Rakesh Kumar Mishra, Bharat Lohani Geoinformatics division. Kanpur, INDIA. Indian Institute of Technology Kanpur 1

Transcription

1 An Object-Oriented Oriented Software Development Approach to Design Simulator for Airborne Altimetric LiDAR, Bharat Lohani Geoinformatics division Indian Institute t of Technology Kanpur Kanpur, INDIA Indian Institute of Technology Kanpur 1

2 Pi Principle i of flidar Laser transmitter INS, GPS, Scanner (x,y,z, direction cosines) d X,Y,Z Indian Institute of Technology Kanpur 2

3 LiDAR Technology Provides accurate topographic data at high speed Data collection with higher density, accuracy & less time Weather and light independent Applications: DEM generation Flood hazard zonation Cellular networks etc. Indian Institute of Technology Kanpur 3

4 Object-Oriented Oriented Software Development Indian Institute of Technology Kanpur 4

5 Initial Investigation (Why simulator?) LiDAR Instrument is very costly LiDAR data is not available in most of the countries LiDAR data is not available for teaching as required LiDAR data is not available for research as required Software for flight planning Indian Institute of Technology Kanpur 5

6 Requirements User friendly GUI Simulation of generic as well as commercial sensors Simulation of earth like surfaces Flight trajectory as in case of actual flight Possibilities of error introduction Output data in common format Help and tutorial Indian Institute of Technology Kanpur 6

7 Feasibility study Sufficient background is available for the system development The system can be engineered using current technology Development can be done within the budget & time Developed system will be useful for the user group Indian Institute of Technology Kanpur 7

8 Object-Oriented Oriented analysis Surface User Trajectory Integration Output Sensor Use case diagram Indian Institute of Technology Kanpur 8

9 Objects in the problem domain is identified Object relationships are made Object state table is developed Inheritance diagram for objects is made Indian Institute of Technology Kanpur 9

10 Object-oriented design INPUT Integration Terrain Sensor Trajectory component component component OUTPUT Indian Institute of Technology Kanpur 10

11 Terrain components Terrain component Polynomial surface Raster surface Fractal surface Indian Institute of Technology Kanpur 11

12 Sensor components Sensor component Generic sensor ALTM ALS 50 Indian Institute of Technology Kanpur 12

13 Trajectory components Trajectory component Location Attitude Acceleration Velocity Indian Institute of Technology Kanpur 13

14 Class design Identify classes from the components Identify subclass within each class Identify abstraction in each class Identify the common behaviour of classes Indian Institute of Technology Kanpur 14

15 System mimpl implementation m ti n Java is used to realize the design Each classes are implemented with its relationship Standard way of coding is used Methods are designed for each basic task Indian Institute of Technology Kanpur 15

16 Complexities handled d Efficient algorithms are designed Threads are used to optimize software execution Special data structures are designed to handle memory problems New file formats are designed to improve I/O Indian Institute of Technology Kanpur 16

17 Software Screenshots Indian Institute of Technology Kanpur 17

18 Polynomial surface Indian Institute of Technology Kanpur 18

19 Raster surface(city model) Indian Institute of Technology Kanpur 19

20 Fractal surface Indian Institute of Technology Kanpur 20

21 Acceleration Indian Institute of Technology Kanpur 21

22 Sensor component Indian Institute of Technology Kanpur 22

23 System defined optimal flight lines Indian Institute of Technology Kanpur 23

24 User defined optimal flight lines Indian Institute of Technology Kanpur 24

25 Attitude Indian Institute of Technology Kanpur 25

26 Error simulation Indian Institute of Technology Kanpur 26

27 Output generation Indian Institute of Technology Kanpur 27

28 Software Results Indian Institute of Technology Kanpur 28

29 3D Raster terrain (Displayed in Surfer) Altitude=210m Overlap=4% Velocity=60m/s Sensor-ALS-50 Firing frequency=20khz Scan frequency=48hz Scan angle=40 Flight area=430m 430m Indian Institute of Technology Kanpur 29

30 Lidar data plot in plan A-AA B-B Indian Institute of Technology Kanpur 30

31 Profile A-A A with and without error Indian Institute of Technology Kanpur 31

32 Profile B-B B with respect to flight lines Indian Institute of Technology Kanpur 32

33 LiDAR data without error Indian Institute of Technology Kanpur 33

34 LiDAR data with error Indian Institute of Technology Kanpur 34

35 Data without attitude variation Indian Institute of Technology Kanpur 35

36 Data with attitude variation Indian Institute of Technology Kanpur 36

37 Fractal surface displayed in Surfer Indian Institute of Technology Kanpur 37

38 LiDAR data of fractal surface Bharat Lohani, IIT Kanpur India Indian Institute of Technology Kanpur 38

39 Terrain with objects Indian Institute of Technology Kanpur 39

40 LiDAR data of terrain with objects Altitude=490m Overlap=2% Velocity=60m/s Sensor-ALS-50 Firing frequency=20khz Scan frequency=48hz Scan angle=50 Flight area=640m 460m Indian Institute of Technology Kanpur 40

41 1 2 Profile view of buildings Indian Institute of Technology Kanpur 41

42 Effect of different flight direction Indian Institute of Technology Kanpur 42

43 Indian Institute of Technology Kanpur 43

44 Use of Simulator Indian Institute of Technology Kanpur 44

45 Education Process of LiDAR data generation Effect of change in various parameters Effect of error in data Effect of different sensors on LiDAR data Generating data for known ground truth Conducting various lab exercises Indian Institute of Technology Kanpur 45

46 Research Evaluation of information extraction algorithms Assessing effect of error on performance of algorithms Study the effect of parameters on data Generate data of different specifications with no cost Finding optimal data specification for an application Indian Institute of Technology Kanpur 46

47 Flight planning Determine the optimal flight line Effect of sensor parameters on data Effect of data density Determine the optimal sensor parameters Indian Institute of Technology Kanpur 47

48 Conclusion Offers a user friendly GUI based interface Simulate the process of LiDAR data collection Freedom to set the sensor parameters Many data sets can be generated for the same terrain Ideal software tool for LiDAR research and education OOSD makes it easily maintainable and scalable software Indian Institute of Technology Kanpur 48

49 Sr. Project Associate Geoinformatics division Indian Institute of Technology Kanpur Kanpur, INDIA i Indian Institute of Technology Kanpur 49