Optimization Model of BIM-based three-dimensional reconstruction technology and engineering model of visual perception
Keywords: Three-dimensional reconstruction, visual perceptual model, engineering optimization, modeling, analysis.
Abstract.Vision-based reconstruction is still there is a big limitation. Through its research-based approach introduces the primary visual three-dimensional reconstruction techniques, advantages and disadvantages of various methods were compared, it is desirable in this area can have a more comprehensive grasp, to further clarify the direction of future research. In order to improve the efficiency of the design and construction of bridge engineering, building information modeling (BIM) is introduced into the bridge project in the past. By analyzing the characteristics of bridge design and construction and the problems proposed bridge design and construction BIM-based optimization solutions, including preliminary design optimization, optimization of construction design, construction process optimization, optimization of the construction schedule and construction management optimization, combined with practical engineering project the applicability and effect analysis. Case application shows, BIM Bridge Project is applicable, can provide effective support for the bridge design and construction, thereby reducing rework and improve efficiency. The study may be large or complex bridge engineering BIM improve the efficiency and effectiveness of the design and construction of reference.
The relevant information and data building information model is based on building projects as the basis for the model, building model were established by the real information of the digital information simulation building has, it has the visibility, coordination, simulated sex, optimality and showing of five characteristics. The BIM technology in the field of bridge engineering construction is currently in the early stages, preliminary exploration in the design, construction, and post-operation maintenance and repair of the entire life cycle of how to use BIM technology to improve design efficiency, improved design quality, strengthen the construction organization and post operations management, specific method and the application of BIM technology can bring benefits, hoping to BIM in bridge engineering to develop ideas.
In recent years, domestic construction projects in the field of non-BIM is none other than the hottest technology in the construction industry has achieved good results after the application, the state began to vigorously promote the railway, highway, water conservancy and hydropower industry application of BIM technology in fields such as engineering, and bridge engineering in the construction field and a large proportion, especially high-speed railway, mountain railway, roads, bridges, often accounting for a larger significance in bridge engineering applications BIM technology on the entire major project, the paper will design, three stages of construction, operation and maintenance of the latter part of the project life cycle are the practical application of research needs and the effect of BIM technology.
In recent years, should the needs of economic development, large, extra large bridge project more and more, such as China, Hangzhou Bay Bridge, the Hong Kong-Zhuhai-Macao Bridge, which bridge design and construction of a higher requirement. Bridge construction project not only involves complex geographical environment, and involves a number of complex projects, the most typical is the Steel Bridge. Currently, the design of large bridge projects usually rely on the traditional two-dimensional drawings and graphs to analyze the design by closing existing in conflict; construction planning is largely dependent on the experience of project managers to develop and implement, and is also a two-dimensional drawings to show. However, since the bridge project their own characteristics, its design complexity, component many rely solely on the traditional two-dimensional drawings difficult to detect in advance or found conflicts existing in the design; these design problems usually can be found in the construction phase, thus affecting the construction schedule and cost, will also affect construction safety. At the same time, in order to resolve the problem of engineering design and construction, we had to deploy more staff, which is also a significant increase in management costs. Therefore, to ensure the feasibility of the bridge may be constructed of engineering design and construction programs for efficient implementation of the bridge project is very important.
Preliminary design stage is divided into two stages of design and post-design, including pre-design project approval, feasibility studies and scheme comparison and other parts, three-dimensional solid model of the bridge by using parametric modeling tools can be easily established according to the actual need to adjust the size, and the actual effect of the bridge into the real-time dynamic display, to achieve WYSIWYG, can intuitively design concept, design effects directly model the three-dimensional visualization of the project as a carrier to deliver policy-makers, which greatly facilitate the adjustment of the design, be revised in accordance with amendments and rendering, and cost control by adding information to keep abreast of changes after the investment increases and decreases, so that the bridge-bridge quickly determine preliminary program is very convenient and efficient. Figure 1 is a railway bridge stayed Bridge main bridge model, we need to establish a special bridge structure according to the characteristics of the family library for complex bridge structure using three-dimensional expression of BIM model than the traditional two-dimensional drawings clearer and easier to understand.
The Proposed Methodology
Three-dimensional modeling techniques.The use of modeling software for three-dimensional modeling is commonly used method, but modeling the need to spend a lot of manpower and material resources are often prohibitive, reconstruction effect is often unsatisfactory. Vision-based reconstruction technique to solve this problem and provides a new way of thinking.
Three-dimensional vision-based three-dimensional reconstruction technology, which uses computer vision methods of three-dimensional model reconstruction of the object, is the use of a digital camera as the image sensor, the integrated use of image processing, visual computing technologies such as non-contact dimensional measurement, obtaining object using a computer program information. The advantage is that the shape of the object is not restricted to rebuild faster, can achieve automatic or semi-automatic modeling, three-dimensional reconstruction is an important direction of development, can be widely used, including autonomous mobile robot navigation systems, remote sensing and aerospace, industrial fields of automation systems, etc., the economic benefits generated by this technology is very impressive.
As an important branch of computer vision technology, vision-based three-dimensional reconstruction of Marr visual theoretical framework is based on the formation of a variety of theoretical approaches. For example, according to the number of cameras can be divided into monocular vision method, binocular vision method, three monocular vision or monocular vision method; according to different principles, vision-based method can be divided into regions, feature-based visual method , model-based and rule-based visual methods; according to the obtained data the way, can be divided into active and passive visual method visual method.
Figure.1 Three dimensional reconstruction technique
According to research at home and abroad in recent years, were selected based on visual presentation of three-dimensional reconstruction of research and practical application of several methods and more comparative analysis, pointed out the main challenges for the future and the future direction of development. Depending on the number of cameras to use, this article will be divided into three-dimensional reconstruction method based on the visual method of monocular vision, binocular vision trinocular vision method and three methods were introduced, focusing on the monocular vision method.
Monocular vision method.Monocular vision method is the use of a camera for three-dimensional reconstruction method. Images used can be a single point of view of single or multiple images can also be a multi-view multiple images. The former is mainly characterized by a two-dimensional image depth information deduced, these features include two-dimensional shading, texture, focus, contour, etc., it is also referred to as X shape recovery method. This simple device structure class methods, the use of single or small number of several images can be reconstructed three-dimensional object model; less than that normally required conditions more idealistic, practical application is not very satisfactory, the effects of reconstruction in general. The latter by matching different images of the same feature points matching using these coordinates in space constraint obtaining information in order to achieve a three-dimensional reconstruction. This method can be implemented in the reconstruction process of camera calibration, to meet the needs of large-scale reconstruction of three-dimensional scene, and in the case of resource-rich image reconstruction is better; the downside is that a greater amount of computing, a long time to rebuild. The following describes several major monocular vision method.
Shading method.Shading method, that the brightness of the shape recovery method (SFS). This approach by analyzing image brightness information, using reflected light model, restore the normal to the surface of three-dimensional reconstruction information. Horn in 1970 first proposed the concept SFS methods, and gives a non-linear relationship between the two-dimensional image showing the brightness of each pixel in the corresponding three-dimensional point of law to the reflectance of light and the direction of Partial Differential Equations , the brightness of the equation.
However, this method is a SFS under-constrained problem and needs to solve other constraints. Therefore, the traditional method of SFS also based on three assumptions. The main advantage of the brightness of the method is that it can recover from a single image in a more precise three-dimensional model can be applied in addition to mirror the object almost all types of objects. However, the brightness of the reconstruction of relying solely on mathematical calculations, results are poor, but because of the lighting conditions more stringent requirements, the need to know the precise position and orientation of the light source and other information, so that the brightness of the method is difficult to apply in the case of an outdoor scene lighting and other complex three-dimensional reconstruction on.
Photometric stereo.Although the shading method to support the reconstruction of three-dimensional model from a single image, but less information is available in a single image, the actual reconstruction of the general effect. So Woodham of SFS method is proposed to improve the photometric stereo.
Photometric stereo by a plurality of non-collinear light source to obtain multiple images of the object, and then a different image brightness simultaneous equations, solving the surface normal direction of the object, and ultimately restore the shape of the object. Technically, the use of two light sources can be obtained method object to the information, but the use of multiple sources of data redundancy can be resolved by the shadows and specular reflections caused by such factors can not solve the problem, better robustness, reconstruction effect It can be improved, so the current method basically using a plurality of (four to six) three-dimensional reconstruction of the light source.
Photometric stereo advantages and brightness of the same law, the use of multiple images at the same time avoids the problems of ill shading method, and the use of multiple light sources also increased constraints, to improve the accuracy and robustness of the method; it the disadvantage is difficult to apply a mirror surface object and three-dimensional reconstruction of outdoor scenes and objects.
Texture law.Humans can surface texture by projection on the retina perceive three-dimensional shape of the object, so the visual image information gradient texture can be used as information for Shape and depth cues. Based on this theory, the analysis can be repeated by surface texture unit image size, the shape, the recovery of the normal object, the depth information to obtain three-dimensional geometric model of the object, i.e., texture profile method for recovery.
Texture is the basic theory of law: For a smooth surface and having a repeating texture units covering the object of which, when projected on the two-dimensional image, texture unit on which will be deformed, this deformation is divided into projection distortion (projective distortion ) and perspective shrinkage. Projection distortion so the farther away from the image plane texture unit looks smaller foreshortening distortion and image plane makes an angle greater texture unit looks shorter. Because these two variants can be measured from the image, so it can be analyzed after deformation texture units, reverse strike the surface normal and depth of information, three-dimensional reconstruction.
Profile method. This method of contour images of objects through a plurality of angles to give a three-dimensional model of the object. Profile method can be divided based on voxel cone prime three methods based on visual and shell.
Figure.2 Visual perception model
Reconstruction of 3D Vision technology is still in the exploratory stage, the practical application of the various methods is still some distance away from a variety of application needs to be urgently met. Therefore, in the future for a long period of time, we also need to do more intensive research in this field. This study shows that, BIM can provide effective support for the bridge design and construction. This study was expected to provide reference for increasing large, complex bridge design and construction efficiency and effectiveness, as well as assist in the promotion and application of BIM in the field of civil engineering. I believe that with the continuous promotion of theory and technology of BIM, BIM applications in civil engineering will become increasingly widespread, so as to improve their quality, efficiency and management level. Safety-critical structural bridge engineering, maintenance and repair of the late, operations management, file management can take advantage of powerful information technology BIM, visualization capabilities to achieve.