WO2009096422A1 - Three-dimensional shape measurement device, method, and program - Google Patents

Abstract

A three-dimensional shape measurement device, method, and program capable of improving precision of three-dimensional shape measurement. A three-dimensional shape measurement part (51) calculates the focusing measures of the respective pixels of images having different focus positions with respect to an object to be measured to detect focusing positions of the respective pixels measuring the three-dimensional shape of the object to be measured. An effective pixel extraction section (72), using at least one of a luminance, hue, and color saturation as judgment values, extracts effective pixels the judgment values of which are each different from the judgment value of an attention pixel by a difference within an allowable range, from pixels within a predetermined range in the vicinity of the attention pixel. A focusing measure calculation section (73) calculates the focusing measure of the attention pixel using the extracted effective pixels. A focusing position detection section (74) detects the focus position where the focusing measure is peaked, as the focusing position of the attention pixel. This invention can be applied to the three-dimensional shape measurement device, for example.

Description

Three-dimensional shape measuring apparatus and method, and program

The present invention relates to a three-dimensional shape measurement apparatus and method, and a program, and more particularly, to a three-dimensional shape measurement apparatus and method, and a program that improve the measurement accuracy of a three-dimensional shape.

Conventionally, an SFF (Shape-From-Focus) method is known as a method for measuring the three-dimensional shape of an object using a two-dimensional image (see, for example, Non-Patent Document 1). The SFF method is an effective method when the surface of the measurement object has a texture. In the SFF method, a plurality of images with different focal positions are photographed while moving the imaging device with respect to the measurement object, and a differential operation is performed for each pixel of the obtained image to indicate the degree of focus. Calculate the measure. Then, the three-dimensional shape of the measurement object is measured based on the focal position when the focus measure of each pixel is maximized.

Further, conventionally, it has been proposed to measure the three-dimensional shape of a measurement object having no texture on the surface by irradiating the measurement object with a predetermined texture pattern using the SFF method. (For example, refer to Patent Document 1).

Japanese Patent No. 3321866 S.K.Nayar, Shape from Focus, tech. Report CMU-RI-TR-89-27, Robotics Institute, Carnegie Mellon University, November, 1989

By the way, in the range where the object to be measured is not in focus, the bright part tends to swell and be photographed, and the tendency becomes stronger as the focus shift increases. For this reason, there are cases where a dark part is originally brightly photographed at a boundary part where the luminance or color of the measurement object changes greatly. In the SFF method, due to this phenomenon, a position where each pixel is in focus (hereinafter referred to as a focus position) is erroneously detected, and measurement accuracy may be reduced.

The present invention has been made in view of such a situation, and is intended to improve the measurement accuracy of a three-dimensional shape.

The three-dimensional shape measurement apparatus according to one aspect of the present invention calculates a focus measure indicating a degree of focus on each pixel of a plurality of images having different focus positions with respect to a measurement target, and the focus measure is calculated as the focus measure. In the three-dimensional shape measurement apparatus that measures the three-dimensional shape of the measurement object by detecting the focus position based on the focus position, at least one of luminance, hue, and saturation is used as a determination value, and the focus of the target pixel is determined. Effective pixel extraction means for extracting effective pixels having the determination value whose difference from the determination value of the target pixel is within an allowable range from pixels within a predetermined range in the vicinity of the target pixel used for calculation of the focus measure And using the extracted effective pixels, a focus measure calculation means for calculating the focus measure at the target pixel, and a focus position at which the focus measure reaches a peak is detected as the focus position. And a focus position detecting means.

The three-dimensional shape measurement method according to one aspect of the present invention calculates a focus measure indicating a degree of focus on each pixel of a plurality of images having different focus positions with respect to a measurement target, and the focus measure is calculated as the focus measure. In the three-dimensional shape measuring method of the three-dimensional shape measuring apparatus for measuring the three-dimensional shape of the measurement object by detecting the in-focus position based on at least one of luminance, hue, and saturation as a determination value An effective pixel having the determination value whose difference from the determination value of the target pixel is within an allowable range from a pixel within a predetermined range in the vicinity of the target pixel used for calculation of the in-focus measure in the target pixel. An effective pixel extraction step to extract, a focus measure calculation step to calculate the focus measure at the pixel of interest using the extracted effective pixel, and a focus position at which the focus measure reaches a peak And a focus position detection step of detecting as said in-focus position.

A program according to an aspect of the present invention calculates a focus measure indicating a degree of focus on each pixel of a plurality of images having different focus positions with respect to a measurement target, and focuses on the basis of the focus measure. In a program that causes a computer to execute a process of measuring a three-dimensional shape of the measurement object by detecting a position, at least one of luminance, hue, and saturation is used as a determination value, and the focus on a target pixel An effective pixel extracting step of extracting effective pixels having the determination value whose difference from the determination value of the target pixel is within an allowable range from pixels within a predetermined range in the vicinity of the target pixel used for calculating the measure; A focus measure calculation step of calculating the focus measure at the target pixel using the extracted effective pixels, and a focus position where the focus measure reaches a peak. To execute processing including a focus position detection step of detecting a focus position on the computer.

In one aspect of the present invention, at least one of luminance, hue, and saturation is used as a determination value, and pixels within a predetermined range near the target pixel used for calculation of the focus measure in the target pixel are An effective pixel having the determination value whose difference from the determination value of the target pixel is within an allowable range is extracted, and using the extracted effective pixel, the in-focus measure at the target pixel is calculated, A focus position where the focus measure reaches a peak is detected as the focus position.

According to the present invention, the measurement accuracy of the three-dimensional shape is improved.

It is a schematic diagram which shows one Embodiment of the three-dimensional shape measurement system to which this invention is applied. It is a figure which shows the example of a structure of the function implement | achieved by computer. It is a flowchart for demonstrating the three-dimensional shape measurement process performed by a three-dimensional shape measurement system. It is a flowchart for demonstrating the three-dimensional shape measurement process performed by a three-dimensional shape measurement system. It is a figure which shows the example of a focus position. It is a figure which shows the example of the focus position in an original image. It is a figure which shows the example of an attention range. It is a figure which shows the example of distribution of the brightness | luminance of an attention range. It is a figure which shows the example of distribution of the hue or saturation of an attention range. It is a graph which shows the example of distribution of a focus measure. It is a graph for demonstrating the interpolation process of a focusing measure.

Explanation of symbols

1 3D shape measurement system, 2 measurement object, 11 imaging device, 12 computer, 51 3D shape measurement unit, 61 imaging control unit, 62 measurement unit, 71 pre-processing unit, 72 effective pixel extraction unit, 73 focusing measure Calculation unit, 74 Focus position detection unit, 75 3D shape data generation unit

Embodiments to which the present invention is applied will be described below with reference to the drawings.

FIG. 1 is a diagram showing an embodiment of a three-dimensional shape measurement system to which the present invention is applied. The three-dimensional shape measurement system 1 of FIG. 1 is configured to include an imaging device 11, a computer 12, and a display 13. The imaging device 11 and the computer 12 are connected via a cable 14, and the computer 12 and the display 13 are connected to each other. They are connected via a cable 15. Further, the image pickup apparatus 11 schematically shown in FIG. 1 is configured to include at least an optical lens 21 and a photodetector 22 including an image pickup device such as a CCD (Charge Coupled Device).

In the three-dimensional shape measurement system 1, as will be described later with reference to FIGS. 3 and 4, the imaging device 11 has a height (in the Z-axis direction) with respect to the measurement object 2 installed on the stage 3 of the measurement microscope. While changing the position), a plurality of images with different focal positions are taken with respect to the measurement object 2, and the computer 12 measures the three-dimensional shape of the measurement object 2 using the plurality of images.

FIG. 2 is a block diagram showing an example of a functional configuration realized when a processor (for example, a CPU (Central Processing Unit), a DSP (Digital Signal Processor), etc.) of the computer 12 executes a predetermined program. The three-dimensional shape measuring unit 51 is realized by the processor of the computer 12 executing a predetermined program. The three-dimensional shape measurement unit 51 includes an imaging control unit 61 and a measurement unit 62.

The imaging control unit 61 controls the imaging device 11 to image the measurement object 2 while changing the focal position. The imaging control unit 61 acquires a captured image (hereinafter referred to as an original image) from the imaging device 11 and supplies the acquired image to the preprocessing unit 71 of the measurement unit 62. In addition, the imaging control unit 61 notifies the preprocessing unit 71 of the end of imaging when imaging of the measurement object 2 at all focal positions is completed.

The measuring unit 62 measures the three-dimensional shape of the measurement object 2 using a plurality of original images having different focal positions. The measurement unit 62 includes a preprocessing unit 71, an effective pixel extraction unit 72, a focus measure calculation unit 73, a focus position detection unit 74, and a three-dimensional shape data generation unit 75.

As will be described later with reference to FIGS. 3 and 4, the preprocessing unit 71 performs predetermined preprocessing on the original image and stores an image generated as a result (hereinafter also referred to as a preprocessed image) in the memory. 52 is stored. Further, when the end of shooting is notified from the shooting control unit 61, the preprocessing unit 71 transfers the notification to the effective pixel extraction unit 72.

As will be described later with reference to FIGS. 3 and 4, the effective pixel extraction unit 72 is a pixel in a predetermined range (hereinafter referred to as a target range) in the vicinity of the target pixel for which a focus measure is calculated. Then, using the luminance, hue, or saturation allowable value set by the user, the effective pixel used for calculating the focus measure at the target pixel is extracted, and the luminance value of the effective pixel is extracted from the preprocessed image. . The effective pixel extraction unit 72 supplies information indicating the luminance value of the extracted effective pixels to the focus measure calculation unit 73.

As will be described later with reference to FIGS. 3 and 4, the focus measure calculation unit 73 calculates the focus measure of the target pixel using the extracted luminance value of the effective pixel, and calculates the calculated focus measure. Is supplied to the in-focus position detector 74.

As will be described later with reference to FIGS. 3 and 4, the in-focus position detection unit 74 uses the luminance tolerance value set by the user based on the calculated in-focus measure, and adjusts the focus pixel of interest. The focus position is detected, and the detected focus position is stored in the memory 52. Further, when the focus position detection unit 74 detects the focus positions of all the pixels, the focus position detection unit 74 notifies the three-dimensional shape data generation unit 75 of the end of the detection of the focus position.

The 3D shape data generation unit 75 generates 3D shape data based on the in-focus position of each pixel stored in the memory 52, and outputs it to the subsequent stage.

Next, the three-dimensional shape measurement process executed by the three-dimensional shape measurement system 1 will be described with reference to the flowcharts of FIGS. This process is started, for example, when the user inputs a command for measuring the three-dimensional shape of the measuring object 2 on the stage 3 via an input unit (not shown) of the computer 12.

In step S1, the effective pixel extraction unit 72 and the focus position detection unit 74 acquire various allowable values. Specifically, for example, the user sets which of luminance, hue, and saturation to use as a determination value used for effective pixel extraction via an input unit (not shown) of the computer 12 and determines The permissible values of luminance, hue, and saturation used in the above are input to the computer 12. The effective pixel extraction unit 72 acquires the input allowable value. Note that any one of luminance, hue, and saturation may be used for the determination value, or two or more types may be used in combination.

Further, for example, the user inputs an allowable luminance value used for determination of the maximum value of the focus measure to the computer 12 via an input unit (not shown) of the computer 12, and the focus position detection unit 74 receives the allowable value. Get the value.

In step S <b> 2, the imaging device 11 captures an image of the measurement object under the control of the imaging control unit 61. The imaging apparatus 11 supplies an original image represented by the three primary colors of RGB obtained as a result of shooting to the shooting control unit 61 via the cable 14. The imaging control unit 61 supplies the acquired original image to the preprocessing unit 71.

In step S3, the preprocessing unit 71 performs preprocessing of the image. Specifically, the preprocessing unit 71 performs an Fourier transform on the acquired original image, removes a low frequency component equal to or lower than a predetermined frequency, and a noise component within a predetermined frequency range, and then performs an inverse Fourier transform. As a result, the low-frequency component and the noise component are removed from the original image, and an RGB image (hereinafter referred to as a preprocessed RGB image) from which the medium-high frequency component is extracted is generated. The preprocessing unit 71 stores the generated preprocessed RGB image in the memory 52. Further, the preprocessing unit 71 converts the generated preprocessed RGB image into an YPbPr image, and stores the converted image (hereinafter referred to as a preprocessed YPbPr image) in the memory 52.

In step S4, the photographing control unit 61 determines whether or not photographing has been performed at all focal positions. If it is determined that shooting has not been performed at all focal positions, the process proceeds to step S5.

In step S5, the imaging control unit 61 changes the focal position. That is, the imaging control unit 61 moves the imaging device 11 in the Z-axis direction so that the focal position becomes the following value.

Thereafter, the process returns to step S2, and the processes of steps S2 to S5 are repeatedly executed until it is determined in step S4 that images have been taken at all the focal positions. Thereby, for example, as shown in FIG. 5, N original images are photographed while moving the focal position from the lowermost end to the uppermost end of the measuring object 2 at a predetermined interval, and preprocessing is performed from each original image. An RGB image and a preprocessed YPbPr image are generated and stored in the memory 52. In FIG. 5, the vertical axis represents the focal position, and on the vertical axis, an index representing the setting order of the focal position is shown instead of the value of the focal position.

FIG. 6 shows an example of an original image taken at the kth focal position in FIG. Each square box in FIG. 6 represents a pixel. In the measurement object 2 in FIG. 6, the region Rf indicates a region in which the imaging apparatus 11 is in focus, the region Rin indicates a region inside the region in focus, and the region Rout indicates the focus. The area outside the area where the That is, the images in the region Rin and the region Rout are so-called defocused images that are out of focus.

Returning to FIG. 3, on the other hand, when it is determined in step S <b> 4 that the photographing control unit 61 has photographed at all the focal positions, it notifies the effective pixel extracting unit 72 of the end of photographing through the preprocessing unit 71. The process proceeds to step S6.

In step S6, the measurement unit 62 selects one of the pixels for which the in-focus position has not been obtained and sets it as the target pixel. Note that the pixel of interest is set, for example, in raster scan order.

In step S7, the measuring unit 62 selects one of the images for which the focus measure at the current target pixel is not obtained and sets it as the target image. Note that the attention image is set, for example, in the order of shooting (index order).

In step S8, the effective pixel extraction unit 72 selects one pixel to be processed from the pixels within the attention range. That is, the effective pixel extraction unit 72 selects, as a pixel to be processed, one of the pixels in the attention range that has not been determined whether or not it is an effective pixel.

FIG. 7 shows an example of the attention range when the pixel PA1 in FIG. 6 is set as the attention pixel. In the example of FIG. 7, a 5 × 5 pixel range RA1 centered on the target pixel PA1 is set as the target range.

In step S9, the effective pixel extraction unit 72 determines whether the difference between the determination values of the target pixel and the pixel to be processed is within an allowable range. Specifically, the effective pixel extraction unit 72 obtains the determination value of the target pixel and the pixel to be processed from the preprocessed RGB image and the preprocessed YPbPr image of the target image stored in the memory 52. The effective pixel extraction unit 72 takes the difference between the determination values of the target pixel and the pixel to be processed, and compares the absolute value of the difference value with the allowable value set in step S1. When the absolute value of the difference value is equal to or smaller than the allowable value, the effective pixel extraction unit 72 determines that the difference between the determination values of the target pixel and the pixel to be processed is within the allowable range, that is, the pixel to be processed is valid The pixel is determined to be a pixel, and the process proceeds to step S10.

In addition, when it is set to use two or more types of determination values, when the absolute value of the difference value of all types of determination values is less than or equal to the allowable value, the difference between the determination values of the target pixel and the pixel to be processed Is determined to be within the allowable range. For example, when three types of determination values of luminance, hue, and saturation are set to be used, the absolute value of the difference value between the target pixel and the pixel to be processed is an allowable value for all of luminance, hue, and saturation. When it is below, it is determined that the difference between the determination values of the target pixel and the pixel to be processed is within the allowable range.

In step S10, the effective pixel extraction unit 72 extracts the luminance value of the pixel to be processed. That is, the effective pixel extraction unit 72 extracts the luminance value of the current pixel to be processed from the preprocessed YPbPr image of the target image stored in the memory 52.

On the other hand, in step S9, when the absolute value of the difference value of at least one kind of determination value exceeds the allowable value, it is determined that the difference between the determination value of the target pixel and the pixel to be processed exceeds the allowable range, The process of step S10 is skipped and the process proceeds to step S11.

In step S11, the effective pixel extraction unit 72 determines whether all the pixels within the attention range have been determined. If it is determined that all the pixels within the attention range have not yet been determined, the process returns to step S8. After that, until it is determined in step S11 that all the pixels in the attention range have been determined, the processing in steps S8 to S11 is repeatedly performed, and the difference in determination value from the pixel in the attention range to the attention pixel is determined. An effective pixel having a determination value within the allowable range is extracted, and further, a luminance value of the effective pixel is extracted.

On the other hand, if it is determined in step S11 that all pixels within the range of interest have been determined, the process proceeds to step S12.

In step S12, the focus measure calculation unit 73 calculates the focus measure at the target pixel of the target image. Specifically, the effective pixel extraction unit 72 supplies information indicating the luminance value of the extracted effective pixel to the focus measure calculation unit 73. The focus measure calculation unit 73 calculates the focus measure at the target pixel of the target image using the following equation (1).

Focus measure = (Σ | Luminance value of target pixel−Luminance value of effective pixel |) ÷ Number of effective pixels (1)

In other words, the focus measure is an average value of absolute values of difference values between the luminance value of the target pixel and the luminance value of the effective pixel within the target range. Therefore, the value of the focus measure increases as the difference in luminance value between the target pixel and the effective pixel increases, that is, as the contrast between the target pixel and the effective pixel increases.

The focus measure calculation unit 73 supplies the calculated focus measure to the focus position detection unit 74. The focus measure calculation unit 73 stores the calculated focus measure, the luminance value of the target pixel of the current target image, and the focal position of the current target image in the memory 52 in association with each other.

For example, as shown in FIG. 8, in the attention range RA1, between the range RA1a including the attention pixel PA1 (the hatched portion in the lower right direction) and the range RA1b (the hatched portion in the lower left direction), When the measurement object 2 has a step, the brightness is greatly different between the range RA1a and the range RA1b. Therefore, when the focus measure is obtained using all the pixels in the attention range RA1, it is affected by both the ranges RA1a and RA1b having different heights, and corresponds to the height between the range RA1a and the range RA1b. The focus measure is calculated.

On the other hand, as described above, by extracting effective pixels from the range of interest and calculating the focus measure using only effective pixels, for example, in the case of the example of FIG. 8, only the luminance of the pixels in the range RA1a is used. Thus, the focus measure at the target pixel PA1 can be calculated more accurately.

FIG. 9 schematically shows the state of hue or saturation distribution within the range RA1a. As shown in FIG. 9, the hue or saturation also varies between the range RA1a and the range RB1b. There is a high possibility that it will be greatly different. Therefore, even if hue or saturation is used as the determination value, the focus measure at the target pixel PA1 can be calculated more accurately by using only the luminance of the pixel within the range RA1a or a range close thereto. Note that effective pixels extracted using luminance, hue, or saturation do not necessarily match, and using all three types as determination values makes it possible to calculate a focus measure more accurately. .

In step S13, the focus position detection unit 74 determines whether the focus measure is the maximum. Specifically, the in-focus position detection unit 74 determines that the in-focus measure is the maximum when the in-focus measure calculated in step S13 is larger than the maximum value of the in-focus measure stored so far in the memory 52. The process proceeds to step S14. Note that when the first focus measure is calculated for the current pixel of interest, since the maximum value of the focus measure is not yet stored in the memory 52, it is determined that the calculated focus measure is unconditionally maximum. Is done.

In step S14, the focus position detection unit 74 determines whether the luminance value of the target pixel is within the allowable range. Specifically, the in-focus position detection unit 74 stores the brightness value of the target pixel at the current focal position (the brightness value of the target pixel of the current target image) and the focus that has been stored in the memory 52 so far. The difference between the luminance value at the focal position where the measure is maximized (the luminance value of the target pixel of the image where the in-focus measure has been maximized so far) is taken, and the allowable value set in step S1. And compare. If the difference value is less than or equal to the allowable value, the in-focus position detection unit 74 determines that the luminance value of the target pixel is within the allowable range, and the process proceeds to step S15.

In step S15, the focus position detection unit 74 stores the current focus measure as a maximum value. That is, the focus position detection unit 74 sets the focus measure at the target pixel of the current target image as the maximum value of the focus measure, and the luminance value of the target pixel of the current target image and the focus position of the current target image. And stored in the memory 52.

On the other hand, if the calculated difference value exceeds the allowable value in step S14, it is determined that the luminance value of the target pixel exceeds the allowable range, the process of step S15 is skipped, and the process proceeds to step S16. That is, the focus measure at the current focus position is not stored as the maximum value.

FIG. 10 is a graph showing an example of the relationship between the focus position and the focus measure in the target pixel PA11 at the boundary portion where the luminance or color of the measurement object changes greatly. As described above, in the boundary portion where the luminance or color of the measurement object changes greatly, the bright portion may expand as the focus shifts, and the originally dark portion may be photographed brightly. Accordingly, as shown in FIG. 10, when the point of the measurement object corresponding to the target pixel PA11 is focused and the target pixel PA11 is photographed with the original dark brightness, the peak PK11 of the focus measure appears. Thereafter, as the focus on the point of the measurement object corresponding to the target pixel PA11 shifts, the luminance of the target pixel PA11 becomes brighter, and the peak PK12 of the focus measure may appear again.

In contrast, the difference between the luminance of the pixel of interest PA11 at the focal point where the focus measure is the peak PK11 and the luminance of the pixel of interest PA11 at the focal point where the focus measure is the peak PK12 is determined by the determination process in step S14. If it is larger, the focus position where the focus measure reaches the peak PK12 and the focus position in the vicinity thereof are excluded from the detection target of the maximum value of the focus measure.
Thereby, the peak of the focus measure is detected at the focus position where the focus is not in focus, and the erroneous detection of the focus position is prevented.

In step S16, the focusing speed calculation unit 73 determines whether all the images have been processed. If it is determined that all the images have not yet been processed, the process returns to step S7. Thereafter, until it is determined in step S16 that all the images have been processed, the processes in steps S7 to S16 are repeatedly performed, and the focus measure at all the focus positions is obtained for the target pixel, and the focus measure The maximum value is detected.

On the other hand, if it is determined in step S16 that all the images have been processed, the process proceeds to step S17.

In step S17, the focus position detection unit 74 detects the focus position. Specifically, the focus position detection unit 74 reads, from the memory 52, the focus position at the focus position where the focus measure is maximum and the focus positions before and after the focus position for the target pixel. For example, when the focus measure becomes the maximum at the k + 1th focus position, the focus measure at the kth to k + 2th focus positions is read from the memory 52. The focus position detection unit 74 performs interpolation of data between the three points by modeling the relationship between the focus position and the focus measure using a Gaussian function from the read focus positions and focus measures of the three points. Then, based on the interpolated data, the focus position detection unit 74 detects the focus position where the focus measure reaches a peak as the focus position at the target pixel. The focus position detection unit 74 stores the detected focus position in the memory 52.

FIG. 11 shows an example of a graph when the relationship between the focus position and the focus measure at points 1 to N is modeled by a Gaussian function. In this way, by modeling the relationship between the focal position and the in-focus measure with a Gaussian function, it is possible to detect the in-focus position at the target pixel with higher accuracy than the interval between the focal positions at which the images were actually captured. Thus, for example, the focal position between k + 1 and k + 2 corresponding to the peak PK21 in FIG. 11 can be detected as the in-focus position.

It should be noted that the relationship between the focal position and the focus measure may be modeled by a Gaussian function using data of four or more points. Further, instead of a Gaussian function, data interpolation may be performed by a quadratic function interpolation calculation. Furthermore, when the interval between the focal positions at which an image is taken is sufficiently small, data interpolation may be performed by taking a moving average of the focus measure.

In step S18, the focus position detection unit 74 determines whether the focus positions of all the pixels have been detected. If it is determined that the focus positions of all the pixels have not been detected yet, the process returns to step S6, and steps S6 to S18 are performed until it is determined in step S18 that the focus positions of all the pixels have been detected. This process is repeatedly executed.

On the other hand, in step S18, when it is determined that the focus position of all the pixels has been detected, the focus position detection unit 74 notifies the end of detection of the focus position to the three-dimensional shape data generation unit 75, and the process Proceed to step S19.

In step S19, the three-dimensional shape data generation unit 75 generates three-dimensional shape data based on the focus position of each pixel stored in the memory 52, and outputs it to the subsequent stage. For example, the three-dimensional shape data is represented by the focus position of each pixel or the distance from a certain reference point to each pixel calculated based on the focus position. Then, the subsequent apparatus or processing unit of the three-dimensional shape measurement unit 51 displays a three-dimensional image of the measurement object 2 on the display 13 based on the three-dimensional shape data, for example. Thereafter, the three-dimensional shape measurement process ends.

In this way, it is possible to improve the measurement accuracy of the three-dimensional shape by calculating the focus measure more accurately and preventing the false peak of the focus measure from being detected at the focus position. Can do.

In the three-dimensional shape measurement unit 51, by adjusting the allowable value, which of the darker and brighter peaks is selected as the maximum value of the focus measure in the processes of steps S14 and S15. If there are three or more focus positions where the focus measure is a peak, the peak to be selected as the maximum value of the focus measure is set based on the luminance value of the pixel of interest at the focus position. It is possible. Thereby, when there are a plurality of focus positions at which the focus measure reaches a peak, the focus position is selected from the focus positions at which the focus measure has a peak, based on the luminance value of the target pixel at each focus position. It becomes like this.

In the above description, the focus measure is obtained by paying attention to the pixels. However, the focus measure may be obtained in units of regions composed of a plurality of pixels.

Further, in the above description, in the preprocessing of the image, after performing the Fourier transform, the low frequency component and the noise component are removed and the inverse Fourier transform is performed. For example, after performing the wavelet transform, Low frequency components and noise components may be removed and inverse wavelet transform may be performed.

Also, as in the conventional SFF method, the in-focus measure may be obtained using an effective pixel within the attention range and a Laplacian filter.

Further, in the above description, an example in which a series of processing is executed by software using the computer 12 is shown, but it is also possible to execute the processing by hardware or a computer incorporated in dedicated hardware. is there.

The program executed by the computer may be a program that is processed in time series in the order described in this specification, or in parallel or at a necessary timing such as when a call is made. It may be a program for processing.

In this specification, the term “system” means an overall apparatus composed of a plurality of apparatuses and means.

The embodiments of the present invention are not limited to the above-described embodiments, and various modifications can be made without departing from the scope of the present invention.

Claims (5)

The measurement target is calculated by calculating a focus measure indicating a degree of focus for each pixel of the plurality of images having different focus positions with respect to the measurement target, and detecting the focus position based on the focus measure. In a three-dimensional shape measuring apparatus for measuring a three-dimensional shape of an object,
At least one of luminance, hue, and saturation is used as a determination value, and the determination value of the target pixel is determined from pixels within a predetermined range in the vicinity of the target pixel used for calculation of the focus measure at the target pixel. Effective pixel extracting means for extracting effective pixels having the determination value whose difference is within an allowable range;
A focus measure calculating means for calculating the focus measure at the target pixel using the extracted effective pixels;
A three-dimensional shape measuring apparatus comprising: a focus position detecting unit that detects a focus position at which the focus measure reaches a peak as the focus position. When there are a plurality of focus positions at which the focus measure reaches a peak, the focus position detection unit, based on the luminance value of the pixel of interest at each focus position, a plurality of focus positions at which the focus measure reaches a peak The three-dimensional shape measuring apparatus according to claim 1, wherein the in-focus position is selected from the list. The image processing apparatus further comprises preprocessing means for generating a preprocessed image obtained by extracting the middle and high frequency components of the image by performing Fourier transform on the image, removing low frequency components and noise components in a predetermined frequency range, and then performing inverse Fourier transform. ,
The pixel extraction means extracts the effective pixels from the pixels within the predetermined range of the preprocessed image,
The focus measurement calculation unit calculates the focus measurement for the target pixel based on a difference in luminance value between the target pixel and the effective pixel in the preprocessed image. The three-dimensional shape measuring apparatus as described. The measurement target is calculated by calculating a focus measure indicating a degree of focus for each pixel of the plurality of images having different focus positions with respect to the measurement target, and detecting the focus position based on the focus measure. In the three-dimensional shape measuring method of the three-dimensional shape measuring apparatus for measuring the three-dimensional shape of an object,
At least one of luminance, hue, and saturation is used as a determination value, and the determination value of the target pixel is determined from pixels within a predetermined range in the vicinity of the target pixel used for calculation of the focus measure at the target pixel. An effective pixel extracting step of extracting effective pixels having the determination value whose difference is within an allowable range;
A focus measure calculating step for calculating the focus measure at the pixel of interest using the extracted effective pixels;
And a focus position detection step of detecting a focus position at which the focus measure reaches a peak as the focus position. The measurement target is calculated by calculating a focus measure indicating a degree of focus for each pixel of the plurality of images having different focus positions with respect to the measurement target, and detecting the focus position based on the focus measure. In a program for causing a computer to execute processing for measuring the three-dimensional shape of an object,
At least one of luminance, hue, and saturation is used as a determination value, and the determination value of the target pixel is determined from pixels within a predetermined range in the vicinity of the target pixel used for calculation of the focus measure at the target pixel. An effective pixel extracting step of extracting effective pixels having the determination value whose difference is within an allowable range;
A focus measure calculating step for calculating the focus measure at the pixel of interest using the extracted effective pixels;
A program that causes a computer to execute processing including a focus position detection step of detecting a focus position at which the focus measure reaches a peak as the focus position.

Images