Entity modeling CSG vs. Brep

3D models in computers are classified based on geometric characteristics. They can be divided into three types: box models, surface models, and solid models. If an object is classified by the method of representation, the entity model can be divided into three categories: Decomposition representation, constructive solid geometry, and boundary representation. Common decomposition methods include: quadtree, octotree, multi-tree, BSP Tree, and so on.
Main Methods for constructing representation: scanning representation, constructing ry representation, feature and parameterized representation. A typical representation of a boundary is a wing-edge structure. In the CSG modeling method, an object is represented as a Boolean operation result of a series of simple basic objects (such as cubes, cylinders, and CONES). The data structure is a tree structure. The leaf is the base ontology or transformation matrix. The node is an operation. The top node corresponds to the object to be modeled. A Brep object is represented as many surfaces (such as patches, triangles, and splines) bond to form a closed space area. Brep advantages:1. There are a lot of information about the surface, edge, point and their relationship.
2. It is conducive to generating and drawing line charts and projection charts, and to calculating geometric characteristics. It is easy to link with two-dimensional drawing software and be associated with the same surface modeling software. Brep limitations:Because its core information is a plane, the ability to describe a geometric object as a whole is relatively poor, so it cannot provide information about the entity generation process, it is also impossible to record the raw data of the elements that constitute the basic element of the ry. At the same time, the amount of information required to describe the object is large, and the expressions of the boundary expressions are not unique. CSG advantages:The method is concise, fast to generate, easy to process, and has no redundant information. It also records in detail the original feature parameters that constitute an object, you can even modify the surerx parameters or surerx parameters to re-splice them if necessary. The data structure is relatively simple, the data volume is small, the modification is easier, and can be easily converted into a boundary (Brep) representation. CSG limitations:Because the information is simple, such a data structure cannot store the final details of an object, such as the boundary and vertex information. Because CSG represents the type of receptor and the restriction on the type of body operation, CSG has great limitations on the coverage of the form and local operations on the form (for example, not easy to implement, display CSGThe result form is also long. Which method is used? Depending on different perspectives, they both have advantages and disadvantages. You can consider the following points:
1. Reality-whether the model needs to approximate the actual material density, thickness, and volume, or whether the actual surface details need to be displayed
2. Precision-what kind of precision-level hybrid model (hybird model) is based on Brep and CSG. In the same system, the two are combined to represent the entity together. The CSG method is used as the external model of the system, and the B-rep method is used as the internal model. the CSG method is suitable for user interfaces and is transformed into a B-rep data model within the computer. It is equivalent to expanding the data structure of the boundary method on the CSG tree node. The hybrid mode is a logical extension based on CSG, and plays a leading role in the CSG structure. The existence of B-rep reduces the mathematical computing workload in the intermediate link, it can completely express the geometric and topological information of objects and facilitate product model construction. There is also a spatial division representation, which uses a quad-tree or eight-Cross data structure to represent a 2D/3D model.