KR101428467B1 - Method and apparatus for automated updating 4d cad construction model - Google Patents

Abstract

The method of generating a 4D CAD construction process model according to the present invention includes the steps of providing a construction plan, a 3D CAD model, and a schedule schedule information, respectively, and image masks of unit components constituting a 3D CAD model from the 3D CAD model, A step of generating a scene processing image from which noises and construction equipment interference are removed from a shot image obtained from an on-site surveillance camera shooting a construction site in the same position and direction as the projection time for generating an image mask Extracting image masks matching the on-scene processed image, identifying the unit operations progressed by the matched image masks, identifying the unit work in the next process order from the point in time when it is determined that the image masks for the particular unit member are matched If it is determined that the image mask is matched Updating the progress schedule information by specifying a task time period of the specific unit member as a time period up to a point and matching a specified task time period with an identified unit task, And updating the 4D CAD construction process model by adding the elementary work items of the 3D CAD model to the immediately preceding 4D CAD construction process model.

Description

[0001] METHOD AND APPARATUS FOR AUTOMATED UPDATING [0002] 4D CAD CONSTRUCTION MODEL [

The present invention relates to a construction process management technique, and more particularly, to a method of updating a 4D CAD construction process model.

4D (4-dimensional) CAD (Computer-Aided Design) is based on 3D CAD and combines the components of the 3D CAD model according to the process schedule so that the 3D CAD model is completed over time It is a tool to assist in effective construction management by visualizing the shape.

4D CAD can visualize construction projects in a wide range of management levels (eg project level, operational level, construction work level) to identify the presence of problems in relation to construction design and planning and further identify problems. have.

For example, a 4D CAD model with a construction sequence allows the designer to perform disturbance elements, workspace analysis, and schedule evaluation prior to actual construction. As 4D CAD improves understanding of construction projects, designers can communicate more effectively with construction personnel.

Despite these anticipated advantages, the development of conventional 4D CAD models is inefficient and has not been widely utilized in the construction industry.

Typically, the process of developing a 4D CAD model begins with the creation of a 3D CAD model, followed by the construction schedule development and the relationship between the 3D CAD model. Techniques for creating 3D CAD models have been greatly improved in recent years, but construction scheduling and scheduling-model relationship generation work still have room for improvement.

Scheduling work involves regular updates to the basic construction schedule, which must be modified in accordance with the latest progress in the construction air. In the past, however, this modification takes a lot of time and labor because a large part of data collection depends on manual and observation.

In addition, the scheduling-model relationship generation task typically requires a lot of time and labor because the user must individually match each component of the 3D CAD model with detailed schedule information.

These hand-based work processes are fundamentally vulnerable to errors and even more so in large-scale construction projects. Therefore, more automated and efficient 4D CAD model development and update techniques are needed.

In response to these problems, there have been recent attempts to automate process monitoring on construction sites using sensing techniques such as radio frequency identification (RFID), 3D laser scanning, and the like.

It is an object of the present invention to provide a method and apparatus for updating an automated 4D CAD construction process model based on image processing.

A method of generating a 4D CAD construction process model according to an aspect of the present invention includes:

A 3D CAD model, and a schedule schedule information, respectively;

Generating image masks of unit components constituting a 3D CAD model from the 3D CAD model for each unit operation;

Generating a field processed image from which noises and construction equipment interference are removed from a captured image obtained from a field surveillance camera capturing a construction site in the same position and direction as the projection time for generating the image mask;

Extracting image masks that match the on-site processed image, and identifying unit work performed by the matched image masks;

The time period from the time when it is determined that the image masks relating to the specific unit member are matched to the time when it is determined that the image masks relating to the unit members in the next process order are matched, Updating the progress schedule information by matching the time period and the identified unit work; And

And updating the 4D CAD construction process model by adding, to the immediately preceding 4D CAD construction process model, the operation time frame of the updated progress schedule information and the member elements per unit work of the 3D CAD model, based on the identified unit work can do.

According to one embodiment, the image mask may be a projected image of a unit element that is installed by a unit operation included in the 3D CAD model.

According to one embodiment, the projection view for generating the image mask may include coordinates in a three-dimensional virtual space of the 3D CAD model corresponding to the position, angle of view, focal length, aperture and direction of the field surveillance camera in the real world And angles.

According to one embodiment, each of the unit element elements is given a unique member identification number, and each of the image masks may be given a mask identification number corresponding to each unit element element.

According to one embodiment, an ongoing unit job may be identified from the identification number of the image mask corresponding to the last of the matched image masks.

According to one embodiment, the step of generating the on-

In the first binary image acquired by binary conversion so that the captured image of the RGB color space has two pixel values of 0 and 1, the color value (H) image of the captured image of the HSV color space converted from the captured image of the RGB color space And subtracting the second binary image obtained by binary conversion so as to have two pixel values of 0 and 1, thereby removing the construction equipment interference from the photographed image.

According to one embodiment, the step of generating the on-

The method may further include the step of removing noise caused by construction equipment interference or included in the shot image by using the noise removal filtering technique and the morphing technique including the closing and enlarging operation, respectively.

According to one embodiment, the step of generating the on-

And removing the background from the on-scene processing image.

According to one embodiment, updating the progress schedule information comprises:

Summing area values of mask areas of the matched image masks;

Specifying a time period in which the summed mask area area values are discontinuously different from a previous value as a start time point of a previous unit work and a start time of a new unit work; And

And updating the progress schedule information by matching the specified work time period with the identified unit work.

A computer-readable recording medium according to another aspect of the present invention may be a computer-readable recording medium on which a program for implementing a 4D CAD construction process model generation method according to an embodiment is recorded.

According to another aspect of the present invention, there is provided an apparatus for updating a 4D CAD construction process model,

A design information database for storing construction plans, 3D CAD models and planning schedule information, respectively;

An image mask generation unit for generating image masks for unit components of the 3D CAD model from the 3D CAD model for each unit operation;

A field processing image generation unit for generating a field processing image from which noises and construction equipment interference are removed from a photographed image obtained from a field surveillance camera photographing a construction site in the same position and direction as the projection time for generating the image mask;

A unit job identification unit for extracting image masks matching the on-site processed image, and identifying unit jobs processed by the matched image masks;

The time period from the time when it is determined that the image masks relating to the specific unit member are matched to the time when it is determined that the image masks relating to the unit members in the next process order are matched, A progress schedule information updating unit for updating the progress schedule information by matching the time period and the identified unit work; And

A 4D CAD model for updating the 4D CAD construction process model by adding the operation time frame of the updated progress schedule information and the member elements for each unit operation of the 3D CAD model to the immediately preceding 4D CAD construction process model based on the identified unit work, And an update unit.

According to one embodiment, the image mask may be a projected image of a unit element that is installed by a unit operation included in the 3D CAD model.

According to one embodiment, the projection time for creating the image mask is determined by the coordinates and angles in the three-dimensional virtual space of the 3D CAD model corresponding to the position, angle of view, focal length, aperture and direction of the field surveillance camera in the real world Lt; / RTI >

According to one embodiment, each of the unit element elements is given a unique member identification number, and each of the image masks may be given a mask identification number corresponding to each unit element element.

According to one embodiment, an ongoing unit job may be identified from the identification number of the image mask corresponding to the last of the matched image masks.

According to one embodiment, the on-site processed image generating unit

In the first binary image acquired by binary conversion so that the captured image of the RGB color space has two pixel values of 0 and 1, the color value (H) image of the captured image of the HSV color space converted from the captured image of the RGB color space And subtracting the second binary image obtained by binary conversion so as to have two pixel values of 0 and 1, thereby removing the construction equipment interference from the photographed image.

According to one embodiment, the on-site processed image generating unit

The noise removing unit may further include a noise removing unit included in the shot image by using the noise removing filtering technique and the morphing technique including the closing and enlarging operations, respectively, or removing the noise caused by construction equipment interference.

According to one embodiment, the on-site processed image generating unit

And a background removal unit removing the background from the on-scene processing image.

According to one embodiment, the progress schedule information updating unit updates the progress schedule information,

Summing the area values of the mask areas of the matched image masks,

The time point at which the summed mask area area values discontinuously differ from the previous value is defined as the end time point of the previous unit work and the start time point of the new unit work,

And to update the progress schedule information by matching the specified task period and the identified unit task.

A computer-readable recording medium according to another aspect of the present invention may be a recording medium on which a program for operating a computer as a 4D CAD construction process model generating apparatus according to an embodiment is recorded.

According to the method and apparatus for updating the automated 4D CAD construction process model of the present invention, by applying the image masks of the members extracted from the 3D CAD model to the on-site processing image obtained through the simple image processing on the photographed image of the construction site, It is possible to identify the actually processed process and automatically construct the actual schedule information.

According to the method and apparatus for updating the automated 4D CAD construction process model of the present invention, the 4D CAD model can be generated by combining planned schedule information of a specific member and actual schedule information with a 3D CAD model.

1 is a flowchart illustrating an automated 4D CAD construction process model update method according to an embodiment of the present invention.
FIG. 2 is a flowchart specifically illustrating steps for identifying an advanced process using a field process image and image masks, among methods of updating an automated 4D CAD construction process model according to an exemplary embodiment of the present invention.
FIG. 3 is a conceptual diagram illustrating a step of identifying an advanced process by applying image masks according to a process sequence of specific members to an on-scene image in a method for updating an automated 4D CAD construction process model according to an exemplary embodiment of the present invention.
FIG. 4 is a graph illustrating a step of updating actual schedule information based on a pixel area value of image masks in a method of updating an automated 4D CAD construction process model according to an exemplary embodiment of the present invention.
5 is a block diagram illustrating an automated 4D CAD construction process model update apparatus in accordance with an embodiment of the present invention.

For the embodiments of the invention disclosed herein, specific structural and functional descriptions are set forth for the purpose of describing an embodiment of the invention only, and it is to be understood that the embodiments of the invention may be practiced in various forms, The present invention should not be construed as limited to the embodiments described in Figs.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. The same reference numerals are used for the same constituent elements in the drawings and redundant explanations for the same constituent elements are omitted.

1 is a flowchart illustrating an automated 4D CAD construction process modeling method in accordance with an embodiment of the present invention.

Referring to FIG. 1, the method for generating a 4D CAD construction process model of the present invention starts with a step of constructing a construction plan and a 3D CAD model in step S11.

Subsequently, in step S12, the planning schedule information is generated from the construction plan. The planning schedule information is structured information in which the planned unit work is matched with the work time interval according to the process sequence.

In step S13, image masks of the unit components constituting the 3D CAD model are generated for each unit operation from the 3D CAD model.

The image mask may be a unit member element in the 3D CAD model, such as, for example, a beam, a girder, a panel, a stringer, etc., It is a projected image of the element.

Each unit element is given a unique member identification number, preferably a member identification number reflecting the process order.

Likewise, an image mask can be given a mask identification number corresponding to each unit element. In some cases, a plurality of mask identification numbers may be associated with one member identification number, since a plurality of image masks having different projection views for one unit element element may be generated.

On the other hand, a plurality of site surveillance cameras are installed on the construction site. The projection point for generating the image mask is a 3D point corresponding to the position, angle of view, focal length, aperture and direction of the field surveillance camera in the real world And may have coordinates and angles within the 3D virtual space of the CAD model.

Accordingly, an image mask of a unit element may be an image obtained by two-dimensionally rendering the shape of the unit element equivalent to an image captured by a specific field surveillance camera.

In step S14, a field processed image is generated by removing noise and construction equipment interference from the captured image obtained from the field surveillance camera capturing the construction site in the same position and direction as the projection time for generating the image mask.

Specifically, in the first binary image obtained by binary conversion of the captured image of the RGB color space to have two pixel values of 0 and 1, the color value of the captured image of the HSV color space converted from the captured image of the RGB color space H) image is subtracted from the second binary image obtained by binary conversion so as to have two pixel values of 0 and 1, so that construction equipment interference can be removed from the photographed image.

Furthermore, using known noise cancellation filtering techniques such as, for example, median filtering and morphing techniques including closure and magnification operations, respectively, can eliminate noise generated in the shot image or caused by construction equipment interference have.

Depending on the embodiment, the background image may also be removed from the in-process image. Known background image extraction techniques such as Gaussian mixture modeling techniques can be utilized to extract and remove background images from the captured images.

In step S15, the image masks matching the generated in-process image are extracted, and the unit operations advanced by the matched image masks are identified.

In particular, it is possible to identify the current ongoing unit operation from the identification number of the image mask corresponding to the latest process sequence among the matched image masks.

In the step S16, the time period from the time when it is determined that the image masks relating to the specific unit member are matched to the time when it is determined that the image masks relating to the unit members in the next process order are matched, And updates the progress schedule information by matching the specified work time interval with the identified unit work.

Specifically, the area values of the mask areas of the matched image masks are summed up, and the point at which the summed mask area area values are discontinuously different from the previous value is defined as the end point of the previous unit operation and the start point of the new unit operation It is possible to update the progress schedule information by specifying the work time interval and matching the specified work time interval with the identified unit work.

The 4D CAD construction process model is added to the immediately preceding 4D CAD construction process model by adding the operation time frame of the updated progress schedule information and the member elements per unit work of the 3D CAD model on the basis of the identified unit work in step S17 Update.

In this way, each time period in which the actual work is performed for each member element of the 3D CAD model is automatically combined to construct a 4D CAD construction process model. It is then possible to visually represent the progressive assembly of the elements of the 3D CAD model according to a timeline based on the 4D CAD construction process model at any time.

The progressive process identification steps are specifically illustrated with reference to Figure 2 below.

FIG. 2 is a flowchart specifically illustrating steps for identifying an advanced process using a field process image and image masks, among methods of updating an automated 4D CAD construction process model according to an exemplary embodiment of the present invention.

Referring to Fig. 2, step S14 of Fig. 1 may consist of steps S141 to S146, and subsequent step S15 of Fig. 1 may consist of steps S151 to S152 .

In step S141, a shot image 21 is acquired from an on-site surveillance camera that photographs the construction site in the same position and direction as the projection time for creating the image mask. Typically, the field surveillance camera can provide a captured image 21 of an RGB color space.

The photographed image 21 may be a single original frame photographed at a specific point in time or an average frame obtained by averaging a plurality of frames photographed for a predetermined time.

In step S142, a first binary image 22 is generated from the captured image 21 in the RGB (Red, Green, Blue) color space.

According to an embodiment, the captured image 21 in the RGB color space may first be converted to a grayscale image, and then the first binary image 22 may be generated from the grayscale image on the basis of a predetermined threshold value.

In step S143, the RGB color space of the shot image 21 is converted into a HSV (Hue, Saturation, Brightness Value) color space, and a color H information image 23 is obtained.

In step S144, the second binary image 24 is generated on the basis of a threshold value capable of highlighting the pixels 231 constituting the construction equipment from the color information image 23. [

Construction equipment is generally distinguished from buildings under construction due to reasons such as safety during construction, and thus has a color with good visibility, for example, red, orange, green, and yellow. In the HSV color space, the saturation (S) and brightness (V) change sensitively to changes in the weather or light conditions of the construction site, but since the color H is relatively insensitive to this change, It is possible to distinguish the building under construction and the construction equipment from each other without regard to shooting conditions or time zone.

Also, the color of these construction equipments can be easily recognized by users at the time of construction. Therefore, the construction equipment can be distinguished by the color value of the HSV color space.

If the construction equipments to be removed from the photographed image have a plurality of kinds of colors, a plurality of binary images are created with a plurality of threshold values for allowing only each equipment color to be highlighted, and these binary images are merged, 2 binary image 24. < / RTI >

In step S145, the construction equipment removal image 25 is generated by subtracting the bright pixels 241 of the second binary image 24 from the first binary image 22. [

At step S146, a noise-free on-scene processing image 26 is generated for the construction equipment removal image 25. [ According to an embodiment, the noise removal operation may be performed using median filtering, morphological method, or both techniques.

Median filtering is a filtering that replaces the value of a particular pixel with the median value of the surrounding pixels surrounding the pixel so that it can remove pixels that are projected relative to the surrounding pixels locally, Less.

Instead of the median filtering, a mean filter, a weighted average filter, an alpha-trimmed filter, a Gaussian filter, a hybrid median filter, a minimum mean square error filter, an anisotropic diffusion filter, have.

The morphing technique includes closing or enlarging (dilation) operations and the like. The morphing technique is a technique of removing holes and / or gaps in an object formed during the process of removing pixels of the second binary image corresponding to the construction equipment from the first binary image, Can be corrected.

Subsequently, in step S151, for the in-situ processed image 26, the image masks 27 are superimposed in the order of the process, and the image masks 27 determined to match the in- (27).

At this time, the image masks 27 to be superimposed are limited to the image masks of the unit member elements of the 3D CAD model corresponding to the positions of the specific field surveillance cameras and the members to be photographed at the angles.

The predetermined decision rule may be such that if there is a bright pixel area in substantially the same position and shape in the field processed image 26, for example, for the masked area 271 of the superimposed image mask 27, It may be a rule to determine that the mask 27 is matched to the in-situ processed image 26.

Depending on the embodiment, the predetermined decision rule may be determined, for example, if the area of bright pixels in the in-process image 26 masked by the masking area 271 of the image mask 27 is less than the area of the masking area 27 of the image mask 27. [ For example, greater than half of the area 271, may be a rule that determines that the image mask 27 is matched to the in-process image 26.

According to the embodiment, the determination rule may be simply a rule for comparing the center-of-gravity position and the area, but it may be a rule for determining based on the image similarity.

In step S152, unit work performed so far is identified based on the unit work corresponding to the last image mask 27 on the process order among the matched image masks 27. [

The areas matched by the image masks 27 among the in-situ processed images 26 may appear like the in-progress processed image 28. [

In particular, a unit operation corresponding to the last image mask in the process order among the matched image masks 27 may be identified as a unit operation in progress.

FIG. 3 is a conceptual diagram illustrating a step of identifying an advanced process by applying image masks according to a process sequence of specific members to an on-scene image in a method for updating an automated 4D CAD construction process model according to an exemplary embodiment of the present invention.

3, 14 image masks 32 to 37 for the 14 unit element elements obtained by projecting the 3D CAD model at the projection time corresponding to the position of the field surveillance camera and the angle are identified in the planned process sequence Are superimposed on the field machining image 31 in the numbered state, respectively.

When each of the image masks 32 to 37 is superimposed on the field processing image 31, only the pixels corresponding to the area corresponding to the projected image of the unit work member, that is, the masking area 321 in the field processing image 31 And the remaining portion is masked, i.e., masked.

Therefore, it is possible to easily obtain the area of the pixels in the area left after the masking of the in-process image 31, and to compare the area of the pixels with the area of the masking area 321, It is possible to easily determine whether or not the task is completed.

On the other hand, each unit work needs to know the start time and the completion time to obtain actual progress schedule information.

4 is a flowchart illustrating a method of updating an automated 4D CAD construction process model according to an exemplary embodiment of the present invention. Referring to FIG. 4, And Fig.

It has been described that the judgment rule for judging whether or not the image mask is matched involves the contrast of the area of the masking area of the image mask with the area of the pixels of the masked area of the field processing image, Of the area of the masking area.

On the other hand, in actual practice, the installation work usually starts by placing the member close to the working position. If the unit member is placed close to the working position, it will appear incompletely on the in-situ machining image. However, according to the above determination rule, it will be determined that the image mask corresponding to the unit member is also matched to the in-situ machining image.

Therefore, regarding the start and end of the installation work of the unit member, the image mask can be identified at a time substantially similar to the actual start time of the installation work, so that the identification time of the image mask is regarded as the start time of the installation work of the unit member can do. Furthermore, the ending point of the installation work of the unit member which has been in the installation work can be regarded as the time when the next unit member is identified.

Accordingly, the time from when the current image mask starts to be matched to when the next image mask starts to be matched can be referred to as the operation time period of the unit member corresponding to the current image mask.

When continuously summing the area values of the masked areas of the matched image masks, the sum value discretely changes from the previous value at the time when it is determined that the image mask for the new unitary member has been matched.

Accordingly, it is possible to specify the operation time interval by defining the point at which the sum value of the masking areas of the matched image masks discontinuously changes as the end time of the previous unit operation and the start time of the new unit operation.

Referring to FIG. 4, the sum of the masking area of the image mask from d-4 to d corresponds to the masking area area of one girder, where d is the number of discontinuous Respectively.

Therefore, the first girder installation work from d-4 to d is considered to be started from the d-date.

On the other hand, the sum of the masking area area on the day d + 6 is discontinuously increased again by the masking area area of the beam, so that the second girder installation work starting from the date d is completed, It can be assumed that the installation work of the beam has started.

5 is a block diagram illustrating an automated 4D CAD construction process model update apparatus in accordance with an embodiment of the present invention.

5, an automated 4D CAD construction process model update apparatus 50 includes a design information database 51, an image mask generation unit 52, a field processing image generation unit 53, a unit operation identification unit 54, A progress schedule information updating unit 55, and a 4D CAD model updating unit 56. [

The design information database 51 stores the construction schedule, the 3D CAD model, and the schedule schedule information generated from the construction schedule.

The image mask generation unit 52 generates image masks for the unit element constituting the 3D CAD model from the 3D CAD model for each unit operation.

The image mask may be given a mask identification number corresponding to each unit member element. In some cases, since a plurality of image masks having different projection times for one unit element can be generated, a plurality of mask identification numbers can be associated with one member identification number.

The projection point for creating an image mask can have coordinates and angles in a 3D virtual space of the 3D CAD model corresponding to the position, angle of view, focal length, aperture and direction of the field surveillance camera in the real world.

Therefore, an image mask of a unit element may be an image obtained by two-dimensionally rendering the shape of a corresponding unit element equivalent to an image picked up by a specific field surveillance camera.

The on-scene processing image generation unit 53 generates a on-scene processing image from which noises and construction equipment interference are removed from the captured image obtained from the field surveillance camera that captures the construction site in the same position and direction as the projection time for generating the image mask .

Specifically, the field processed image generating unit 53 may include a construction equipment interference removing unit 531, a noise removing unit 532, and a background image removing unit 533.

In the first binary image obtained by binary conversion of the photographed image of the RGB color space to have two pixel values of 0 and 1, the construction equipment interference removing section 531 extracts the HSV color space converted from the photographed image of the RGB color space It is possible to remove the construction equipment interference from the photographed image by subtracting the second binary image obtained by binary conversion of the color value H image among the photographed images so as to have two pixel values of 0 and 1.

Further, the noise remover 532 may be included in the shot image by using a known noise cancellation filtering technique such as, for example, median filtering and a morphing technique including a closure and an enlarging function, respectively, Thereby eliminating the noise caused by the noise.

According to the embodiment, the on-scene image generating unit 53 may further include a background image removing unit 533 that removes the background image from the on-scene processing image. Known background image extraction techniques such as Gaussian mixture modeling techniques can be utilized to extract and remove background images from the captured images.

The unit operation identifying unit 54 extracts image masks matching the generated on-scene processing image, and identifies the unit operations advanced by the matched image masks.

In particular, it is possible to identify the current ongoing unit operation from the identification number of the image mask corresponding to the latest process sequence among the matched image masks.

The progress schedule information updating unit 55 updates the process schedule information of the specific unit member from the point of time when it is determined that the image mask of the specific unit member is matched to the point of time when it is determined that the image mask of the unit member of the next process order is matched And updates the progress schedule information by matching the specified work time interval with the identified unit work.

Specifically, the area values of the mask areas of the matched image masks are summed up, and the point at which the summed mask area area values are discontinuously different from the previous value is defined as the end point of the previous unit operation and the start point of the new unit operation It is possible to update the progress schedule information by specifying the work time interval and matching the specified work time interval with the identified unit work.

The 4D CAD model updating unit 56 adds 4D CAD construction process models to the immediately preceding 4D CAD construction process based on the identified unit work and adds the operation time frame of the updated progress schedule information and the member elements for each unit operation of the 3D CAD model, Update the process model.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, but, on the contrary, Modification is possible. Accordingly, the spirit of the present invention should be understood only in accordance with the following claims, and all of the equivalent or equivalent variations will fall within the scope of the present invention.

Further, the apparatus according to the present invention can be implemented as a computer-readable code on a computer-readable recording medium. A computer-readable recording medium includes all kinds of recording apparatuses in which data that can be read by a computer system is stored. Examples of the recording medium include a ROM, a RAM, an optical disk, a magnetic tape, a floppy disk, a hard disk, a nonvolatile memory, and the like, and a carrier wave (for example, transmission via the Internet). The computer-readable recording medium may also be distributed over a networked computer system so that computer readable code can be stored and executed in a distributed manner.

21 Shooting Image 22 First Binary Image
23 Color Information Image 24 Second Binary Image
241 bright pixels 25 construction equipment removal image
25 Field Processing Image 27 Image Mask
271 Masking area
31 Field Processing Image 32 to 37 Image Mask
321 masking area
50 Automated 4D CAD construction process model update device
51 design information database 52 image mask generation section
53 Field Processing Image Generator
531 Construction equipment Interference elimination 532 Noise elimination
533 Background image removal
54 unit job identification unit 55 progress schedule information update unit
56 4D CAD model update section

Claims (20)

A 3D CAD model, and a schedule schedule information, respectively;
Generating image masks of unit components constituting a 3D CAD model from the 3D CAD model for each unit operation;
Generating a field processed image from which noises and construction equipment interference are removed from a captured image obtained from a field surveillance camera capturing a construction site in the same position and direction as the projection time for generating the image mask;
Extracting image masks that match the on-site processed image, and identifying unit work performed by the matched image masks;
The time period from the time when it is determined that the image masks relating to the specific unit member are matched to the time when it is determined that the image masks relating to the unit members in the next process order are matched, Updating the progress schedule information by matching the time period and the identified unit work; And
And updating the 4D CAD construction process model by adding, to the immediately preceding 4D CAD construction process model, the operation time frame of the updated progress schedule information and the member elements per unit work of the 3D CAD model, based on the identified unit work To create a 4D CAD construction process model. 2. The method of claim 1, wherein the image mask is a projected image of a unit element installed by a unit operation included in the 3D CAD model. 3. The method of claim 2, wherein the projection view for generating the image mask comprises coordinates in a three-dimensional virtual space of a 3D CAD model corresponding to a position, an angle of view, a focal length, Wherein the 4D CAD construction process model has an angle. The 4D CAD construction process model according to claim 1, wherein each unit element is assigned a unique member identification number, and each image mask is assigned a mask identification number corresponding to each unit element. Generation method. 5. The method according to claim 4, wherein a current ongoing unit job is identified from an identification number of the image mask corresponding to the last of the matched image masks. The method according to claim 1, wherein the step of generating the on-
In the first binary image acquired by binary conversion so that the captured image of the RGB color space has two pixel values of 0 and 1, the color value (H) image of the captured image of the HSV color space converted from the captured image of the RGB color space And subtracting the second binary image obtained by binary conversion so as to have two pixel values of 0 and 1, thereby eliminating construction equipment interference from the photographed image. 7. The method of claim 6, wherein generating the on-
Further comprising the step of eliminating noise generated by construction equipment interference or included in the photographed image by using a noise cancellation filtering technique and a morphing technique including closure and magnification operations, respectively, or both together. ≪ RTI ID = Process model generation method. 7. The method of claim 6, wherein generating the on-
Further comprising removing the background from the on-scene processing image. ≪ RTI ID = 0.0 > 5. < / RTI > The method of claim 1, wherein updating the progress schedule information comprises:
Summing area values of mask areas of the matched image masks;
Specifying a time period in which the summed mask area area values are discontinuously different from a previous value as a start time point of a previous unit work and a start time of a new unit work; And
And updating the progress schedule information by matching the specified work time period with the identified unit work. A computer-readable recording medium on which a program for implementing a method of generating a 4D CAD construction process model according to any one of claims 1 to 9 is recorded in a computer. A design information database for storing construction plans, 3D CAD models and planning schedule information, respectively;
An image mask generation unit for generating image masks for unit components of the 3D CAD model from the 3D CAD model for each unit operation;
A field processing image generation unit for generating a field processing image from which noises and construction equipment interference are removed from a photographed image obtained from a field surveillance camera photographing a construction site in the same position and direction as the projection time for generating the image mask;
A unit job identification unit for extracting image masks matching the on-site processed image, and identifying unit jobs processed by the matched image masks;
The time period from the time when it is determined that the image masks relating to the specific unit member are matched to the time when it is determined that the image masks relating to the unit members in the next process order are matched, A progress schedule information updating unit for updating the progress schedule information by matching the time period and the identified unit work; And
A 4D CAD model for updating the 4D CAD construction process model by adding the operation time frame of the updated progress schedule information and the member elements for each unit operation of the 3D CAD model to the immediately preceding 4D CAD construction process model based on the identified unit work, A 4D CAD construction process model generation device including an update unit. 12. The apparatus of claim 11, wherein the image mask is a projected image of a unit element installed by a unit operation included in the 3D CAD model. 13. The method of claim 12, wherein the projection time for generating the image mask is at least one of a three-dimensional virtual space of the 3D CAD model corresponding to the position, angle of view, focal length, Wherein the 4D CAD construction process model generation device is a 4D CAD construction process model generation device. The 4D CAD construction process model according to claim 11, wherein each unit element is assigned a unique member identification number, and each image mask is assigned a mask identification number corresponding to each unit element element Generating device. 15. The apparatus of claim 14, wherein a current ongoing unit job is identified from an identification number of the image mask corresponding to the last of the matched image masks. [Claim 12] The system of claim 11,
In the first binary image acquired by binary conversion so that the captured image of the RGB color space has two pixel values of 0 and 1, the color value (H) image of the captured image of the HSV color space converted from the captured image of the RGB color space And a second binary image obtained by performing binary conversion so as to have two pixel values of 0 and 1, thereby removing the construction equipment interference from the photographed image. Device. 18. The apparatus of claim 16, wherein the on-
Further comprising a noise canceling unit which is included in the photographed image by using the noise canceling filtering technique and the morphing technique including the closing and enlarging operation, or removes noise generated by construction equipment interference, Construction process model generation device. 18. The apparatus of claim 16, wherein the on-
Further comprising a background removal unit for removing the background from the on-scene processing image. The system according to claim 11, wherein the progress schedule information updating unit comprises:
Summing the area values of the mask areas of the matched image masks,
The time point at which the summed mask area area values discontinuously differ from the previous value is defined as the end time point of the previous unit work and the start time point of the new unit work,
And to update the progress schedule information by matching the specified work time period with the identified unit work. A computer-readable recording medium having recorded thereon a program for causing a computer to operate as a 4D CAD construction process model generation apparatus according to any one of claims 11 to 19.

Images