JP6770826B2 - Automatic collimation method and automatic collimation device for measuring the placement position of structures - Google Patents

Description

The present invention specifies the shape and placement position of the object to be measured.
The present invention relates to an automatic collimation method and an automatic collimation device for measuring the arrangement position of a structure . Specifically, automatic vision for measuring the placement position of a structure that performs image processing from the captured image obtained by capturing the measurement target with an imaging device, recognizes a specific color and shape, and specifies the placement position of the measurement target. Regarding quasi-methods and automatic collimation devices.

In recent years, the recognition accuracy of three-dimensional images by image recognition technology has become more sophisticated, and it has become possible to recognize from human face recognition in photographs to recognition of structures such as buildings with high accuracy. Further, as a measuring device such as a total station for measuring a distance to an object, there are models equipped with a digital camera in addition to a distance measuring means for measuring a distance using a laser beam. This digital camera captures the measurement direction and uses the image to be used in place of the telescope used for setting the measurement direction, collimation, etc. in the surveying instrument.

Therefore, the surveying device can stream not only the measurement data obtained by measuring the distance and the direction, but also the image data of the still image and the moving image taken by the digital camera. A technique of processing an image taken by a photographing device and recognizing and identifying the shape of the photographed object is widely used. For example, a technique for identifying a target area from the background of a captured image and detecting a mark figure for collimation and inclination measurement in the target area has been conventionally known (see Patent Document 1). At this time, even if the target is tilted, its color and shape can be detected and identified.

Japanese Unexamined Patent Publication No. 11-63952

However, conventional surveying instruments do not combine a measurement function and a photographing function to recognize the shape of an image obtained by photographing an object to be measured, and calculate and provide a displacement of an object in the image.
An object of the present invention is to combine the measurement function and the photographing function of a measuring device to automatically recognize the shape of an image of a measurement object and to measure the placement position of a structure for identifying the position of the shape-recognized object. A collimation method and an automatic collimation device are provided.

Another object of the present invention is to combine the measurement function and the photographing function of the measuring device to recognize the shape of the photographed image of the object to be measured, and to identify the position deviation of the shape-recognized object from the design installation position. An automatic collimation method and an automatic collimation device for measuring the arrangement position of a structure are provided.

The automatic collimation method for measuring the arrangement position of the structure according to the first aspect of the present invention is
A surveying instrument equipped with an imaging device that can swing around the horizontal and vertical axes,
An image / arithmetic processing means capable of recognizing a specific color and shape by performing image processing from the captured image captured by the imaging device.
It is an automatic collimation method for measuring the arrangement position of a structure for measuring the arrangement position of a measurement object by an automatic collimation device including a storage means for storing the setting data of the structure in advance.
A target having a preset two-dimensional outer shape is attached to one plane constituting the structure.
Wherein the image-processing means, the target recognition to identify to the by automatically image recognition of the picture taken by the imaging device, and recognizes a section in which the target is stuck, the Estimate the center position of the cross section
With the surveying instrument, the distance and orientation data at the center position are measured to obtain the measurement data.
The image / arithmetic processing means is used to compare the set data with the measurement data, calculate the positional displacement of the structure, and output the data.

The automatic collimation method for measuring the arrangement position of the structure according to the invention 2 of the present invention is described in the invention 1.
And displaying the measurement data superimposed on the setting data to the display.

The automatic collimation method for measuring the arrangement position of the structure according to the invention 3 of the present invention is described in the invention 1 or 2 .
The structure is characterized by being a three- dimensional object .

The automatic collimation device for measuring the arrangement position of the structure according to the invention 4 of the present invention.
A surveying instrument equipped with an imaging device that can swing around the horizontal and vertical axes,
An image / arithmetic processing means capable of recognizing a specific color and shape by performing image processing from the captured image captured by the imaging device.
It consists of a storage means for storing the setting data of the structure in advance, and is an automatic collimation device for measuring the arrangement position of the structure for measuring the arrangement position of the object to be measured.
The image / arithmetic processing means is
A shape setting unit that sets the shape of the target in order to detect the target attached to the structure from the captured image.
It has a target image identification unit that automatically recognizes and identifies the target based on the shape setting unit.
The image / arithmetic processing means has a function of recognizing a cross section of the structure to which the target is attached and estimating the center position of the cross section.
The surveying instrument has a function of measuring the distance and direction data of the central position and outputting the measurement data.
The image / calculation processing means has a function of comparing the setting data with the measurement data, calculating the position displacement of the structure, and outputting the data.

The automatic collimation device for measuring the arrangement position of the structure according to the invention 5 of the present invention is described in the invention 4 .
A detection color setting unit that sets a color in order to detect the target from the captured image,
It is characterized by having the target image identification unit that identifies the target based on the detection color setting unit.

The automatic collimation device for measuring the arrangement position of the structure according to the invention 6 of the present invention is described in the invention 4 or 5.
The image / arithmetic processing means is
A target image detecting means for detecting a target image from the captured image,
It is characterized by having a target image center point calculation means for calculating the center point of the target image.

The automatic collimation device for measuring the arrangement position of the structure according to the invention 7 of the present invention is described in the inventions 4 to 6 .
The structure is characterized by being a three- dimensional object .

According to the present invention, the imaging device mounted on the surveying device can recognize the object to be measured and grasp the displacement of the object to be measured from the design position.
According to the present invention, it has become possible to recognize a mark attached to a measurement object from an image taken by a photographing device mounted on a surveying instrument and specify the position thereof. This makes it possible to grasp the position of the object to be measured.

FIG. 1 is a conceptual diagram showing an operating state of the automatic collimation device 1 according to the first embodiment of the present invention. FIG. 2 is a flowchart showing an operation example of the automatic collimation device 1 according to the first embodiment of the present invention. FIG. 3 is a flowchart showing an example of an operation procedure of the automatic collimation device 1 according to the first embodiment of the present invention. FIG. 4 is a flowchart showing an example of image recognition by the automatic collimation device 1 according to the first embodiment of the present invention. FIG. 5 is a drawing showing an example of image recognition (color) by the automatic collimation device 1 according to the first embodiment of the present invention. FIG. 6 is a drawing showing an example of image recognition (color) by the automatic collimation device 1 according to the first embodiment of the present invention. FIG. 7 is a drawing showing an example of image recognition (shape) by the automatic collimation device 1 according to the first embodiment of the present invention. FIG. 8 is a diagram illustrating an example of image recognition by the first automatic collimation device 1 of the present invention and showing the positional displacement of the object to be measured.

Hereinafter, the first embodiment of the automatic collimation method and the automatic collimation device for measuring the arrangement position of the structure of the present invention will be described with reference to the drawings. FIG. 1 is a conceptual diagram showing an operating state of an automatic collimation device 1 (hereinafter, referred to as an automatic collimation device 1) for measuring the arrangement position of a structure according to the first embodiment of the present invention. The automatic collimation device 1 is a device for measuring whether or not the structure deviates from the design position. The automatic collimation device 1 includes a surveying device 2 for measuring a distance to a structure, a photographing device 3 built therein, a computer 4 for analyzing imaging data, and the like.

Specifically, the imaging device 3 photographs the structure or its direction to obtain imaging data. The computer 4 analyzes the photographed data, recognizes the image of the structure reflected in the photographed data, feeds back the direction to the measuring device 2, and the measuring device 2 measures the distance to the structure and the like. Using this measurement result, the photographing apparatus 3 finally calculates whether or not the structure is in the design position, and outputs the deviation from the design position.

A structure basically refers to any three-dimensional object having a predetermined size and installed in a space in the present embodiment. Buildings such as buildings, parts of them, and their steel members are also treated as structures. In addition, two-dimensionally printed signs and images are also used as structures.
The structure is basically installed in place in the space as designed. However, the structure may deviate from the designed installation position.

The cause may be due to changes such as underground fluctuations and environmental changes, and may be due to changes due to aging of structures and loads. In addition, when installing a structure, it may deviate from the installation position due to work mistakes, errors, malfunctions, etc. in the installation work of humans or machines. Further, the dimensions of the structure do not match the design due to a manufacturing error or the like at the time of manufacturing or construction of the structure, and as a result, when the structure is installed, the structure does not reach the design position and shifts. The deviation of the structure from the design installation position due to these causes is hereinafter referred to as the positional displacement of the structure or simply the displacement.

Hereinafter, the structure will be described by taking a rod having a quadrangular cross section as an example. In the example of FIG. 1, the structure is one end of the square bar indicated by reference numeral 10. The cross section of the square bar 10 is a rectangle, in this example, a rectangle, and the end face 10a at one end thereof is a rectangle. A plurality of square bars 10 are arranged side by side, and in a structure such as a normal building, the end faces 10a are arranged in a straight line. For example, one lower side of the end face 10a is arranged side by side on the line 11.

However, due to the installation work of the square bar 10, the installation position may be displaced from the design position, which is the position to be installed, or the installation position may be displaced for some reason after installation. In other words, the square bar 10 The position is displaced. In this example, in FIG. 1, the position-displaced one is illustrated by a square bar 10b. The square bar 10b is displaced from the design position. The design position of the square bar 10b is illustrated by the broken line square bar 10c. Therefore, the end face 10d of the square bar 10b is displaced from the end face 10a of the other square bar 10 and is displaced from the line 11.

The automatic collimation device 1 according to the first embodiment of the present invention photographs such a structure, automatically recognizes the image, performs measurement, compares the measured value with the design value, and obtains the same. It calculates and outputs the displacement. Image recognition is a first method of recognizing an image of a structure to be measured as it is or a part thereof from shooting data, and a second method of attaching a predetermined mark to the structure in advance and recognizing the mark from the shooting data. Is adopted.

The automatic collimation device 1 according to the first embodiment of the present invention can use the first method or the second method alone, but can also use both methods in combination. In the case of the second method, in the example of FIG. 1, a triangular mark 12 is adopted as a mark. As the mark 12, basically any mark such as numbers, geometric figures, images, and combinations thereof can be used. The mark can have any color, but colors such as red, yellow, and orange that are easy for the human eye to see are preferable.

For example, the mark 12 is attached to the substantially center of the cross section 10a of the plurality of square bars 10 and the cross section 10d of the square bars 10b shown in FIG. The square bar 10c shown by the broken line is installed at the original installation position (design position) of the square bar 10b. The positional displacement of the square bar 10c and the square bar 10b is the positional displacement obtained in the present invention by using the mark 12.

Therefore, the mark 12 is recognized and its position is also recognized from the photographed image of the structure, and then the structure to which it is attached is image-recognized, and finally the structure and / or the mark 12 is reached. Measure the distance and calculate the position displacement. Alternatively, the mark 12 is recognized from the image of the structure, the structure around the mark 12 and its position are image-recognized and estimated, and finally, the distance to the structure is measured and the position displacement is calculated. To do. Hereinafter, the structure of the surveying instrument 2 and the outline of its operation will be described, and then the details of image recognition and calculation of position displacement will be described.

The automatic collimation device 1 photographs the structure, recognizes the mark 12, measures the distance to the mark 12 of the structure with the measuring device 2, and analyzes the photographed data and the measured data of the computer 4. It is displayed on the display 9 as a display device. The display 9 displays the design value of the structure, the measurement data, the displacement value of the design value and the measurement data, and the like in an easy-to-see manner. For example, as shown in FIG. 1, the square bar 10c which is the design value, the square bar 10d which is the actual installation result, and the displacement value thereof are displayed. This makes it easier for site managers to see and understand the progress of construction.

While looking at the structure displayed on the display 9, the site manager proceeds with the construction so as to meet the design value or the preset reference value. As shown in FIG. 1, the surveying instrument 2 is installed in the vicinity of the structure to be measured, particularly in a place where the structure to be measured and the entire measurement target portion can be clearly seen. The surveying instrument 2 is mounted on a tripod 5 and installed. The computer 4 is installed in the vicinity of the surveying instrument 2 or brought by an operator and operated by the operator. The surveying instrument 2 and the computer 4 have a communication function for bidirectional communication with each other.

In the first embodiment of the present invention, the surveying device 2 performs wireless bidirectional communication with the computer 4, but the surveying device 2 provides a wired communication function in which the surveying device 2 and the computer 4 are connected by a communication cable or the like. May be prepared. The surveying instrument 2 includes an antenna 8 for wireless communication. The computer 4 has a built-in antenna (not shown) for wireless communication. The surveying instrument 2 communicates with the computer 4 via the antenna 8 and the built-in antenna of the computer 4.

For communication between the surveying device 2 and the computer 4, for example, a general-purpose wireless communication method such as a wireless LAN (Local Area Network), Wi-Fi (registered trademark), or Bluetooth (registered trademark) is used. In this example, a wireless LAN will be described as an example. The computer 4 can also communicate with the management center (not shown) via the built-in antenna. The management center manages the entire operation of the automatic collimation device 1 including the computer 4, the surveying instrument 2, and the like. The surveying instrument 2 generates a laser beam and irradiates the measurement object with the laser beam, detects the reflected laser beam reflected by the measurement object, and performs measurement.

The laser beam is reflected by irradiating the object with the surveying instrument 2 in a substantially linear shape, and is shown by a broken line extending from the surveying instrument 2 in FIG. 1, and the tip of the broken line reaches. An arrow is attached to indicate a point. As the surveying instrument 2, a general-purpose device such as a total station is used. As a total station, the Leica Nova TM series (for example, TM50 to TM60 models) of Leica Geosystems AG (Headquarters: Heerbrugg, Canton St. Gallen, Switzerland) can be exemplified.

This device has a built-in telescope camera with coaxial collimation. The structure of the surveying instrument 2 to be described is a general-purpose structure, and any known measuring device, measuring device, or distance measuring device can be used as long as the same function can be realized even if there are detailed differences in the structural parts. , Surveying equipment and other equipment can be used. Hereinafter, the structural outline of the surveying instrument 2, the measuring method, and the like will be briefly described, and the detailed structure and the measuring method will be omitted because they are general-purpose instruments. The surveying instrument 2 includes a laser beam (phase) detection distance calculator (not shown).

The laser light detection distance calculator includes a laser projector 6 for oscillating the projected laser light, and a two-axis rotation servo mechanism 7 for controlling the optical axis to a two-dimensional angle corresponding to the projection direction of the laser projector 6. , A laser projection angle electronic control system (not shown) that controls the laser projection angle according to a program, and a light wave phase analysis distance calculator (not shown) that detects the reflected and returned laser beam and measures the phase. Is equipped with.

The reflected laser light projected from the laser projector 6 is reflected at the measurement target point (not shown) and returned, and is received by the light wave phase analysis distance calculator. The light wave phase analysis distance calculator is based on the phase difference corresponding to the reflected laser light and the projected laser light, respectively, and the distance (reference point) between the mechanical origin (reference point) set in the surveying instrument 2 and the measurement target point ( R) is calculated and output as measurement data. Since the measurement data has three-dimensional polar coordinate data (R, θx, θy) and the distance R is measured corresponding to the two-dimensional angular coordinates (θx, θy), the distance within the three-dimensional polar coordinate data. The data R is represented as R (θx, θy).

The three-dimensional polar coordinate data (R, θx, θy) is a combination of the two-dimensional angle data (θx, θy) output from the laser projection angle electronic control system and the distance data R output from the light wave phase analysis distance calculator. It is created by the 3D polar coordinate data creator.

The surveying instrument 2 transmits the measured measurement data to the computer 4. Although not shown, the computer 4 includes a central processing unit (hereinafter referred to as a CPU), a main storage device (RAM), a ROM, a communication bus, an input device, a display 9 which is a display as an output device, and various communication interfaces. It is an electronic computer. In the first embodiment of the present invention, the main body of the computer 4 includes a CPU, RAM, ROM, a communication bus, and various communication interfaces. Further, the main body of the computer 4 has an auxiliary storage device such as an internal and / or external hard disk and a non-volatile memory.

Although not shown, this auxiliary storage device is connected to a communication bus by a predetermined interface and stores application software. The automatic collimation device 1 according to the first embodiment of the present invention has an automatic collimation management program 13 that operates on a computer 4. The automatic collimation management program 13 controls and operates a computer 4, has a function of reading design data, a function of analyzing shooting data and recognizing an image, and a function of comparing measurement results with design data to calculate position displacement. This is an application program that allows the computer 4 to realize a function of displaying the calculation result on the display 9.

The automatic collimation management program 13 is usually stored in the auxiliary storage device of the computer 4, and is automatically started by the computer 4 or manually operated by the administrator. However, the automatic collimation management program 13 may be stored and operated in another computer connected to the computer 4 via a network or a storage device on the network. In this case, the computer 4 reads the automatic collimation management program 13 from its storage location and operates it.

In the present embodiment, the automatic collimation management program 13 adopts a mode in which the automatic collimation management program 13 is stored in the auxiliary storage device of the computer 4 and operates, and the other modes are applied mutatis mutandis to this. In addition, the automatic collimation management program 13 also has a function of displaying measurement data measured by the surveying instrument 2 and a three-dimensional or two-dimensional diagram calculated from the measurement data. Further, when the computer 4 communicates with the surveying instrument 2 and / or the management center, the auxiliary storage device of the computer 4 stores communication software for controlling the communication.

These softwares are called from the basic software that controls the computer 4 and operate. Further, the main body of the computer 4 includes a built-in power supply unit (not shown) for supplying power to the computer 4. Although not shown, the input device of the computer 4 is a device such as a mouse, a keyboard, and a touch panel. The output device comprises a display 9. The built-in antenna of the computer 4 is preferably connected to the communication interface and built in the computer 4.

As described above, the computer 4 includes a main body, an input device connected to the main body, a display 9, a built-in antenna, and the like, but the computer 4 may be an integrated type in which the main body has these built-in. Computer 4 is a general-purpose computer-like operating system such as Microsoft's Windows (registered trademark), open source LINUX (registered trademark), Apple's Mac OS (registered trademark), and Google's Android (registered trademark). , Apple's iOS®, Modira Foundation's Firefox OS®, Microsoft's Windows Mobile® and other smartphones, controlled by any basic software such as operating systems for tablet computers Works.

As the computer 4, any electronic computer can be used as long as it can perform the processing described in the first embodiment of the present invention. The detailed operation of the computer 4 is omitted because it is not the gist of the present invention. Based on the measurement data transmitted from the surveying instrument 2, the situation at the site is displayed on the display 9 of the computer 4.

[Self-fixing of surveying instrument 2]
At the site, the surveying instrument 2 is installed and the coordinates of the two reference points whose coordinates are known are input to the surveying instrument 2. The computer 4 calculates, but is not limited to, the self-position where the surveying instrument 2 is installed by the backward interaction method. The backward association method is well known, and to explain the outline thereof, the coordinates of the known points are input to the computer 4, and the distances from the surveying instrument 2 to the prisms installed at the two known points by the surveying instrument 2 are set respectively. measure.

Then, from the distance formed by these two distances, the straight line connecting the surveying instrument 2 and one known point, and the angle formed by the straight line connecting the surveying instrument 2 and the other known point, the self-position where the surveying instrument 2 is installed by the computer 4 The reference point (mechanical origin) is calculated and obtained. The range of the measurement target is set in advance in the surveying instrument 2, or the operator takes the initiative in setting the range, and the surveying instrument 2 takes a picture.

[Overall flow]
FIG. 2 is a flowchart showing a procedure example of the entire operation of the automatic collimation device 1. First, the workers pre-mark the structure to be measured (steps 1 and 2). Alternatively, the mark may be installed from the beginning, for example, when the structure is constructed. The mark 12 shown in FIG. 1 is attached to the cross sections 10a and 10d of each of the square bars 10. Power is supplied to the surveying instrument 2, and the surveying instrument 2 operates (step 3).

When the surveying device 2 is activated, the surveying device 2 takes a picture of the marked structure with the photographing device 3. Alternatively, the surveying instrument 2 photographs the direction of the marked structure with the photographing device 3 (step 4). In the example of FIG. 1, the square bar 10 with the mark 12 is photographed by the photographing device 3. Then, the surveying instrument 2 transmits the photographed photographed data to the computer 4. The computer 4 analyzes and analyzes the photographed data and recognizes the mark and the structure on which the mark is placed (step 5). As a result of this recognition, the mark and the structure in which the mark is installed are identified from the photographed direction, and the data in that direction is transmitted to the surveying instrument 2.

In the example of FIG. 1, the photographed data is image-recognized, the mark 12 is recognized and specified, the cross-sections 10a and 10d to which the mark 12 is affixed are recognized, and the central position of the cross-sections 10a and 10d is estimated. The coordinates of the center position, in other words, the orientation data, are transmitted to the surveying instrument 2. When the mark 12 is attached to the center of the cross sections 10a and 10d, the center position of the cross sections 10a and 10d overlaps with the center position of the mark 12. If the mark 12 is uniformly affixed to the center of the cross sections 10a and 10d, or if the affixed position is within a range in which an error can be ignored (error range), the center position of the mark 12 is obtained and the surveying instrument is used. Send to 2.

The surveying instrument 2 acquires this direction data and measures the direction (step 6). The measurement data obtained by the measurement is the distance to the marked structure, and this measurement data is transmitted to the computer 4. In the example of FIG. 1, the measurement data is the data obtained by measuring the distance to the center position of each of the cross sections 10a and 10d. The computer 4 receives the measurement data, compares the measurement data with the design data of the structure, calculates the positional displacement of the structure to be measured, and displays these data on the display 9 (step 7). ).

In the example of FIG. 1, the measurement data of the square bar 10 and the design data are compared and displayed in an easy-to-understand manner for the operator like the broken line square bar 10c which is the design data and the solid line square bar 10b which is the measurement data. .. This can be grasped by the operator, and the position of the displaced structure can be corrected or adjusted (step 8). In the example of FIG. 1, the square bar 10b whose position is displaced and the square bar 10c which is the design position thereof are displayed at each position, and the operator can grasp the site situation by looking at this. Therefore, the operator can adjust the position of the square bar 10b. When these operations are completed, the computer 4 moves to the next process (step 9).

FIG. 3 is a flowchart showing an operation example of the surveying instrument 2 and the computer 4. Power is supplied to the surveying instrument 2, and the surveying instrument 2 operates (steps 20 and 21). When the surveying instrument 2 operates, the surveying instrument 2 performs an initialization process. For example, the self-position adjustment of the surveying instrument 2, the on-site coordinate input to the surveying instrument 2, the coordinates of the structure, and the like can be input by the above-mentioned backward meeting method (step 22). These coordinates may be transmitted from the computer 4.

The surveying instrument 2 uses the built-in photographing device 3 to take a picture in a designated direction or direction or in a designated coordinate (step 23). The surveying instrument 2 transmits the photographed data to the computer 4 (step 24). The computer 4 receives the captured data, analyzes the captured data, and recognizes and detects the target in the captured data (step 25). The target is a mark or structure whose shape and color are designated in advance, and is input to and set in the computer 4 in advance.

The computer 4 calculates the position of the target and calculates the displacement from that position (step 26). It is determined whether this displacement meets the set criteria (step 27). If the displacement does not meet the reference, an instruction to correct the imaging direction is transmitted to the surveying instrument 2, and an instruction to perform imaging again is given (step 28). As a result, the processes of steps 23 to 27 described above are repeated until the displacement criteria are satisfied. The reference here refers to the case where the center of the target is within a predetermined range. For example, it may be located within a predetermined range from the center point of the image of the shooting data.

When the displacement satisfies the reference, the computer 4 transmits an instruction to measure to the target to the surveying instrument 2, and the surveying instrument 2 performs the measurement (step 29). Then, the surveying instrument 2 transmits the measurement result to the computer 4, and the computer 4 calculates and outputs the displacement of the target using the measurement result (steps 30 and 31). The computer 4 displays the target and its displacement on the display 9, and proceeds to the next process (steps 31 and 32).

FIG. 4 is a flowchart showing an example of image recognition, and FIG. 5 is a diagram conceptually illustrating this image recognition. When the computer 4 recognizes an image, the computer 4 receives the photographed data from the surveying device 2 (steps 40 and 41). The computer 4 reads the target data (step 42). A target is a mark or a structure whose shape or color is specified in advance, and is a target of image recognition in an image. The target data is data about the shape, size, design drawing, etc. of the target.

The target data is set in the computer 4 in advance, or parameters necessary for image recognition such as images and features of the target data are input at a specific place. The computer 4 reads this and uses it for image recognition. Then, the computer 4 analyzes the captured data and identifies the target data in the photographed data (step 43). The identification method at this time will be described below.

The computer 4 deletes the figure or the like of the identified object other than the target data in the photographed data (step 44). Subsequently, when the computer 4 identifies the target data in the shooting data, the background color of the shooting data is deleted, and then the noise is deleted (steps 45 and 46). At the end, almost the target data remains, the position of the target data is calculated, and the position is corrected if necessary (step 47). In this position calculation, the direction position of the target data is calculated from the direction when the shooting data is taken and the size of the image.

In two-dimensional photography, even if the shape of an object in the photographed data is recognized, the distance to each object cannot be known. Therefore, if the shooting direction of the target data is known from the shot data, the distance to the target can be grasped by measuring the shooting direction with the measuring device 2. Here, as shown in FIG. 1, when there are a plurality of cross sections 10a to which the mark 12 is affixed, the mark 12 is recognized from the shooting data, and the cross section 10a of the shape of the object to which the mark 12 is affixed is recognized. To do. After that, as a recognition result, the direction of each cross section 10a is output.
Alternatively, the mark 12 may be recognized from the shooting data, and the direction of the mark 12 may be output as the recognition result.

(Explanation of image recognition algorithm)
Technology that analyzes and recognizes images or videos taken with digital cameras is evolving day by day. In the first embodiment of the present invention, the photographing device 3 is a digital camera capable of capturing both still images and moving images. To the human eye, a continuously changing still image is recognized as a moving image. Even for moving images, in the standardized moving image format, it is a continuous frame of still images. Therefore, when shooting with a digital camera, only the difference in the shooting scenery is shot and output. As the performance of digital cameras increases, the amount of shooting data such as shot images and shot images to be acquired tends to increase. In the shooting data, pixels as points are arranged in a two-dimensional array, and color information and gradation information (number of steps of lightness and shade) are given to each pixel.

In the case of a color image, one point is represented by three kinds of colors. In the present embodiment, the captured image adopts an arbitrary format including a compression method such as JPEG, GIF, PNG, TIFF and the like. The format of the shot video is AVI (.avi), ASF (.asf, .asx, .wmv, .wvx, .wma, .wax), MP4 (.mp4, .m4v, .m4a), Flash Video (.flv) ), And any format including its compression method. Since the present invention is not an invention for a photographed image or a photographed moving image and uses data in a general-purpose photographing method and a format based on the method, further description of the detailed structure and format of the photographed data will be omitted.

As the image recognition method, various algorithms and recognition methods are used depending on the purpose. In the present embodiment, it is assumed that a mark composed of a figure having a predetermined shape and a color and a structure having a predetermined shape are detected. In the first embodiment of the present invention, when recognizing an object having a predetermined shape and color from shooting data, the following method can be used without limitation. Noise is removed by removing outliers from the histogram using filtering such as smoothing, averaging filter, and Gaussian filter. Image recognition can be used for three-dimensional filters such as volume data and voxels.

In addition, the captured data can be extracted by repeatedly using expansion and contraction calculations to extract contours and remove small noises, lines, and points to emphasize and extract large change points. Obtain the processed image required for recognition. Further, a predetermined shape and color may be compared from the processed image, and when the coincidence rate becomes equal to or higher than the predetermined value, it may be output as a recognition result. In this way, various algorithms for image recognition can be used. Since the present invention is not an invention of the image recognition algorithm itself, detailed description thereof will be omitted.

5 to 7 are diagrams showing an example in which the shooting data shot by the shooting device 3 is analyzed by the computer 4 and the target image in the shooting data is recognized. FIG. 5 illustrates an example of identifying a mark of this color when a specific color is set as a target, for example. The computer 4 receives the shooting data including the image data 20 as shown in FIG. 5A from the surveying instrument 2. The image data 20 was taken by the photographing device 3. The image data 20 has a quadrangle in the drawing and is in a format taken by a general-purpose digital camera.

The image data 20 does not have to be a quadrangle in this way, and any format and format output by the photographing device 3 can be used. The image data 20 shown in FIG. 5A conceptually illustrates general shooting data, and in the figure, the objects are shown as a quadrangle as a first object 22 and a second object 23. Each is illustrated in a different color. The colors are illustrated here in separate hatches. Here, the quadrangle 22 indicates the target, and the quadrangle 23 indicates other objects.

The target color is preferably as basic as possible, such as red, blue, yellow, and the like. If it is a neutral color, it is difficult to distinguish it from noise in the image data, which complicates image processing. The first object 22 is illustrated with a first hatch that indicates the first color. The second object 23 is illustrated by a second hatch that indicates a second color. The image data 20 has a background 21. The background 21 is basically an image of a color or an object different from these colors such as the first color and the second color. In the figure, various symbols indicated by reference numeral 24 indicate noise.

The computer 4 analyzes the image data 20 and first deletes the object 23 having a color other than the first color of the target. Then, when the background 21 is deleted, the image data 20a as shown in FIG. 5B is obtained, and only the noise 24 and the first object 22 of the first color of the target remain. The background 21 is deleted and becomes a black background 21a here, but in the image data 20a, there is no color information and it becomes transparent. Then, when only the first color of the target is left and the others are deleted, the image data 20b as shown in FIG. 5C remains.

In this, the position of the first object 22 of the first color is specified, and the direction thereof is transmitted to the surveying instrument 2. As illustrated in FIGS. 5 (a) and 5 (c), the coordinate system is defined in the image data 20 and 20b through the center point, and usually has a horizontal axis of 25x and a vertical axis of 25y. .. Here, the target first object 22, especially its central point, is defined in this coordinate system. In the image, the horizontal axis 25x and the vertical axis 25y are defined, which are converted into polar coordinate data (θx, θy) indicating the direction to be measured by the surveying instrument 2 and output.

The computer 4 obtains a measurement direction from the converted polar coordinate data in the direction indicated by the polar coordinate data, turns and adjusts the measurement, and performs measurement. In other words, the surveying instrument 2 measures the first object 22 of the target color. The surveying instrument 2 makes a measurement and transmits the measurement data to the computer 4. The computer 4 can calculate the position displacement by grasping the position of the target first object 22 by using the measurement data and comparing it with the design data.

FIG. 6 illustrates an example in which the first object 22 is located at a position other than the center point in the image data 20. The image recognition is basically the same as that described in the description of FIG. 5, but only the different parts will be described here. When the image data 20 shown in FIG. 6A is recognized and the second object 23, the background 21, and the noise 24, which are objects other than the first object 22, are deleted, the image data 20a'shown in FIG. 6B is obtained. The first object 22 is shown at the bottom.

At this time, the computer 4 corrects the coordinates of the image so that the first object 22'is located at the center point of the image 20b' as shown in FIG. 6C, and obtains the coordinates thereof. Then, these coordinates are converted into polar coordinates to obtain the turning angle and output. FIG. 7 illustrates an example of identifying a mark of this shape when a specific shape is set as a target. The computer 4 receives the shooting data including the image data 30 as shown in FIG. 7A from the surveying instrument 2. The image data 30 is taken by the photographing device 3 in the same manner as the image data 20 described above.

Illustrated in FIG. 7A, the image data 30 illustrates the objects as the first object 32 of the first shape and the second object 33 of the second shape. Here, the objects of each shape have different colors shown by different hatching, and even in the figure, the quadrangles and the triangles are shown in different shapes. The quadrangle indicates the first object 32, and the triangle 33 indicates other objects. In the image recognition, the background 31, the second object 33, the noise 34, and the like are deleted from the image data 30, and the image data 30a remains. The background of the image data 30a has no color information and becomes transparent, but is shown in black in the figure.

Then, the position of the first object 32 is specified, and the direction is transmitted to the surveying instrument 2. The directions are converted from polar coordinate data into angle data and transmitted from the computer 4 to the surveying instrument 2. The surveying instrument 2 turns at a designated angle, adjusts the measurement direction, and performs measurement. In other words, the surveying instrument 2 measures the target first object 32. The surveying instrument 2 makes a measurement and transmits the measurement data to the computer 4. By grasping the position of the target first object 32 using the computer 4 measurement data and comparing it with the design data, the position displacement can be calculated.

In the image recognition of the image data 30 illustrated in FIG. 7A, the shape recognition may be combined with a specific color. Further, even if the target object is far from the center point in the image data, the coordinate correction can be performed in the same manner as described with reference to FIG. The description is omitted here. FIG. 8 illustrates an example of the result of image recognition. FIG. 8 shows the result of image recognition of the captured data by the computer 4. Here, the marks 41 to 45 are recognition targets, and the mark is taken as an example.

At this time, as a result of image recognition of the marks from the shooting data, the marks 41 to 45 are recognized, and when the positions are corrected and the positions are obtained, the marks are as shown in FIG. As described above, each mark 41 to 45 is image-recognized from the shooting data, the position is obtained, the direction direction data of this position is transmitted to the surveying instrument 2, and the distance to each mark 41 to 45 is measured. There is. The positional relationship of each mark 41 to 45 is calculated using the measurement data of the distance. Therefore, when the center points of the marks 41 to 45 are obtained and a line 46 passing through them is drawn as shown in the figure, the mark 44 deviates from this line.

This is displayed on the display 9. At the time of this display, highlight with color and text information. Instead of the line 46, it may be determined whether the mark 44 is within a predetermined range, such as two lines 47 having a predetermined interval. In this example, the mark 44 is not between the two lines 47. As a result, the operator adjusts the installation position of the structure to which the marks 41 to 45 are affixed while observing the positional relationship of the marks 41 to 45 displayed on the display 9.

[Other]
As described above, the mark has been described using a two-dimensional mark as an example, but a three-dimensional mark can also be used. When the mark becomes three-dimensional, it is converted so that it can be seen from each direction to create an image, which is compared with the photographed data and recognized. The image illustrated in FIG. 9A can be used as a three-dimensional mark.

1 ... Automatic collimation device 2 ... Surveying equipment 3 ... Imaging device 4 ... Computer 9 ... Display 10, 10b ... Square bars 10a, 10d ... End face 12 ... Mark 13 ... Automatic collimation management program

Claims (7)

A surveying instrument equipped with an imaging device that can swing around the horizontal and vertical axes,
An image / arithmetic processing means capable of recognizing a specific color and shape by performing image processing from the captured image captured by the imaging device.
It is an automatic collimation method for measuring the arrangement position of a structure for measuring the arrangement position of a measurement object by an automatic collimation device including a storage means for storing the setting data of the structure in advance.
A target having a preset two-dimensional outer shape is attached to one plane constituting the structure.
Wherein the image-processing means, the target recognition to identify to the by automatically image recognition of the picture taken by the imaging device, and recognizes a section in which the target is stuck, the Estimate the center position of the cross section
With the surveying instrument, the distance and orientation data at the center position are measured to obtain the measurement data.
An automatic collimation method for measuring the arrangement position of a structure, which comprises comparing the setting data with the measurement data by the image / calculation processing means, calculating and outputting the position displacement of the structure. In the automatic collimation method for measuring the arrangement position of a structure according to claim 1.
Autocollimation method for position determination of structures and displaying superimposed on the configuration data of the measurement data on the display. In the automatic collimation method for measuring the arrangement position of a structure according to claim 1 or 2.
The automatic collimation method for measuring the arrangement position of a structure, characterized in that the structure is a three-dimensional object. A surveying instrument equipped with an imaging device that can swing around the horizontal and vertical axes,
An image / arithmetic processing means capable of recognizing a specific color and shape by performing image processing from the captured image captured by the imaging device.
It consists of a storage means for storing the setting data of the structure in advance, and is an automatic collimation device for measuring the arrangement position of the structure for measuring the arrangement position of the object to be measured.
The image / arithmetic processing means is
A shape setting unit that sets the shape of the target in order to detect the target attached to the structure from the captured image.
It has a target image identification unit that automatically recognizes and identifies the target based on the shape setting unit.
The image / arithmetic processing means has a function of recognizing a cross section of the structure to which the target is attached and estimating the center position of the cross section.
The surveying instrument has a function of measuring the distance and direction data of the central position and outputting the measurement data.
The image / calculation processing means has a function of comparing the setting data with the measurement data, calculating the position displacement of the structure, and outputting the automatic collimation for measuring the arrangement position of the structure. apparatus. In the automatic collimation device for measuring the arrangement position of the structure according to claim 4.
A detection color setting unit that sets a color in order to detect the target from the captured image,
An automatic collimation device for measuring the arrangement position of a structure, which comprises the target image identification unit that identifies the target based on the detection color setting unit. In the automatic collimation device for measuring the arrangement position of the structure according to claim 4 or 5.
The image / arithmetic processing means is
A target image detecting means for detecting a target image from the captured image,
An automatic collimation device for measuring the arrangement position of a structure, which comprises a target image center point calculation means for calculating the center point of the target image. In the automatic collimation device for measuring the arrangement position of the structure according to claim 1, which is selected from claims 4 to 6.
The structure is an automatic collimation device for measuring the arrangement position of a structure, which is characterized by being a three-dimensional object.