corresponding to each coordinate is created. Furthermore, by shifting the phase of the stripe pattern for each coordinate by one pixel, multiple stripe patterns with different phases are created. The 24 stripe patterns exemplified in FIG. 6 are the result of creating six stripe patterns with different phases at each of the four coordinates in the sidebands.

[0043] Note that the size of the stripe pattern is preferably the size of the defect area or less, e.g., one-fourth of the size of the defect area. However, the size of the stripe pattern has to be 16 pixels (44) or more such that the stripe pattern is clearly shown.

[0044] (S304)

[0045] The operational unit 102 creates a patch image using the stripe pattern created in S303, and corrects the interference fringe image by pasting the created patch image onto the interference fringe image. FIG. 7 shows an example of a corrected interference fringe image. The stripe pattern, which is blurred before correction, becomes clearer by pasting the patch image.

[0046] The patch image is created, for example, by weighted addition of the plurality of stripe patterns. Assuming that the period and direction of the stripe patterns in the patch image are formed of sparse stripe pattern pairs, the sparse coding algorithm can be used, and the patch image can be formed with a combination of a fewer stripe pattern, and thus it is possible to reduce processing time.

[0047] (S305)

[0048] The operational unit 102 creates a phase image from the corrected interference fringe image in S304. For example, (Formula 2) is used to create the phase image. FIG. 8 shows an example of a phase image created from the corrected interferogram image.

[0049] In the phase image created in step S305, the presence or absence of a phase singularity is determined in S301. When any phase singularity remains, the process from step S302 to step S305 is repeated. Note that the iterative process may be terminated when the number of phase singularities becomes less than a predetermined number, or when the number of iterations from S302 to S305 reaches a predetermined number.

[0050] As described above, by the flow of the described process, it is possible to highly accurately correct a phase singularity included in a phase image. Note that parameters used to prepare the interference fringe image in S302 and to create the fringe pattern in S303 may be set by the operator.

[0051] Referring to FIG. 9, an example of a parameter input screen displayed on the display device 107 will be described. The parameter input screen exemplified in FIG. 9 has a fringe pitch input unit 901, a fringe angle input unit 902, an interference fringe preparation button 903, an area size input unit 904, a pattern size input unit 905, a step input unit 906, an SCP input unit 907, and a start button 908.

[0052] To the fringe pitch input unit 901, the cycle of the interference fringe is input. To the fringe angle input unit 902, an angle indicating the direction of the interference fringe is input. The interference fringe preparation button 903 is pressed when the interference fringe image is prepared.

[0053] To the area size input unit 904, the size of the area corresponding to the defect area is input. To the pattern size input unit 905, the size of the fringe pattern is input. To the step input unit 906, the step when laying out the patch image in the region corresponding to the defect region is input. To the SCP input unit 907, a parameter used for the sparse coding algorithm is input. The start button 908 is pressed when starting the creation and pasting of the patch image.

[0054] As described above, the embodiment according to the present invention has been described. The present invention is not limited to the foregoing embodiment, and can be embodied by modifying the components without deviating from the gist of the invention. The plurality of components disclosed in the foregoing embodiment may be appropriately combined. Furthermore, some components may be deleted from all the components shown in the foregoing embodiment.

REFERENCE SIGNS LIST

[0055] 101: phase-image processing apparatus [0056] 102: operational unit [0057] 103: memory [0058] 104: storage unit [0059] 105: network adapter [0060] 106: system bus [0061] 107: display device [0062] 108: input device [0063] 109: network [0064] 110: interference fringe image imaging device [0065] 111: image database [0066] 901: fringe pitch input unit [0067] 902: fringe angle input unit [0068] 903: interference fringe preparation button [0069] 904: area size input unit [0070] 905: pattern size input unit [0071] 906: step input unit [0072] 907: SCP input unit [0073] 908: start button