Information processing device, correction method, and correction program

Abstract

An information processing device according to one embodiment of the present disclosure comprises a calculating unit and a correcting unit. The calculating unit calculates a point cloud of the shape of an object on the basis of the result of imaging of the object performed by a distance measuring camera. The correcting unit corrects the point cloud calculated by the calculating unit on the basis of the shape of the object estimated from the point cloud.

Description

Information processing device, correction method, and correction program

[0001] The present disclosure relates to an information processing device, a correction method, and a correction program.

[0002] Distance measuring devices that measure the distance to an object are known. For example, Patent Document 1 discloses a distance measuring device that measures the distance to an object by emitting modulated light into space and receiving light that includes light reflected from the object.

[0003] JP 2016-090268 A

[0004] However, when a point cloud representing the shape of an object such as luggage is detected using a distance measuring sensor such as a distance measuring device and the dimensions of the object are measured and verified from the point cloud, a phenomenon in which the dimensions of the object shrink is observed, making it impossible to accurately detect the dimensions of the object.

[0005] Therefore, the present disclosure proposes an information processing device, a correction method, and a correction program that can improve the detection accuracy of the dimensions of an object.

[0006] In order to solve the above problem, an information processing device according to one embodiment of the present disclosure includes a calculation unit and a correction unit. The calculation unit calculates a point cloud of a shape of an object based on an image of the object captured by a distance measuring camera. The correction unit corrects the point cloud based on the shape of the object estimated from the point cloud calculated by the calculation unit.

[0007] FIG. 1 is a diagram illustrating an example of a schematic configuration of a dimension detection device according to an embodiment. FIG. 2 is a block diagram illustrating an example of a schematic configuration of a control device according to an embodiment. FIG. 3A is a diagram illustrating an example of a phenomenon in which a point cloud contracts according to an embodiment. FIG. 3B is a diagram illustrating an example of a phenomenon in which a point cloud contracts according to an embodiment. FIG. 4A is a diagram illustrating a phenomenon in which a point cloud contracts according to an embodiment. FIG. 4B is a diagram illustrating a phenomenon in which a point cloud contracts according to an embodiment. FIG. 5A is a diagram illustrating an example of a point cloud according to an embodiment. FIG. 5B is a diagram illustrating an example of a frequency distribution of a point cloud in the height direction according to an embodiment. FIG. 6A is a diagram illustrating an example of correction of points in a point cloud according to an embodiment. FIG. 6B is a diagram illustrating an example of correction of an X coordinate according to an embodiment. FIG. 6C is a diagram illustrating an example of correction of a Y coordinate according to an embodiment. FIG. 7A is a diagram illustrating an example of a correction range according to an embodiment. FIG. 7B is a diagram illustrating an example of a correction range according to an embodiment. FIG. 8A is a diagram illustrating an example of a point cloud before correction according to an embodiment. FIG. 8B is a diagram illustrating an example of a point cloud before correction according to an embodiment. FIG. 9A is a diagram illustrating an example of a point cloud after correction according to an embodiment. FIG. 9B is a diagram illustrating an example of a point cloud after correction according to an embodiment. FIG. 10A is a diagram illustrating an example of a point cloud before correction according to an embodiment. FIG. 10B is a diagram showing an example of a point cloud before correction according to the embodiment. FIG. 11A is a diagram showing an example of a point cloud after correction according to the embodiment. FIG. 11B is a diagram showing an example of a point cloud after correction according to the embodiment. FIG. 12A is a diagram showing an example of a result of superimposing a point cloud after correction and a point cloud after correction. FIG. 12B is a diagram showing an example of a result of superimposing a point cloud after correction and a point cloud after correction. In particular, FIG. 13 is a flowchart showing an example of the flow of a dimension detection process according to the embodiment. FIG. 14 is a flowchart showing an example of the flow of a correction / detection process according to the embodiment. FIG. 15A is a diagram schematically illustrating an example of verification. FIG. 15B is a diagram schematically illustrating an example of verification. FIG. 15C is a diagram schematically illustrating an example of verification. FIG. 16 is a diagram showing an example of a change in the shape of a room due to correction of a point cloud according to the embodiment. FIG. 17 is a hardware configuration diagram showing an example of a computer that realizes the functions of a control device.

[0008] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. In the following embodiments, the same components are designated by the same reference numerals, and redundant description will be omitted.

[0009] The present disclosure will be described in the following order: 1. Embodiment 1-1. Configuration of dimension detection device according to embodiment 1-2. Configuration of control device according to embodiment 1-3. Processing flow 1-4. Verification 2. Other embodiments 3. Hardware configuration

[0010] (1. Embodiment) (1-1. Configuration of dimension detection device 10 according to embodiment) An example of a dimension detection device 10 to which the technology of the present disclosure is applied will be described using FIG. 1. FIG. 1 is a diagram showing an example of a schematic configuration of the dimension detection device 10 according to the embodiment. The dimension detection device 10 has a distance measuring camera 11 and a control device 20.

[0011] The distance measuring camera 11 is held by a stand 12 and is placed above a stage 13 with its imaging range facing downward. The stage 13 has a flat upper surface on which an object whose dimensions are to be detected is placed. In this embodiment, the object is a package 14 having a rectangular parallelepiped shape, such as a box.

[0012] The ranging camera 11 is configured to be capable of measuring the distance to an object within a shooting range. For example, the ranging camera 11 is configured as a ranging sensor employing a so-called active ranging method, which emits measuring light such as infrared light and measures the distance to the object by detecting reflected light of the emitted measuring light. The ranging camera 11 can use, for example, an indirect time of flight (iToF) method, a direct time of flight (dToF) method, or a frequency modulated continuous wave (FMCW) method. The ranging camera 11 measures the distance to an object within the shooting range by emitting measuring light into the shooting range and receiving reflected light. The ranging camera 11 outputs distance data indicating the distance to the object. For example, the ranging camera 11 includes an iToF distance image sensor capable of capturing distance images. The ranging camera 11 emits measuring light into the shooting range, detects reflected light to measure the distance, and outputs a depth image in which the measured distance is represented by a pixel value for each pixel as distance data. The distance measuring camera 11 is connected to the control device 20 .

[0013] Distance data is input to the control device 20 from the distance measuring camera 11. The control device 20 detects the dimensions of an object placed within the range of shooting by the distance measuring camera 11 based on the distance data.

[0014] (1-2. Configuration of the control device 20 according to the embodiment) Next, the configuration of the control device 20 will be described. FIG. 2 is a block diagram showing an example of a schematic configuration of the control device 20 according to the embodiment. The control device 20 is, for example, a computer such as a personal computer. In the embodiment, the control device 20 corresponds to the information processing device of the present disclosure.

[0015] 2, the control device 20 has an external I / F (interface) unit 21, a display unit 22, an input unit 23, a storage unit 24, and a control unit 25. Note that the control device 20 may have various functional units that are included in known computers in addition to the functional units shown in FIG.

[0016] The external I / F unit 21 is an interface for inputting and outputting various types of information to and from other devices. For example, the external I / F unit 21 is a communication interface such as a USB (Universal Serial Bus) port or a LAN port. The distance measuring camera 11 is connected to the external I / F unit 21. Distance data from the distance measuring camera 11 is input to the external I / F unit 21.

[0017] The display unit 22 is a display device that displays various types of information. Examples of the display unit 22 include a liquid crystal display (LCD) and a cathode ray tube (CRT). The display unit 22 displays various types of information.

[0018] The input unit 23 is an input device for inputting various types of information. For example, the input unit 23 may be an input device such as a mouse or a keyboard. The input unit 23 receives an operation input from a user and inputs operation information indicating the received operation content to the control unit 25.

[0019] The storage unit 24 is a storage device that stores various types of data. For example, the storage unit 24 is a storage device such as a hard disk, a solid state drive (SSD), or an optical disk. The storage unit 24 may also be a data-rewritable semiconductor memory such as a random access memory (RAM), a flash memory, or a non-volatile static random access memory (NVSRAM).

[0020] The storage unit 24 stores an operating system (OS) and various programs executed by the control unit 25. For example, the storage unit 24 stores various programs including a program for executing a dimension detection process and a correction / detection process, which will be described later. Furthermore, the storage unit 24 stores various data used in the processing of the programs executed by the control unit 25.

[0021] The control unit 25 is a device that controls the control device 20. The control unit 25 may be an electronic circuit such as a central processing unit (CPU) or a micro processing unit (MPU), or an integrated circuit such as an application specific integrated circuit (ASIC) or a field programmable gate array (FPGA). The control unit 25 has an internal memory for storing programs defining various processing procedures and control data, and executes various processes using these. The control unit 25 functions as various processing units when various programs are executed. For example, the control unit 25 has a reception unit 25a, a distance measurement control unit 25b, a calculation unit 25c, a correction unit 25d, a detection unit 25e, and an output unit 25f.

[0022] The receiving unit 25a receives input of various operation instructions and settings. For example, the receiving unit 25a displays an operation screen on the display unit 22 and receives input of operation instructions and settings for the operation screen from the input unit 23.

[0023] The distance measurement control unit 25b controls the distance measurement camera 11 via the external I / F unit 21. For example, the distance measurement control unit 25b controls the distance measurement camera 11 in accordance with operation instructions and settings on the operation screen, and causes the distance measurement camera 11 to capture a distance image. The distance measurement camera 11 measures the distance within the shooting range under the control of the distance measurement control unit 25b, and outputs the distance measurement image as distance data.

[0024] The calculation unit 25c calculates a point cloud of the shape of the object based on the image capture result of the object by the distance measuring camera 11. Specifically, the calculation unit 25c generates point cloud data that indicates the shape of the object as a point cloud in three-dimensional space from the distance measuring image input as distance data from the distance measuring camera 11. For example, the calculation unit 25c performs coordinate transformation to convert the position on the image of each pixel of the distance measuring image into XY coordinates and each pixel from the distance measurement into a point in three-dimensional space, calculates the three-dimensional coordinates of each point corresponding to each pixel, and generates point cloud data that indicates the shape of the object as a point cloud in three-dimensional space.

[0025] When measuring and verifying the dimensions of an object from the point cloud of point cloud data, a phenomenon of shrinkage of the object's dimensions is observed. Figures 3A and 3B are diagrams illustrating an example of the phenomenon of shrinkage of point clouds according to an embodiment. Figure 3A illustrates the results of photographing the top surfaces 15a-15c of luggage 14a-14c. The luggage 14a-14c have the same size for the top surfaces 15a-15c, but are made of materials with different colors and reflectivities. For example, luggage 14a is made of a white board with a white surface. An example of such a white board is KAPA (registered trademark) board. The luggage 14b is made of black construction paper. The luggage 14c is made of cardboard. Figure 3B enlarges the boundary between the top surfaces 15a and 15b in the point cloud data, showing the positions of each point in the point cloud at the edge of the top surfaces 15a and 15b. In Figure 3B, even though the actual dimensions are the same, the top surface 15b is shrunk more than the top surface 15a in the point cloud.

[0026] 4A and 4B are diagrams illustrating the phenomenon of point cloud shrinkage according to an embodiment. FIGS. 4A and 4B schematically illustrate an example of an image of a package 14 captured by a distance measuring camera 11. The inventors of the present application discovered that, among the points 16 in the point cloud representing the top surface 15 of the package 14, point 16a at the edge of the top surface 15 appears to shrink when it deviates from its original position on the top surface 15. In this embodiment, points 16 that deviate from their original positions, such as point 16a, are referred to as flying pixels. Flying pixels change depending on the reflection conditions, such as the material of the object (e.g., package 14) whose dimensions are to be detected and the imaging angle.

[0027] The conditions under which a point cloud shrinks are complex, making it difficult to correct the shrinkage of the point cloud using a uniform formula such as a correction coefficient or offset.

[0028] On the other hand, although the shrinkage conditions are complicated, one trend can be seen: Flying pixels exist on an extension of the position of the distance measuring camera 11 and the position where the flying pixel should originally be.

[0029] Therefore, the correction unit 25d corrects the flying pixels to positions corresponding to their original positions. For example, the correction unit 25d corrects the point cloud based on the shape of the object estimated from the point cloud calculated by the calculation unit 25c. For example, the correction unit 25d estimates the shape of the object from the point cloud of the point cloud data generated by the calculation unit 25c. For example, the correction unit 25d identifies a high distribution part from the distribution of the point cloud to estimate the shape of the object. The correction unit 25d corrects the point cloud based on the estimated shape of the object.

[0030] A specific example of the correction will be described below, taking as an example a case where the dimensions of a package 14 having a rectangular parallelepiped shape as an object to be detected.

[0031] The calculation unit 25c performs coordinate conversion of each pixel of the ranging image into a three-dimensional space to generate point cloud data that indicates the shape of the object as a point cloud in the three-dimensional space. For example, the calculation unit 25c generates point cloud data that indicates the shape of the package 14 as a point cloud in three-dimensional coordinates in which two orthogonal directions on the plane of the flat stage 13 are defined as the X and Y directions and the height relative to the plane of the stage 13 is defined as the Z axis direction.

[0032] 5A is a diagram illustrating an example of a point cloud according to an embodiment, showing an example of the height (Z-axis direction) of each point in point cloud data for a rectangular piece of luggage 14 placed on a flat stage 13.

[0033] The correction unit 25d identifies high-distribution areas from the point cloud distribution and estimates the shape of the object. For example, the correction unit 25d calculates a frequency distribution of the point cloud in the height direction of the object from the point cloud data. Then, the correction unit 25d identifies a correction candidate height for each point of the point cloud based on the calculated frequency distribution. For example, the correction unit 25d determines the height at which the frequency peaks as the correction candidate height, and identifies the closest correction candidate height for each point of the point cloud. FIG. 5B is a diagram illustrating an example of a frequency distribution of the point cloud in the height direction (Z-axis direction) according to an embodiment. FIG. 5B shows an example of a frequency distribution of the point cloud in the height direction (Z-axis direction) shown in FIG. 5A. The point cloud shown in FIG. 5A has many points located on the top surface 15 of the luggage 14 and the stage 13. Therefore, in the height direction point cloud distribution shown in FIG. 5B, a frequency peak occurs near the height of the top surface 15 of the luggage 14 and the height of the stage 13. The correction unit 25d determines the height ha of the peak 17a and the height hb of the peak 17b as the heights of the correction candidates, and specifies the heights of the correction candidates that are closest to each point of the point cloud. The correction unit 25d specifies the heights of the correction candidates that are closest to each point of the point cloud.

[0034] The correction unit 25d corrects each point in the point cloud to the height of the closest correction candidate. Here, as described above, flying pixels exist on an extension line between the position of the ranging camera 11 and the desired position of the flying pixel. Therefore, the correction unit 25d corrects each point in the point cloud to a position at a height specified on a line connecting the point and a viewpoint position corresponding to the shooting position by the ranging camera 11. FIG. 6A is a diagram showing an example of correction of points in a point cloud according to an embodiment. In FIG. 6A, the viewpoint position corresponding to the shooting position by the ranging camera 11 is shown as the origin 18. Point 16a is a flying pixel, and the height of the top surface 15 of the luggage 14 is specified as the height of the closest correction candidate. The correction unit 25d corrects point 16a to a position at the specified height of the top surface 15 of the luggage 14 on a line 19 connecting point 16a and the origin 18. For example, the correction unit 25d corrects the coordinates of point 16a to the coordinates of point 16b on the line 19, which is at the height of the top surface 15 of the luggage 14. 6B is a diagram illustrating an example of X-coordinate correction according to the embodiment. For example, if the X and Z coordinates of point 16a before correction are (dxa, dza) and the height of the correction candidate is dzb, the correction unit 25d identifies an X-coordinate dxb of the correction target on line 19 at a height of dzb. FIG. 6C is a diagram illustrating an example of Y-coordinate correction according to the embodiment. For example, if the Y and Z coordinates of point 16a before correction are (dya, dza) and the height of the correction candidate is dzb, the correction unit 25d identifies an Y-coordinate dyb of the correction target on line 19 at a height of dzb. The correction unit 25d corrects the X, Y, and Z coordinates of point 16a to (dxb, dyb, dzb).

[0035] In the above description, the correction unit 25d specifies the height of the correction candidate as the height at which the frequency peaks. However, the correction unit 25d may specify the average or median of the heights within a certain range (for example, within a half-width range) from the height at which the frequency peaks as the height of the correction candidate.

[0036] The correction unit 25d may also set a correction range for the point cloud and correct the point cloud within the set correction range. For example, the correction unit 25d may set the edge portion of the object as the correction range and correct the point cloud of the edge portion of the object based on the estimated shape of the object. For example, the correction unit 25d may set a correction range within a predetermined range from each of the four sides constituting the top surface of the object, and correct the point cloud within the correction range based on the estimated shape of the object. FIG. 7A is a diagram showing an example of the correction range according to the embodiment. FIG. 7A shows the top surface of the luggage 14. The correction unit 25d may set the correction range within ranges 30 that are a certain distance from each of the four sides constituting the top surface of the object in the X-axis and Y-axis directions, and correct the point cloud within range 30 based on the estimated shape of the object.

[0037] The correction unit 25d may set the correction range by setting a correction exclusion range. The correction unit 25d may correct the point cloud constituting the object by excluding the periphery of the object. Furthermore, when the object is placed on a plane, the correction unit 25d may correct the point cloud that is higher than a predetermined height from the plane based on the estimated shape of the object. For example, when the object is placed on the stage 13, the correction unit 25d may correct the point cloud that is higher than a predetermined height from the stage 13 based on the estimated shape of the object. Fig. 7B is a diagram showing an example of the correction range according to the embodiment. Fig. 7B shows luggage 14 placed on the stage 13. Height Z ｔｈ is set to a height lower than the upper surface 15 of the luggage 14. ｔｈ Points higher than height Z may be corrected based on the estimated object shape. ｔｈ may be set by the user. ｔｈ may be set according to the height of the upper surface 15 of the luggage 14. For example, the height Z ｔｈ may be set to a certain percentage (for example, 20%) of the highest height of the correction candidates. For example, the corrector 25d may identify the highest height of the correction candidates and set the height of a certain percentage of the highest height of the correction candidate as height Z ｔｈ You can also set it as follows.

[0038] Here, an example of the result of correcting the position of the point cloud will be described. FIGS. 8A and 8B are diagrams showing an example of a point cloud before correction according to an embodiment. FIG. 8A shows a point cloud image of a top view of the vicinity of the edge of the top surface 15 of a rectangular piece of luggage 14. FIG. 8B shows a point cloud image of a side view of the top surface 15 of the rectangular piece of luggage 14. As shown in FIG. 8A, the shape of the edge of the top surface 15 of the piece of luggage 14 is unclear due to flying pixels, and the edge shape is slanted. Furthermore, as shown in FIG. 8B, the top surface 15 of the piece of luggage 14 has noise that jumps up and down in height due to flying pixels.

[0039] 9A and 9B are diagrams showing an example of a point cloud after correction according to an embodiment. FIGS. 9A and 9B show the results of correcting the point clouds of FIGS. 8A and 8B. As shown in FIG. 9A , the correction has made the edges of the edge of the upper surface 15 of the luggage 14 clearer, and the shape of the edge has become closer to a right angle. Furthermore, as shown in FIG. 9B , the correction has made the height of the upper surface 15 of the luggage 14 uniform, and noise has been eliminated.

[0040] Another example of the correction result will be described. Figures 10A and 10B are diagrams showing an example of a point cloud before correction according to the embodiment. Figure 10A shows an image of a point cloud obtained by viewing the top surface 15 of a piece of luggage 14 from above (Top View). Figure 10B shows an image of a point cloud obtained by viewing the top surface 15 of a rectangular piece of luggage 14 from a side (Side View). As shown in Figure 10B, the top surface 15 of the piece of luggage 14 has noise that jumps up and down in height due to flying pixels.

[0041] 11A and 11B are diagrams illustrating an example of a point cloud after correction according to an embodiment. Figures 11A and 11B show the results of correcting the point clouds of Figures 10A and 10B. As shown in Figure 11B, the height of the top surface 15 of the luggage 14 has become uniform and noise has been eliminated through the correction.

[0042] 12A and 12B are diagrams showing an example of the result of overlaying the point cloud before and after correction. FIG. 12A shows a diagram in which the point cloud before correction shown in FIG. 10A and the point cloud after correction shown in FIG. 11A are overlaid. FIG. 12B shows an enlarged view of region A1, including the vicinity of the edge of FIG. 12A. The top surface 15 of the rectangular luggage 14 has become slightly larger. For example, as shown in FIG. 12B, the points of the post-correction point cloud are larger near the edge than the points of the post-correction point cloud due to the correction of flying pixels, indicating that shrinkage has been improved.

[0043] Returning to FIG. 2 , the detection unit 25e detects the dimensions of the object based on the point cloud corrected by the correction unit 25d. For example, the detection unit 25e uses AI for object detection to identify the area of ​​the object from the corrected point cloud. For example, the detection unit 25e generates a three-dimensional view of the object from the point cloud. The detection unit 25e projects the object represented by the point cloud in each of the three axes of the three-dimensional coordinate system to generate a front view, a top view, and a side view as three-dimensional views. The detection unit 25e inputs the generated three-dimensional views to AI for object detection to identify the area of ​​the object in each three-dimensional view. The detection unit 25e measures the area of ​​the identified object in the three-dimensional view to detect the dimensions of the object. For example, the detection unit 25e measures the area of ​​the identified object in the three-dimensional view to detect the width, depth, and height of the package 14. Note that the method for detecting the dimensions of the object is not limited to this. The detection unit 25e may use any detection method as long as it can detect the dimensions of the object.

[0044] The output unit 25f outputs the processing results obtained by the detection unit 25e. For example, the output unit 25f outputs the three-orthographic drawings or the detection results generated by the detection unit 25e to the display unit 22. For example, the output unit 25f outputs the dimensions of the object detected by the detection unit 25e to the display unit 22. The output unit 25f may output the data of the three-orthographic drawings or the detection results to another device.

[0045] (1-3. Processing Flow) Next, the flow of various processes performed by the dimension detection device 10 according to the embodiment will be described. First, the flow of the dimension detection process for detecting the dimensions of an object according to the embodiment will be described. FIG. 13 is a flowchart showing an example of the flow of the dimension detection process according to the embodiment. The dimension detection process is started, for example, when an object is placed on the stage 13 and a predetermined operation is performed on the operation screen displayed on the display unit 22 to instruct the start of dimension detection.

[0046] The distance measurement control unit 25b controls the distance measurement camera 11 to capture a distance image (step S10). The distance measurement camera 11 outputs the distance measurement image as distance data.

[0047] The calculation unit 25c calculates a point cloud of the shape of the object based on the image of the object captured by the distance measuring camera 11 (step S11). For example, the calculation unit 25c calculates the three-dimensional coordinates of each point corresponding to each pixel in the distance measuring image, and generates point cloud data that indicates the shape of the object as a point cloud in three-dimensional space.

[0048] The calculation unit 25c determines whether the automatic tilt correction setting is on (step S12). For example, the dimension detection device 10 allows the user to set the automatic tilt correction on or off from the operation screen. To perform automatic tilt correction, the user sets the automatic tilt correction on from the operation screen. If the automatic tilt correction setting is off (step S12: No), the process proceeds to step S14, which will be described later.

[0049] On the other hand, if the automatic tilt correction setting is on (step S12: Yes), the calculation unit 25c corrects the tilt of the three-dimensional coordinates of the point cloud data (step S13) based on the plane of the stage 13. For example, the calculation unit 25c corrects the coordinates of each point of the point cloud data to three-dimensional coordinates in which the height direction relative to the plane of the stage 13 is defined as the Z-axis direction and the plane of the stage 13 is defined as the X-axis direction and the Y-axis direction.

[0050] The calculation unit 25c sets a height in the point cloud data (step S14). For example, the dimension detection device 10 allows the user to set the height of the ranging camera 11 relative to the stage 13 from the operation screen. The user measures the height of the ranging camera 11 relative to the stage 13 and sets the height of the ranging camera 11 relative to the stage 13 from the operation screen. The calculation unit 25c corrects the coordinates of each point of the point cloud data based on the height of the ranging camera 11 relative to the stage 13. For example, the calculation unit 25c corrects the three-dimensional coordinates of each point of the point cloud data so that the set height of the ranging camera 11 relative to the stage 13 becomes the height of the position of the ranging camera 11 from a plane of the stage 13 in the point cloud data.

[0051] The detection unit 25e determines whether the three-side estimation setting is on (step S15). For example, the dimension detection device 10 can set three-side estimation on or off from the operation screen. To perform three-side estimation, the user sets three-side estimation on from the operation screen. If the three-side estimation setting is off (step S15: No), the detection unit 25e generates three-view drawings of the object from the point cloud data. For example, the detection unit 25e projects the object in each of the three axes of the three-dimensional coordinate system to generate a front view, a plan view, and a side view as three-view drawings (step S16), and then proceeds to step S23, which will be described later.

[0052] On the other hand, if the three-side estimation setting is on (step S15: Yes), the distance measurement control unit 25b determines whether the trigger photography setting is on (step S17). Trigger photography is a process of detecting dimensions from distance images captured a specified number of times of the same object. For example, the dimension detection device 10 can set trigger photography on / off and the number of images to be captured from the operation screen. To perform trigger photography, the user sets trigger photography on and specifies the number of images to be captured from the operation screen. If the trigger photography setting is off (step S17: No), the process proceeds to step S22, which will be described later.

[0053] On the other hand, if the trigger photography setting is on (step S17: Yes), the distance measurement control unit 25b determines whether photography has been performed the specified number of times (step S18). If photography has been performed the specified number of times (step S18: Yes), the process proceeds to step S24, which will be described later.

[0054] If the specified number of photographs have not been taken (step S18: No), the calculation unit 25c calculates the difference from the immediately preceding distance measurement image and extracts the change in the distance measurement image (step S19).

[0055] The ranging control unit 25b determines whether to execute the trigger for trigger photography (step S20). If there is a change in the ranging image, the ranging control unit 25b determines that the ranging image is not a ranging image of the same object and therefore cannot execute the trigger (step S20: No), and proceeds to step S24, which will be described later.

[0056] On the other hand, if there is no change in the ranging image, the ranging control unit 25b determines that the ranging image is a ranging image of the same object and that the trigger can be executed (step S20: Yes).The ranging control unit 25b waits for stability for a predetermined time (step S21) and proceeds to step S22, which will be described later.

[0057] In the correction and detection process (step S22), the point cloud is corrected and the dimensions of the object are detected. The details of the correction and detection process will be described later. When the correction and detection process is completed, the process proceeds to step S23, which will be described later.

[0058] The output unit 25f outputs the processing result by the detection unit 25e (step S23). For example, the output unit 25f outputs the three-orthographic views or the detection results generated by the detection unit 25e to the display unit 22. For example, the output unit 25f outputs the dimensions of the object detected by the detection unit 25e to the display unit 22. When the trigger photography setting is turned on and the distance measurement images are taken a specified number of times, the output unit 25f outputs the average value of the dimensions of the object detected from each of the taken distance measurement images to the display unit 22.

[0059] The distance measurement control unit 25b determines whether to end the process (step S24). If an operation to instruct the end of dimension detection is performed on the operation screen displayed on the display unit 22, the distance measurement control unit 25b determines that the process is to end (step S24: Yes), and ends the dimension detection process.

[0060] On the other hand, if the process is not completed (step S24: No), the process proceeds to step S10.

[0061] Next, the flow of the correction and detection process according to the embodiment will be described. Fig. 14 is a flowchart showing an example of the flow of the correction and detection process according to the embodiment. The correction and detection process is executed from step S22 of the dimension detection process shown in Fig. 13.

[0062] The calculation unit 25c corrects the tilt of the three-dimensional coordinates of the point cloud data with reference to the plane of the stage 13 (step S50). For example, the calculation unit 25c corrects the coordinates of each point of the point cloud data to three-dimensional coordinates in which the height direction relative to the plane of the stage 13 is defined as the Z-axis direction and the plane of the stage 13 is defined as the X-axis direction and the Y-axis direction. Note that if the rotation correction has already been performed in step S13, the processing of step S60 may be skipped.

[0063] The correction unit 25d sets a correction range for the point cloud (step S51). For example, the correction unit 25d sets an edge portion of the object as the correction range.

[0064] The correction unit 25d identifies high-frequency distribution parts from the point cloud distribution and estimates the shape of the object (step S52). For example, the correction unit 25d calculates the frequency distribution of the point cloud in the height direction of the object from the point cloud data. Then, the correction unit 25d determines the height with the high frequency as the height of the correction candidate and identifies the height of the correction candidate closest to each point in the correction range of the point cloud.

[0065] The correction unit 25d corrects each point in the correction range of the point cloud based on the estimated shape of the object (step S53). For example, the correction unit 25d corrects each point in the correction range of the point cloud to the height of the closest correction candidate. For example, the correction unit 25d corrects each point in the correction range of the point cloud to a position at a specified height on a straight line connecting the point and the viewpoint position corresponding to the shooting position by the ranging camera 11.

[0066] The detection unit 25e detects the dimensions of the object based on the corrected point cloud (step S54). For example, the detection unit 25e detects the width, depth, and height of the package 14 based on the point cloud corrected by the correction unit 25d. After the processing of step S54 is completed, the process proceeds to step S23 of the dimension detection processing shown in FIG. 13.

[0067] In this way, the dimension detection device 10 according to the embodiment corrects the flying pixels included in the point cloud to their original positions, thereby improving the detection accuracy of the dimensions of the object.

[0068] (1-4. Verification) Next, the results of detecting the dimensions of various objects using the dimension detection device 10 according to the embodiment and verifying the detected dimensions with the actual dimensions will be described.

[0069] The items (1) to (6) that were verified are explained below.

[0070] In item (1), verification was performed by changing the material (reflectivity) of the object (subject). For example, as shown in Figure 3A, a white board, black drawing paper, and cardboard packages 14a-14c were lined up, and distance-measuring images were taken by the distance-measuring camera 11 to detect the dimensions.

[0071] In item (2), verification was performed by changing the angle at which the object was placed. For example, in the scene in item (1), distance measurement images were taken with the object placed parallel and at an angle to the angle of view of the distance measurement camera 11, and the dimensions were detected.

[0072] In item (3), verification was performed by changing the shooting angle of the object. For example, in the scene in item (1), the installation position of the ranging camera 11 was shifted relative to the object, ranging images were captured, and dimensions were detected. FIG. 15A is a diagram schematically illustrating an example of verification. It is a diagram illustrating shifting the installation position of the ranging camera 11. FIG. 15A shows a case where the installation position of the ranging camera 11 is shifted. In FIG. 15A, the height of the ranging camera 11 relative to the object, i.e., the baggage 14, is kept constant, and the installation position of the ranging camera 11 is moved laterally.

[0073] In item (4), verification was performed by changing the distance (WD) between the object and the ranging camera 11. For example, ranging images were taken when the ranging camera 11 was brought closer to the object and when the object was brought closer to the ranging camera 11, and dimensions were detected. FIG. 15B is a diagram schematically illustrating an example of verification. FIG. 15B shows a case where the distance between the object and the ranging camera 11 was changed. In FIG. 15B, the height of the ranging camera 11 was changed from 2.5 m to 1.5 m, thereby changing the distance between the baggage 14 and the ranging camera 11.

[0074] In item (5), verification was performed when objects were overlapped. In item (5), verification was performed on a scene in which a small package was placed on top of another package. FIG. 15C is a diagram that schematically explains an example of verification. In FIG. 15C, a scene in which a small package 14b was placed on top of a package 14a was photographed by the distance measuring camera 11, and the dimensions were detected.

[0075] In item (6), we verified scenes where the detection accuracy was expected to deteriorate. In item (6), we used the ranging camera 11 to capture images of scenes in which the camera was angled significantly relative to the luggage 14, and then detected the dimensions. We also wrapped a transparent film around the luggage 14, captured the image with the ranging camera 11, and then detected the dimensions.

[0076] The dimension detection device 10 according to the embodiment has an improvement effect on the problem of point cloud shrinkage due to flying pixels in any of items (1) to (6). The dimension detection device 10 according to the embodiment can correct flying pixels regardless of the shooting conditions (such as differences in the material of the object, differences in the shooting angle, and differences in the distance between the object and the ranging camera 11). Therefore, the dimension detection device 10 according to the embodiment can improve the detection accuracy of the dimensions of the object.

[0077] (2. Other Embodiments) The processing according to the above-described embodiment may be implemented in various different forms other than the above-described embodiment.

[0078] In the above embodiment, the object is described as a package 14 having a rectangular parallelepiped shape, such as a box. However, the object is not limited to this. The object may have any shape. The correction unit 25d can estimate the shape of the object regardless of its shape by identifying a high distribution portion from the distribution of the point cloud. Furthermore, the control device 20 may set a representative shape of the object. The correction unit 25d identifies a high distribution portion from the distribution of the point cloud and identifies a representative shape of the object that corresponds to the shape of the high distribution portion from the set representative shape of the object. For example, the correction unit 25d identifies a representative shape of the object that is most similar to the shape of the high distribution portion. The correction unit 25d may correct each point of the point cloud to match the identified representative shape of the object.

[0079] Here, a description will be given of the results of verifying an object other than the luggage 14 using the dimension detection device 10 according to the embodiment. In the following, a case will be described in which the object is a room.

[0080] When a point cloud representing the shape of a room is calculated from a ranging image of the interior of a room captured by the ranging camera 11, the corners of the room may be distorted due to flying pixels. Generally, the walls of a room are flat. Furthermore, walls intersect perpendicularly at corners of the room. Therefore, the correction unit 25d identifies high-distribution areas from the point cloud distribution and estimates the shape of the object. For example, the correction unit 25d identifies high-distribution areas from the point cloud distribution as walls and estimates the shape by treating the walls as flat. The correction unit 25d then corrects the point cloud based on the shape of the object estimated from the point cloud. For example, the correction unit 25d corrects the point cloud at the corner of the room by treating the walls as flat. FIG. 16 is a diagram illustrating an example of a change in the shape of a room due to point cloud correction according to an embodiment. FIG. 16 illustrates a shape 40 of a corner of a room represented by a point cloud before correction and a shape 41 of a corner of a room represented by a point cloud after correction. In the corner shape 40 before correction, the wall near the corner is distorted, and the corner does not intersect perpendicularly. On the other hand, in the corner shape 40 before correction, the wall is flat and the corner intersects perpendicularly. In this way, the dimension detection device 10 according to the embodiment can improve the detection accuracy of the interior dimensions of a room. As a result, for example, when the dimension detection device 10 estimates its own position or creates an environmental map using a technique such as SLAM (simultaneous localization and mapping), the dimension detection device 10 can accurately estimate its own position and create an environmental map.

[0081] The point cloud data in the above embodiment has been described with reference to an example in which the viewpoint position corresponding to the shooting position by the ranging camera 11 is used as the origin. However, the point cloud data is not limited to this. The origin may be any coordinate position. The point cloud data may also include the coordinate position of the viewpoint position corresponding to the shooting position by the ranging camera 11. In this case, the correction unit 25d corrects each point in the point cloud data to a position at the height of the closest correction candidate on a line connecting each point in the point cloud and the viewpoint position.

[0082] Furthermore, the dimension detection device 10 according to the above embodiment has been described with reference to an example in which the control device 20 is a computer such as a personal computer, and the distance measuring camera 11 is connected to the control device 20. However, the present disclosure is not limited to this. The control device 20 may be any device capable of implementing the processes described in the embodiment. The control device 20 may be implemented by a computer such as a server device or a cloud system. For example, the control device 20 may be connected to the distance measuring camera 11 via a network, and distance measuring image data may be input from the distance measuring camera 11 via the network. The control device 20 may perform the dimension detection process and correction / detection process according to the embodiment on the input distance measuring image. Furthermore, the control device 20 may be configured integrally with the distance measuring camera 11. For example, the functions of the control device 20 may be implemented by one or more integrated circuits (ICs) included in the distance measuring camera 11.

[0083] In the above embodiment, the dimension detection device 10 calculates a point cloud from the image capture results of the object captured by the distance measuring camera 11, corrects the calculated point cloud, and detects the dimensions of the object based on the corrected point cloud. However, the present disclosure is not limited to this. The control device 20 may store point cloud data of the generated point cloud in the storage unit 24, correct the point cloud of the point cloud data stored in the storage unit 24, and detect the dimensions of the object based on the corrected point cloud.

[0084] Among the processes described in the above embodiments of the present disclosure, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically using known methods. In addition, the process procedures, specific names, and information including various data and parameters shown in the above documents and drawings can be changed as desired unless otherwise specified. For example, the various information shown in each drawing is not limited to the illustrated information.

[0085] Furthermore, the components of each device shown in the figure are conceptual functional components and do not necessarily have to be physically configured as shown in the figure. In other words, the specific form of distribution and integration of each device is not limited to that shown in the figure, and all or part of them can be functionally or physically distributed and integrated in any unit depending on various loads, usage conditions, etc.

[0086] Furthermore, the above-described embodiments of the present disclosure can be combined as appropriate within the scope of the processing content without causing inconsistencies. Furthermore, the order of the steps shown in the sequence diagrams or flowcharts of the present embodiments can be changed as appropriate. For example, the steps may be processed in chronological order, repeatedly, or partially in parallel.

[0087] Furthermore, the effects described in this specification are merely examples and are not limiting, and other effects may also be present.

[0088] As described above, the information processing device (control device 20 in the embodiment) according to the present disclosure includes a calculation unit (calculation unit 25c) and a correction unit (correction unit 25d). The calculation unit calculates a point cloud of the shape of the object (luggage 14) based on the image capture results of the object (luggage 14) by the ranging camera (ranging camera 11). The correction unit corrects the point cloud based on the shape of the object estimated from the point cloud calculated by the calculation unit. This allows the information processing device according to the present disclosure to improve the detection accuracy of the dimensions of the object.

[0089] The target object is an object (luggage 14) or a room. The calculation unit calculates a point cloud of the shape of the object or room based on the image of the object or room captured by the distance measuring camera. The correction unit corrects the point cloud of the edge of the object or the corner of the room. This allows the information processing device according to the present disclosure to improve the detection accuracy of the dimensions of the object or room.

[0090] The correction unit corrects the point cloud based on the shape of the object estimated from the distribution of the point cloud, thereby enabling the information processing device according to the present disclosure to correct the point cloud to match the shape of the object.

[0091] The correction unit also identifies high-distribution portions from the distribution of the point cloud to estimate the shape of the object, thereby enabling the information processing device according to the present disclosure to accurately estimate the shape of the object.

[0092] The object is placed on a plane. The correction unit calculates a frequency distribution of the point cloud of the object in a height direction from the plane, identifies a correction candidate height for each point of the point cloud based on the calculated frequency distribution, and corrects each point of the point cloud to a position at the identified correction candidate height on a line connecting the point and a viewpoint position corresponding to the shooting position of the ranging camera. This allows the information processing device according to the present disclosure to accurately correct the shape of the object in the height direction from the plane.

[0093] The correction unit determines the height of the correction candidate that is closest to each point of the point cloud, taking the height at which the frequency peaks in the frequency distribution as the height of the correction candidate, and corrects each point of the point cloud to a position on a straight line that corresponds to the height of the specified closest correction candidate. This enables the information processing device according to the present disclosure to accurately correct the shape of the object in the height direction from a plane.

[0094] The correction unit corrects the point cloud within the set correction range based on the estimated shape of the object. This allows the information processing device according to the present disclosure to correct only the point cloud within the correction range. Furthermore, since the information processing device according to the present disclosure corrects only the point cloud within the correction range, the correction processing load can be reduced.

[0095] The object is assumed to have a rectangular parallelepiped shape. The correction unit corrects the point cloud of the edge portion of the object based on the estimated shape of the object. This allows the information processing device according to the present disclosure to correct the shape of the edge portion of the object.

[0096] The correction unit also corrects the point clouds within a predetermined range from each of the four sides constituting the top surface of the object based on the estimated shape of the object, thereby enabling the information processing device according to the present disclosure to correct the shape of the four sides constituting the top surface of the object.

[0097] The correction unit also corrects the point cloud that is higher than a predetermined height from the plane based on the estimated shape of the object, thereby enabling the information processing device according to the present disclosure to correct the shape of the part of the object that is placed on the plane and that is higher than the predetermined height from the plane.

[0098] The predetermined height is set according to the height of the upper surface of the object, thereby enabling the information processing device according to the present disclosure to set the predetermined height according to the upper surface of the object.

[0099] The correction unit also identifies the highest correction candidate among the heights of the correction candidates and sets a certain percentage of the height of the highest correction candidate as the predetermined height, thereby enabling the information processing device according to the present disclosure to set the predetermined height according to the height of the highest correction candidate.

[0100] The correction unit also identifies a high distribution portion from the distribution of the point cloud, identifies a representative shape of the target object that corresponds to the shape of the high distribution portion from the representative shapes of the set target object, and corrects each point of the point cloud to match the identified representative shape of the target object. This allows the information processing device according to the present disclosure to correct each point of the point cloud to match the representative shape of the set target object even if the target object is unknown.

[0101] The information processing device according to the present disclosure further includes a detection unit (detection unit 25e). The detection unit detects the dimensions of the object based on the point cloud corrected by the correction unit. This allows the information processing device according to the present disclosure to accurately detect the dimensions of the object.

[0102] (3. Hardware Configuration) The control device 20 and the like according to the embodiments of the present disclosure described above are realized by, for example, a computer 1000 configured as shown in FIG. 17 . The control device 20 will be described as an example. FIG. 17 is a hardware configuration diagram showing an example of a computer 1000 that realizes the functions of the control device 20. The computer 1000 has a processing circuitry 1100, a RAM 1200, a ROM 1300, a secondary storage device 1400, a communication interface 1500, an input / output interface 1600, and a display unit 1700. The various units of the computer 1000 are connected by a bus 1050.

[0103] The processing circuit 1100 operates and controls each unit based on programs stored in the ROM 1300 or the secondary storage device 1400. For example, the processing circuit 1100 loads the programs stored in the ROM 1300 or the secondary storage device 1400 into the RAM 1200 and executes processing corresponding to the various programs.

[0104] The ROM 1300 stores boot programs such as a Basic Input Output System (BIOS) that is executed by the processing circuit 1100 when the computer 1000 is started up, and programs that depend on the hardware of the computer 1000 .

[0105] The secondary storage device 1400 is a computer-readable recording medium that non-temporarily records programs executed by the processing circuit 1100 and data used by such programs. Specifically, the secondary storage device 1400 is a recording medium that records programs for each process of the control device 20 according to an embodiment of the present disclosure, which are examples of program data 1450.

[0106] The communication interface 1500 is an interface for connecting the computer 1000 to an external network 1550. For example, the processing circuit 1100 receives data from other devices and transmits data generated by the processing circuit 1100 to other devices via the communication interface 1500.

[0107] The input / output interface 1600 is an interface for connecting the input / output device 1650 and the computer 1000. The input / output interface 1600 corresponds to the external I / F unit 21 provided in the control device 20. For example, the processing circuit 1100 receives data from an input device such as a touch panel via the input / output interface 1600. The processing circuit 1100 also transmits data to an output device such as a display unit 1700 via the input / output interface 1600. The input / output interface 1600 may also function as a media interface that reads programs and the like recorded on a predetermined recording medium. Examples of the media include optical recording media such as DVDs (Digital Versatile Discs) and PDs (Phase Change Rewritable Discs), magneto-optical recording media such as MOs (Magneto-Optical Disks), tape media, magnetic recording media, and semiconductor memories.

[0108] The display unit 1700 is an interface for displaying information processed by the computer 1000. The display unit 1700 is, for example, a liquid crystal display or an organic electroluminescence display (EL display). The display unit 1700 may also be a touch panel display device or a video projection device.

[0109] The components of the computer 1000 are connected by a bus 1050. Each interface does not necessarily have to be provided inside the computer 1000, but may be provided outside the computer 1000 via a network or the like. Furthermore, the components constituting the computer 1000 may be controlled by a circuit different from the processing circuit 1100. For example, the display unit 1700 may be controlled not by the processing circuit 1100 but by a circuit dedicated to display processing provided in the display unit 1700.

[0110] For example, when the computer 1000 functions as the control device 20 according to an embodiment of the present disclosure, the processing circuit 1100 of the computer 1000 functions as the control unit 130 by executing a program loaded onto the RAM 1200. The secondary storage device 1400 stores a correction program according to the present disclosure and various data stored in the storage device 120. The processing circuit 1100 reads and executes program data 1450 from the secondary storage device 1400. Alternatively, the processing circuit 1100 may obtain these programs from another device via an external network 1550. That is, the secondary storage device 1400 need not be located inside the computer 1000, but may also be located outside the computer 1000. The processing circuit 1100 is an example of an integrated circuit, and a CPU, an MPU, a GPU, an APU, an ASIC, and an FPGA can all be considered integrated circuits.

[0111] The present technology can also be configured as follows. (1) An information processing device having: a calculation unit that calculates a point cloud of a shape of an object based on an image of the object captured by a distance measuring camera; and a correction unit that corrects the point cloud based on the shape of the object estimated from the point cloud calculated by the calculation unit. (2) The information processing device described in (1), wherein the object is an object or a room, the calculation unit calculates a point cloud of the shape of the object or the room based on an image of the object or the room captured by the distance measuring camera, and the correction unit corrects a point cloud of an edge of the object or a corner of the room in the point cloud. (3) The information processing device described in (1) or (2), wherein the correction unit corrects the point cloud based on the shape of the object estimated from the distribution of the point cloud. (4) The information processing device described in (3), wherein the correction unit identifies a high distribution portion from the distribution of the point cloud to estimate the shape of the object. (5) The information processing device according to any one of (1) to (4), wherein the object is placed on a plane, and the correction unit calculates a frequency distribution of a point cloud of the object in a height direction from the plane, specifies a height of a correction candidate for each point of the point cloud based on the calculated frequency distribution, and corrects each point of the point cloud to a position where the height of the correction candidate is specified on a straight line connecting the point and a viewpoint position corresponding to the shooting position of the ranging camera. (6) The correction unit specifies a height at which the frequency peaks in the frequency distribution as a height of the correction candidate, specifies a height of the closest correction candidate for each point of the point cloud, and corrects each point of the point cloud to a position where the height of the closest correction candidate is specified on the straight line. (7) The information processing device according to any one of (1) to (6), wherein the correction unit corrects the point cloud within a set correction range based on an estimated shape of the object. (8) The information processing device according to any one of (1) to (7), wherein the object has a rectangular parallelepiped shape, and the correction unit corrects a point cloud of an edge portion of the object based on an estimated shape of the object.(9) The information processing device according to (8), wherein the correction unit corrects a point cloud within a predetermined range from each of four sides constituting the upper surface of the object based on an estimated shape of the object. (10) The information processing device according to (5), wherein the correction unit corrects a point cloud that is higher than a predetermined height from the plane based on an estimated shape of the object. (11) The information processing device according to (10), wherein the predetermined height is set according to the height of the upper surface of the object. (12) The information processing device according to (10), wherein the correction unit identifies the highest correction candidate height among the heights of the correction candidates, and sets a height that is a certain percentage of the height of the highest correction candidate as the predetermined height. (13) The information processing device according to any one of (1) to (12), wherein the correction unit identifies a high distribution portion from the distribution of the point cloud, identifies a representative shape of the object that corresponds to the shape of the high distribution portion from a set representative shape of the object, and corrects each point of the point cloud to match the identified representative shape of the object. (14) The information processing device according to any one of (1) to (13), further comprising a detection unit that detects dimensions of the object based on the point cloud corrected by the correction unit. (15) A correction method including, by a computer, calculating a point cloud of the shape of the object based on an image of the object captured by a distance measuring camera, and correcting the point cloud based on the shape of the object estimated from the calculated point cloud. (16) A correction program that causes a computer to function as: a calculation unit that calculates the point cloud of the shape of the object based on an image of the object captured by a distance measuring camera, and a correction unit that corrects the point cloud based on the shape of the object estimated from the point cloud calculated by the calculation unit.

[0112] REFERENCE SIGNS LIST 10 Dimension detection device 11 Distance measuring camera 12 Stand 13 Stage 20 Control device 21 External I / F section 22 Display section 23 Input section 24 Storage section 25 Control section 25a Reception section 25b Distance measurement control section 25c Calculation section 25d Correction section 25e Detection section 25f Output section

Claims

1. An information processing device having: a calculation unit that calculates a point cloud of the shape of an object based on the image capture results of the object by a distance measuring camera; and a correction unit that corrects the point cloud based on the shape of the object estimated from the point cloud calculated by the calculation unit.

2. The information processing device according to claim 1, wherein the target object is an object or a room, the calculation unit calculates a point cloud of the shape of the object or the room based on the image capture results of the object or the room by the ranging camera, and the correction unit corrects the point cloud of an edge of the object or a corner of the room.

3. The information processing device according to claim 1, wherein the correction unit corrects the point cloud based on a shape of the object estimated from a distribution of the point cloud.

4. The information processing device according to claim 3, wherein the correction unit estimates the shape of the target object by identifying a portion with a high distribution from the distribution of the point cloud.

5. The information processing device according to claim 1, wherein the object is placed on a plane, and the correction unit calculates a frequency distribution of the point cloud of the object in a height direction from the plane, identifies a correction candidate height for each point of the point cloud based on the calculated frequency distribution, and corrects each point of the point cloud to a position that is the height of the identified correction candidate on a straight line connecting the point and a viewpoint position corresponding to the shooting position by the ranging camera.

6. The information processing device according to claim 5, wherein the correction unit determines the height of a correction candidate to be the height at which the frequency peaks in a frequency distribution, identifies the height of the closest correction candidate for each point of the point cloud, and corrects each point of the point cloud to a position on the straight line that corresponds to the height of the closest correction candidate identified.

7. The information processing device according to claim 1, wherein the correction unit corrects the point cloud within the set correction range based on the estimated shape of the target object.

8. The information processing device according to claim 1, wherein the object has a rectangular parallelepiped shape, and the correction unit corrects the point cloud of an edge portion of the object based on an estimated shape of the object.

9. The information processing device according to claim 8, wherein the correction unit corrects a group of points within a predetermined range from each of the four sides constituting the top surface of the object based on an estimated shape of the object.

10. The information processing device according to claim 5, wherein the correction unit corrects a point group that is higher than a predetermined height from the plane based on an estimated shape of the target object.

11. The information processing device according to claim 10, wherein the predetermined height is set according to the height of the upper surface of the object.

12. The information processing device according to claim 10, wherein the correction unit identifies the highest correction candidate among the heights of the correction candidates, and sets a certain percentage of the height of the highest correction candidate as the predetermined height.

13. The information processing device according to claim 1, wherein the correction unit identifies a high distribution part from the distribution of the point cloud, identifies a representative shape of the object that corresponds to the shape of the high distribution part from the set representative shape of the object, and corrects each point of the point cloud to match the identified representative shape of the object.

14. The information processing device according to claim 1, further comprising a detection unit that detects dimensions of the object based on the point cloud corrected by the correction unit.

15. A correction method comprising: a computer calculating a point cloud of the shape of an object based on the image of the object captured by a distance measuring camera; and correcting the point cloud based on the shape of the object estimated from the calculated point cloud.

16. A correction program for causing a computer to function as: a calculation unit that calculates a point cloud of the shape of an object based on the image of the object captured by a distance measuring camera; and a correction unit that corrects the point cloud based on the shape of the object estimated from the point cloud calculated by the calculation unit.

Images