JP2011252803A - Photographic measuring target and photographic measuring method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a photographic measuring target and a photographic measuring method that reduce troubles in removing noise due to sunlight reflection or the like and moreover make possible precise and automatic recognition of high luminance reflection of flash light at measuring points.SOLUTION: A photographic measuring target 1 used in a photographic measuring method, by which a work W is flash-photographed with a camera in a plurality of directions, subjects a plurality of pictures thereby obtained to image processing and calculates three-dimensional coordinates of measuring points incorporated into each picture, is provided with a light reflection unit 2 that reflects flash light, vividly highlights measuring points in pictures to thereby cause high luminance points to be recognized, and a color information providing unit 3 that reflects flash light at a higher luminance value than the luminance value of noise arising elsewhere than the measuring points on the work W.

Description

  The present invention relates to a photographic measurement target and a photographic measurement method used in a photographic measurement method for three-dimensionally measuring a measurement object such as a structure.

  In the above-mentioned photo measurement method, a structure (measurement object) is flash-photographed from multiple directions using a camera, and the resulting image is processed by a computer so that multiple measurements are captured in each image. This is a technique for calculating the three-dimensional coordinates of the plurality of measurement points based on the two-dimensional coordinates of the points.

  In this photo measurement method, a target that reflects flash light is affixed to these measurement points in order to sharpen the multiple measurement points captured in the image of the structure taken with flash from multiple directions. By using these photographic measurement targets, it is possible to more accurately recognize the image portions of a plurality of measurement points in the structure image.

  Conventionally, as a photographic measurement target used in the above photographic measurement method, for example, there is one described in Patent Document 1. This photo measurement target is equipped with a circular light reflector that reflects flash light from multiple directions in the irradiation direction without directivity. This photo measurement target uses flash photography to photograph structures from various directions. Even so, the luminance in the light reflecting portion is constant.

Japanese Patent Laid-Open No. 2001-249020

  However, in the above-mentioned target for photo measurement, when taking a photo under daylight conditions, all the images with high brightness in the image are recognized as measurement points during the computer processing of the image obtained by flash photography. End up. That is, all the parts where the brightness by sunlight reflection increases, such as parts other than the measurement point in the measurement object (for example, a polished surface) and parts existing around the measurement object (for example, the water surface and clouds) are used as measurement points. Misunderstood.

  As a result, the analysis work that separates the high-intensity reflection at the target for photo measurement by flash light from the high-intensity reflection at a part other than the measurement point due to disturbance such as sunlight is complicated and difficult. However, there is a problem that it cannot be said that the accuracy of recognizing high-intensity reflection on a photometric target by flash light is high, and it has been a conventional problem to solve these problems.

  The present invention has been made by paying attention to the above-described conventional problems, and reduces high-intensity reflection at a measurement point by flash light with high accuracy, while minimizing the time for removing noise due to sunlight reflection or the like. Another object of the present invention is to provide a photographic measurement target and a photographic measurement method that can be automatically recognized.

  In the invention according to claim 1 of the present invention, a measurement object is flash-photographed from a plurality of directions using a digital camera or a silver halide camera, and a plurality of images obtained thereby are image-processed and imprinted on each image. A photographic measurement target used in a photographic measurement method for calculating the three-dimensional coordinates of a measurement point, which recognizes a high-luminance point by reflecting flash light and clearly displaying the measurement point or its vicinity on the image. In the photo measurement target including the light reflecting unit to be configured, a configuration including a color information providing unit that reflects the flash light with a luminance value higher than a luminance value of noise generated in a part other than the measurement point of the measurement object; The photographic measurement target having this configuration is used as a means for solving the above-described conventional problems.

  Here, the color image expresses a color by mixing three color information of R (red), G (green), and B (blue). As shown in FIG. 4, R (red), G (green), and B (blue) each have a frame F, and each frame F stores luminance information of that color for each pixel P. Yes.

  The color of the pixel P (xn, yn) can be expressed as P (xn, yn) = R (xn, yn) + G (xn, yn) + B (xn, yn). However, R (xn, yn) is a luminance value in the (xn, yn) pixel of the R frame shown in FIG. 5, and G (xn, yn) is a luminance value in the (xn, yn) pixel of the G frame shown in FIG. B (xn, yn) is a luminance value in the (xn, yn) pixel of the B frame shown in FIG.

When one pixel P is 8 bits, each color can be expressed with 2 ⁸ = 256 levels of luminance, and thus a color image is approximately 256 (R) × 256 (G) × 256 (B). 16.77 million colors can be expressed.

  In addition to the method of mixing the three colors R (red), G (green), and B (blue), there is a method of expressing color information by luminance, hue, and saturation. A digital camera image is represented by three colors, R (red), G (green), and B (blue), and about 10 million pixels are represented by R (red), G (green), and B In order to convert from (blue) to luminance, hue, and saturation, depending on the computer's ability to perform image processing, it takes 10 to 20 seconds. We decided to select an expression method that can extract only specific color information with high speed and high accuracy by simple calculation.

  In other words, the base color frame is selected, and the luminance value of the other color frame is subtracted from the luminance value of the color frame. For example, when the R frame is used as a base, P (xn, yn) = R (xn, yn) −G (xn, yn) −B (xn, yn). However, when P (xn, yn) <0, P (xn, yn) = 0 (black).

  Since the sensitivity of R (red) information is high, it is desirable to extract R (red) information from R (xn, yn) -G (xn, yn) -B (xn, yn). As shown in FIG. 8, it is possible to obtain an image in which only the red portion is white. However, the use of B (blue) information is also worth considering depending on the later image processing method (addition of histogram adjustment processing). For comparison, FIGS. 9 and 10 show a case where the G frame is used as a base and a case where a B frame is used as a base, respectively.

  When recognizing a target that is mixed with noise outdoors, first, the image obtained by flash photography is subjected to the above processing based on the R frame. For example, the color number in the color sample of PANTONE is 811 C 2X , Orange 021 C, 485 C 2X (color with double circle in Fig. 8), Process Magenta C, 813 2X, Process Yellow C (color with circle in Fig. 8) Extract.

  In this processing, as shown in FIG. 11 and Table 1, since the luminance values of the three kinds of colors of 811 C 2X, Orange 021 C, and 485 C 2X which are orange colors are particularly high, these 811 C 2X, Orange 021 C, 485 C 2X part is extracted.

Next, a frame including the extracted part is set. There is a high possibility that one target exists in this frame. Therefore, if there is a high luminance value in a range corresponding to the frame in a normal image, that is, if there is a reflector surrounded by an orange color, it is processed as a target.
In addition, the tendency of the luminance value was the same even if the specifications of the cameras such as the digital cameras of the CCD element and the CMOS element were different.

  Therefore, in the photographic measurement target according to the present invention, the color information providing unit has a color that reflects the flash light with a high luminance value, that is, a so-called orange color when performing flash photography outdoors. It is desirable to have a configuration, and with such a configuration, noise due to sunlight reflection or the like can be reliably removed.

  As the photometric target according to the present invention, both a seal type and a sheet magnet type can be adopted.

  The present invention can be used as a measurement position recognition target placed at a measurement position set on a measurement object, and can also be used as a code target used for measurement automation by clarifying the positional relationship between a plurality of photographs. Can also be adopted.

  On the other hand, in the invention according to claim 2 of the present invention, the measurement object is obtained by photographing the measurement object in a flash using a camera from a plurality of directions, and performing image processing on the plurality of images obtained thereby and imprinting on each image. A photographic measurement method for calculating three-dimensional coordinates, wherein the photographic measurement target according to claim 1 is arranged at the measurement point or in the vicinity thereof, followed by flash photography of the measurement object from a plurality of directions. Then, extract the color information providing unit of the photographic measurement target from each of the plurality of images obtained in this way to identify the position of the photographic measurement target, and then image-process the plurality of images, The three-dimensional coordinates of the measurement point are calculated based on the high-intensity point of the light reflecting portion of the target for photo measurement that has already been specified in each image.

  In the present invention, when photo measurement of a measurement object is performed, first, a photo measurement target is placed at a plurality of measurement points set on the measurement object, and then the measurement object is flashed with a camera from a plurality of directions. Take a picture.

  In each of the plurality of images obtained by this shooting, a lot of reflected light (noise) other than the target is mixed, but the luminance value of the color information providing unit of the photo measurement target is higher than the luminance value of the noise. This color information providing unit is extracted from the inside, and the position of the target for photo measurement is specified.

  Thereafter, the plurality of images are processed to recognize the high luminance value of the light reflection portion of the specified photo measurement target, and based on this, the three-dimensional coordinates of the measurement point are calculated.

  Therefore, in the photographic measurement target and the photographic measurement method according to the present invention, it is not necessary to perform complicated and difficult work for removing noise due to sunlight reflection or the like. It can be recognized automatically with high accuracy.

  In the photographic measurement target and the photographic measurement method according to the present invention, the above-described configuration is adopted. Therefore, high-intensity reflection at the measurement point by flash light is suppressed while reducing the time for removing noise due to sunlight reflection or the like. A very excellent effect of being able to recognize automatically with high accuracy is brought about.

It is a plane explanatory view (a) which shows one example of the photographic measurement target concerning the present invention, a plane explanatory view (b) which shows other composition examples, and a plane explanatory view (c) which shows other adoption examples. It is the situation explanatory drawing (a) during the target specification of the photo measurement performed using the target for photo measurement in Fig.1 (a), and the situation explanatory drawing (b) after target specification. It is the situation explanatory drawing (a) during the target specification of the photo measurement performed using the code target in FIG.1 (c), and the situation explanatory drawing (b) after target specification. It is a schematic diagram which shows the relationship between the frame of color information, and a pixel. It is a color sample photograph image figure of R frame. It is a color sample photograph image figure of G frame. It is a color sample photograph image figure of a B frame. FIG. 6 is a color sample photographed image diagram based on an R frame. FIG. 6 is a color sample photographed image diagram based on a G frame. FIG. 5 is a color sample photographed image diagram based on a B frame. It is a graph which shows the imaging | photography result of six types of colors selected based on R (red) information.

Hereinafter, the present invention will be described with reference to the drawings.
FIG. 1A shows an example of a photo measurement target according to the present invention. In this example, the photo measurement target according to the present invention is a target for recognizing a measurement position of a workpiece (measurement object). An example will be described.

  As shown in FIG. 1 (a), this photographic measurement target 1 has a circular sheet shape of a type to be affixed to a workpiece, and has a circular light reflecting portion 2 located in the center and positions around it. A donut-shaped color information providing unit 3 is provided.

The light reflecting portion 2 is formed by fixing transparent glass beads in a dense state on the silver-white mount surface, and reflects flash light from a plurality of directions in the irradiation direction without directivity.
On the other hand, the color information providing unit 3 has a luminance value higher than the luminance value of noise (high luminance reflection at a part other than the measurement point due to disturbance such as sunlight), for example, a high luminance value in outdoor photographing using flash light. The color numbers reflected in the so-called PANTONE color swatches are painted with orange colors such as 811 C 2X, Orange 021 C, and 485 C 2X.

  Note that the overall shape of the photo measurement target 1 may be rectangular as shown in FIG. 1B in accordance with the situation of the measurement points of the workpiece.

Further, in this embodiment, the case where the photo measurement target 1 is used for measuring position recognition is shown. However, as shown in FIG. 1C, the present invention clarifies the positional relationship between a plurality of photos. It can employ | adopt as the code target 11 utilized for measurement automation.
In FIG.1 (c), the codes | symbols 12-14 are classification light reflection parts, and show that they are for identification of a measuring point itself. Reference numerals 15 to 17 denote code light reflecting portions, which indicate predetermined code numbers according to the relative positions with respect to the type light reflecting portions 12 to 14. Reference numeral 18 denotes a color information providing unit which, like the color information providing unit 3 of the photographic measurement target 1, is coated with a color that reflects with a high luminance value in outdoor shooting using flash light, a so-called orange color. It is attached.

  When performing photo measurement of a workpiece placed on a riverside using the photo measurement target 1, first, as shown in FIG. 2 (a), a plurality of measurement points set on the peripheral portion of the plate-like workpiece W are set. Subsequent to the placement of the photometry target 1, the workpiece W is photographed with a digital camera (both CCD elements and CMOS elements) from a plurality of directions.

  In each of the plurality of images obtained by this photographing, a large number of reflected light N (noise) of the river surface H behind the work W is mixed, but the luminance value of the color information providing unit 3 of the photographic measurement target 1 is greater than the luminance value of noise. Therefore, the color information providing unit 3 is extracted from each image, and the position of the photo measurement target 1 is specified.

  Thereafter, by processing a plurality of images, as shown in FIG. 2 (b), the high brightness value of the light reflecting portion 2 of the identified photo measurement target 1 is recognized, and based on this, the three-dimensional measurement point is obtained. Calculate the coordinates.

  When performing photo measurement of a workpiece using the code target 11, first, as shown in FIG. 3A, a code is placed near a plurality of measurement points set on the surface of the workpiece W joined together by welding. Subsequent to the placement of the target 11, the workpiece W is photographed with a digital camera from a plurality of directions.

In each of a plurality of images obtained by this photographing, a large number of reflected light N (noise) of the polished surface Wa and scale S by the grinder on the surface of the work W is mixed, but the luminance value of the color information providing unit 18 of the code target 11 is Since it is higher than the luminance value of noise, this color information providing unit 18 is extracted from each image, and the position of the code target 11 is specified.
Thereafter, by processing a plurality of images, as shown in FIG. 3B, the high luminance values of the light reflectors 12 to 14 for types and the light reflectors 15 to 17 for codes of the identified code target 11 are recognized. Then, the positional relationship of the code target 11 between a plurality of photographs is clarified and used for measurement automation.

  Therefore, in the photo measurement method using the photo measurement target 1 and the code target 11, the work of removing noise N due to sunlight reflection or the like is reduced, and high-intensity reflection at the measurement point by flash light is high. It can be recognized automatically with accuracy.

  The configurations of the photographic measurement target and the photographic measurement method according to the present invention are not limited to the configurations of the above-described embodiments.

1 Photo measurement target 2 Light reflection part 3, 18 Color information provision part 11 Code target (photo measurement target)
12 to 14 light reflecting parts for types 15 to 17 light reflecting parts for codes W work (measurement object)

Claims (2)

A photograph used for a photo measurement method that takes a measurement object with a camera from multiple directions and uses a camera to perform flash processing, performs image processing on multiple images, and calculates the three-dimensional coordinates of measurement points captured in each image. A measurement target,
In a photometric target having a light reflecting portion that reflects flash light and clearly displays the measurement point or its vicinity on the image to recognize a high luminance point,
A photographic measurement target comprising: a color information providing unit that reflects flash light with a luminance value higher than a luminance value of noise generated at a part other than the measurement point of the measurement object. A photo measurement method for photographing a measurement object from a plurality of directions using a camera, calculating a three-dimensional coordinate of a measurement point obtained by performing image processing on a plurality of images obtained by the image processing,
After arranging the target for photo measurement according to claim 1 at or near the measurement point,
Following the flash photographing of the measurement object from a plurality of directions, the color information providing unit of the photographic measurement target is extracted from each of the plurality of images thus obtained, and the position of the photographic measurement target is determined. Identify,
Next, the plurality of images are subjected to image processing, and the three-dimensional coordinates of the measurement points are calculated based on the high-intensity points of the light reflection part of the photographic measurement target already specified in each image. Photo measurement method.