S1: Csci 8701 - Week 3
- Administrative
- HW 2 - Pick up peer HW1 for review
- submit peer review and project data seperately
- put peer review and project data on web
- Goals:
- Spatial Concepts and Data Models
- Recommended Readings:
- Chapter 2 , Supplementary Book
Copyright: S. Shekhar, C. S. Dept.,
University of Minnesota, Minneapolis, MN 55455. ,,.(home)
S2: Chapter Outline
- 2.1 Models of Spatial Information
- 2.2 Three step Database Design
- 2.3 Extending ER with spatial concepts
- 2.4 OO Data Modeling with UML
Copyright: S. Shekhar, C. S. Dept.,
University of Minnesota, Minneapolis, MN 55455. ,,.(home)
S3: %Models of Spatial Information
- %Why is spatial special?
- Continuous domains (e.g. space)
- complex objects (e.g. polygons)
- Field vs. Object based models
- Fig. 2.1 (pp. 54)
- Modeling a continuous phenomenon
- Field functions (Fig. 2.1(b))
- Set of polygon objects (Fig. 2.1(c))
Copyright: S. Shekhar, C. S. Dept.,
University of Minnesota, Minneapolis, MN 55455. ,,.(home)
S4: %Key Concepts - Field Model
- Components:
- Spatial framework: (a partitioning of space)
- field functions : framework -- > attribute domain
- field function operators
- Operations on spatial objects
- Local(f,g): f+g, f-g, f or g, ...
- Focal(f) : derivative(f)
- Ex. drainage direction (gradient) from elevation
- Zonal(f,z): aggegate f values group by z values
- Ex. average elevation of each state in USA
- Closure under operations
Copyright: S. Shekhar, C. S. Dept.,
University of Minnesota, Minneapolis, MN 55455. ,,.(home)
S5: %Key Concepts - Object Model
- Components:
- Classes of Spatial objects
- Spatial operators
- Classes of spatial objects
- points, point collections
- polygons, polygon collections
- linestrings, linestring collections
- Closure under operations
- Operations on spatial objects
- Set oriented
- Topological - see Table 2.1 (pp. 49)
- ... nine intersection model
- Metric
- Euclidean
Copyright: S. Shekhar, C. S. Dept.,
University of Minnesota, Minneapolis, MN 55455. ,,.(home)
S6: %Comparison with Relational Model
- Conceptual Model
- Entity Relationship, or UML
- Spatial relationship - crowd the diagram
- Pictograms to simplify the diagrams
- Make spatial relationships implict
- Example
- Figure 2.3 (pp. 59)
- Figure 2.6 (pp. 68)
- Pictogram specification (sec. 2.3.1, pp 65)
- Q? Do pictogram simplify logical level schema too?
Copyright: S. Shekhar, C. S. Dept.,
University of Minnesota, Minneapolis, MN 55455. ,,.(home)
S7: %Comparison with Relational Model
- Logical- Relational, OO, Object Relational
- User defined data types
- User defined functions
- Spatial data types can be incorporated
- Simplifies logical schema
- ... Removes FKs, tables for spatial relationships
- Physical- relations, indexes
- Storage of large spatial objects
- Spatial indexes (e.g. R-tree)
Copyright: S. Shekhar, C. S. Dept.,
University of Minnesota, Minneapolis, MN 55455. ,,.(home)
S8: %Exercises
Copyright: S. Shekhar, C. S. Dept.,
University of Minnesota, Minneapolis, MN 55455. ,,.(home)