Study of LIDAR mapping technology

Abstract

High-resolution digital terrain models (DTMs) are essential and key component for topographic applications such as earth moving industry, flood modeling and etc. They are generated by LIght Detection and Ranging technology which is short for LIDAR technology. It is one of the latest optical remote sensing technologies used for measuring the distance from a laser which transmits pulses to some target. Usually, airborne LIDAR systems return a three-dimensional cloud of point measurements which gives irregular spacing. Measurements from unwanted features and objects such as trees, vehicles should be classified and removed so as to generate a DTM. The exact location of a DTM in a real map is a critical job when GPS service is not available.

This project is mainly divided into two parts. The first part is to remove unwanted points from LIDAR datasets. Gridded LIDAR datasets are needed for generating DTM if measurements from wanted features are expected to be identified and separated from unwanted features. If LIDAR datasets are needed to be converted into either a regular or a square, grayscale grid image in terms of elevation, objects will be classified by the change of gray tone. Morphological filter (MF) and progressive morphological filter (PMF) have a task which is to extract the wanted points from LIDAR measurements. The second part is to find and map the exact location of the LIDAR image in the real Google map of Singapore Science Park. The real image which is retrieved form Google contains almost all information needed, while LIDAR image is formed through LIDAR datasets collected. Two types of LIDAR images are discussed in detail, which are of multi-color and BW (Black & Color). Elevation filter which leads to this difference between the two types is also investigated. In summary, all tasks assigned are accomplished successfully and fulfill all the requirements given.