Method and system for volume determination using a structure from motion algorithm

A construction site object volume determination system, the system comprising:

• a mobile camera for capturing images, and
• a control and evaluation unit, the control and evaluation unit having a stored program with code so as to control and execute a volume determination functionality in which using a series of images of the construction site object being captured with the mobile camera, the series comprising a plurality of images captured from different points on a path around the construction site object and with different orientations of the mobile camera, without the need of maintaining a certain orientation of the camera, as well as using given information about a known absolute reference regarding scale, wherein the volume determination functionality comprises:
• a structure from motion evaluation with a defined algorithm using the series of images is performed and a spatial representation comprising a surface of the construction site object is generated therefrom,
• the spatial representation is scaled with the given known absolute reference,
• a ground surface for the construction site object is defined, wherein the ground surface is derived by an automatically performed predefined evaluation or wherein an inquiry is made to a user to input information defining the ground surface, and is applied onto the spatial representation and
• the absolute volume of the construction site object is calculated based on the scaled spatial representation and the defined ground surface and is output,
• the system further comprising at least one of:
• an elongated reference body that defines a long axis due to its elongated shape and comprises at least two defined visually detectable markings,
• wherein at least two points are defined on the reference body by the markings, and
• wherein an absolute distance in between the at least two points and the spatial relationship of the at least two points with respect to the long axis of the reference body, being known, and
• wherein the reference body is foreseen to be placed by a user with its long axis aligned vertically on or next to the construction site object of which the volume is to be determined before capturing the series of images,
• wherein, when the series of images comprises a sub-group of images of the plurality of images, the reference body being placed on or next to the construction site object being captured in the images of the sub-group, in the functionality controlled and executed by the control and evaluation unit,
• at least two spatial positions relative to the spatial representation are determined for the at least two points also based on the structure from motion evaluation, and
• the spatial representation is:
• scaled based on a measured distance in between the determined at least two spatial positions and the known absolute distance in between the at least two points and vertically oriented based on the determined at least two spatial positions relative to the spatial representation and the known spatial relationship of the at least two points with respect to the long axis;
• and/or
• at least two defined visually detectable markings, whereby at least two points are defined, are placed on or next to the construction site object of which the volume is to be determined before moving the camera and capturing the series of images, wherein:
• for each point of the at least two points an absolute spatial position is determined with absolute reference by use of a surveying instrument,
• the series of images comprises images, in which images the markings, being placed on or next to the construction site object, are also captured,
• for each point of the at least two points a spatial position is determined also based on the structure from motion evaluation, and
• the spatial representation is scaled based on an assignment of its determined absolute spatial position to its spatial position being determined based on the structure from motion evaluation, wherein the spatial representation is also vertically oriented based on an assignment of its determined absolute spatial position to its spatial position being determined based on the structure from motion evaluation.

A non-transitory computer readable medium comprising:

• computer-executable instructions stored on the non-transitory computer readable medium that when executed perform a method for controlling and executing a functionality for determining a volume of a construction site object on a construction site, the method comprising:
• obtaining a series of images of the construction site object being captured with a camera, the series comprising a plurality of images captured from different points on a path around the construction site object and with different orientations of the camera as well as using information about a known absolute reference regarding scale,
• performing a structure from motion evaluation with a defined algorithm using the series of images and a spatial representation, comprising a surface of the construction site object is generated therefrom,
• scaling the spatial representation with the given known absolute reference,
• defining a ground surface for the construction site object, wherein the ground surface is derived by an automatically performed predefined evaluation or wherein an inquiry is made to a user to input information defining the ground surface, and is applied onto the spatial representation, and
• calculating the absolute volume of the construction site object based on the scaled spatial representation and the defined ground surface and is output,
• the method further comprising at least one of:
• an elongated reference body that defines a long axis due to its elongated shape and comprises at least two defined visually detectable markings,
• wherein at least two points are defined on the reference body by the markings, and
• wherein an absolute distance in between the at least two points where the spatial relationship of the at least two points with respect to the long axis of the reference body, being known,
• wherein the elongated reference body is placed with its long axis aligned vertically on or next to the construction site object of which the volume is to be determined before moving the camera and capturing the series of images,
• wherein the series of images comprises a sub-group of images of the plurality of images, in which images of the sub-group the reference body, being placed on or next to the construction site object, is also captured,
• wherein at least two spatial positions relative to the spatial representation are determined for the at least two points also based on the structure from motion evaluation, and
• wherein the spatial representation is scaled based on a measured distance in between the determined at least two spatial positions and the known absolute distance in between the at least two points and vertically oriented based on the determined at least two spatial positions relative to the spatial representation and the known spatial relationship of the at least two points with respect to the long axis;
• and/or
• at least two defined visually detectable markings, whereby at least two points are defined, are placed on or next to the construction site object of which the volume is to be determined before moving the camera and capturing the series of images, wherein:
• for each point of the at least two points an absolute spatial position is determined with absolute reference by use of a surveying instrument,
• the series of images comprises images, in which images the markings, being placed on or next to the construction site object, are also captured,
• for each point of the at least two points a spatial position is determined also based on the structure from motion evaluation, and
• the spatial representation is scaled and also vertically oriented based on an assignment of its determined absolute spatial position to its spatial position being determined based on the structure from motion evaluation.

A method for generating and geo-referencing a spatial representation of a construction site object to be measured, the method comprising:

• placing at least three defined visually detectable markings, whereby at least three points are defined, on or next to the construction site object,
• determining for each point of the at least three points an absolute spatial position with geo-reference by use of a surveying instrument,
• moving a mobile camera along a path around the construction site object while orienting the camera repeatedly onto the construction site object,
• in the course of this movement, capturing a series of images of the construction site object with the camera, the series comprising a plurality of images captured from different points on the path and with different orientations of the camera, without the need of maintaining a certain orientation of the camera, the plurality of images also including, for each of the at least three markings, several images where the construction site object is captured together with the respective marking, and the series being represented by an image data set collected thereby,
• performing a structure from motion evaluation with a defined algorithm using the series of images,
• generating the spatial representation of the construction site object and determining, for each point of the at least three points, a relative spatial position with relation to the spatial representation, both based on the structure from motion evaluation,
• assigning, for each point of the at least three points, its absolute spatial position to its relative spatial position and
• using the geo-reference of the absolute spatial positions together with the respectively assigned relative spatial positions and its respective relation to the spatial representation for geo-referencing the spatial representation.

A system for generating and geo-referencing a spatial representation of a construction site object to be measured, the system comprising:

• a mobile camera for capturing images,
• a control and evaluation unit, and
• at least three defined visually detectable markings, whereby at least three points are defined, wherein the visually detectable markings are foreseen to be placed by a user on or next to the construction site object and to be surveyed by a surveying instrument, whereby for each point of the at least three points an absolute spatial position with geo-reference is determined,
• wherein the control and evaluation unit has a non-transitory computer readable medium with computer-executable instructions that when executed perform a method for controlling and executing a functionality for generating and geo-referencing a spatial representation in which, using a series of images of the construction site object being captured with the camera, the series comprising a plurality of images captured from different points on a path around the construction site object and with different orientations of the camera, without the need of maintaining a certain orientation of the camera, wherein the plurality of images include, for each of the at least three markings, several images where the construction site object is captured together with the respective marking, as well as using determined absolute spatial positions for the at least three points based on surveying the at least three markings with a surveying instrument,
• a structure from motion evaluation with a defined algorithm using the series of images is performed,
• a spatial representation of the construction site object is generated and for each point of the at least three points, a relative spatial position with relation to the spatial representation is determined, both based on the structure from motion evaluation,
• for each point of the at least three points, the absolute spatial position is assigned to the respective relative spatial position, and
• the geo-reference of the absolute spatial positions together with the respectively assigned relative spatial positions and its respective relation to the spatial representation are used for geo-referencing the spatial representation.

A method for determination of an absolute volume of a construction site object on a construction site, the method comprising:

• moving a mobile camera along a path around the construction site object while orienting the camera repeatedly onto the construction site object,
• wherein the camera comprises a mobile field controller of a surveying system comprising a total station and/or GNSS-components,
• in the course of this movement, capturing a series of images of the construction site object with the camera, the series comprising a plurality of images captured from different points on the path and with different orientations of the camera, the series being represented by an image data set collected thereby,
• performing a structure from motion evaluation with a defined algorithm using the series of images and generating a spatial representation comprising a surface of the construction site object therefrom,
• scaling the spatial representation with given scaling information about a known absolute reference regarding scale,
• defining a ground surface for the construction site object and applying it onto the spatial representation and
• calculating and outputting the absolute volume of the construction site object based on the scaled spatial representation and the defined ground surface,
• the method further comprising
• determination of positions and/or changes of position and/or accelerations with absolute reference in dependency of the motion of the camera along its path around the construction site object, using
• an inertial measurement unit fixed with respect to the camera,
• a GNSS-Sensor fixed with respect to the camera,
• a tachymeter, a total station, a laser tracker or a tracking 3D-position-determining camera-system, and/or
• a stereo-camera-system or a pattern-projector-camera-system, measuring positions of the camera, and
• the spatial representation is scaled and vertically oriented based on the determined positions and/or changes of position and/or accelerations, respectively, in relation to derived relative positions of points on the path where images of the series of images have been captured by the camera, which positions and orientations are derived based on the structure from motion evaluation relative to the spatial representation.

A construction site object volume determination system, the system comprising:

• a mobile camera for capturing images,
• wherein the camera comprises a mobile field controller of a surveying system comprising a total station and/or GNSS -components, and
• a control and evaluation unit, the control and evaluation unit having a stored program with code so as to control and execute a volume determination functionality in which using a series of images of the construction site object being captured with the mobile camera, the series comprising a plurality of images captured from different points on a path around the construction site object and with different orientations of the mobile camera, without the need of maintaining a certain orientation of the camera, as well as using given information about a known absolute reference regarding scale, wherein the volume determination functionality comprises:
• a structure from motion evaluation with a defined algorithm using the series of images is performed and a spatial representation comprising a surface of the construction site object is generated therefrom,
• the spatial representation is scaled with the given known absolute reference,
• a ground surface for the construction site object is defined, wherein the ground surface is derived by an automatically performed predefined evaluation or wherein an inquiry is made to a user to input information defining the ground surface, and is applied onto the spatial representation and
• the absolute volume of the construction site object is calculated based on the scaled spatial representation and the defined ground surface and is output, the system further comprising:
• the camera comprises a built-in component of a hand-held unit, the hand-held unit further comprising a sensor for determination of positions and/or changes of position and/or accelerations with absolute reference in dependency of motion of the camera, wherein an inertial measurement unit fixed with respect to the camera and/or a GNSS-sensor fixed with respect to the camera,
• wherein in the functionality controlled and executed by the control and evaluation unit the spatial representation is scaled based on the positions and/or changes of position and/or accelerations determined by the inertial measurement unit and/or GNSS -sensor, respectively, in relation to derived relative positions of points on the path where images of the series of images have been captured by the camera, which positions are derived based on the structure from motion evaluation relative to the spatial representation, wherein the spatial representation is also vertically oriented based on the positions and/or changes of position and/or accelerations determined by the inertial measurement unit and/or GNSS-sensor, respectively, in relation to derived relative positions of points on the path where images of the series of images have been captured by the camera, which positions are derived based on the structure from motion evaluation relative to the spatial representation.