Origin of a Point Cloud

Wikipedia has defined a “Point Cloud” as “a set of data points in some coordinate system”. In a three-

dimensional coordinate system each point is commonly defined by an X, Y, and Z value, and is often employed to represent the surface of an object. Point cloud geometry is often referred to, rather mistakenly, as LiDAR. LiDAR is a technology that can be used to make point clouds; so measurements produced with LiDAR technology is more aptly known as LiDAR data. Most commonly, point clouds are the result of data being output by 3D scanners. Aerial and Terrestrial LiDAR sensors are two such examples of 3D scanners that perform a large number of measurements on an object's surface to output a point cloud, but as you’ll read, point clouds may originate from various other sources and technologies.

Photogrammetric Detection and Ranging (PhoDAR) is an alternative technology to conventional photogrammetry and LiDAR to generate point clouds. Though, PhoDAR and LiDAR are both capable of producing point clouds, the primary difference being PhoDAR uses photographic data (i.e. pictures) collected with a camera to generate a point cloud, whereas LiDAR involves the collection of data measured directly from the reflection of an object using a laser.

Another technology from which topographical point clouds are regularly created is IFSAR (Interferometric Synthetic Aperture Radar). . IFSAR sensors are typically flown in fixed-wing aircraft or mounted to satellites, and primarily uses X-band data of the electromagnetic spectrum to perform measurements. IFSAR data collection involves placing two radar (radio detection and ranging) antennas on the vehicle such that one antenna transmits a radar beam, and then both antennae receive the radar beam reflected from the target (usually the earth’s surface). Captured signals are combined with aircraft telemetry and positional data are processed to generate an interferogram, which is employed to generate elevation products.

Sound Navigation and Ranging (SONAR) technology uses the propagation of sound in an underwater environment to target objects. It is most commonly used to in the surveying profession to determine water depth (bathymetry). Sonar is applied to water-based activities because sound waves attenuate (taper off) less in water as they travel than do radar and light waves.

In future posts I’ll provide detailed analysis of the pros/cons, uses, and accuracies of the various technologies, but for now, the following table provides a brief overview of each.

Technology	Coverage	Accuracy	Use of Technology
Terrestrial LiDAR (Mobile or Static)	Small-to-Medium area projects	1cm	Buildings, Corridor mapping, Asset Inventory, Engineering Design
Aerial LiDAR	Large area projects, in varied terrain conditions	10 cm	Flood plain mapping, Disaster Management, Transportation and Engineering design, Impervious surface mapping, vegetation mapping
PhoDAR	Medium-area projects	15+ cm (dependent on GSD)	Same as above
IFSAR	Large area projects in varied terrain conditions	30 cm	Same as above
SONAR	Small-to-Medium area projects	5-10cm	Underwater studies, dredging, navigation

Cheers!
Srini

Belly of the BEAST

A bank of 2 TB hard drives in our Data Center.

In an effort to better support processing and managing “Big Data,” Michael Baker implemented the Baker Enterprise Architecture for Spatial Technology (BEAST) platform. The BEAST is a fully integrated enterprise solution that allows for server-side and cloud-based processing, production, storage, and hosting within a single framework.

The BEAST hosts a variety of tools that are available to all Michael Baker offices and enables our staff to collaborate on projects in an effective and efficient environment across offices, regions and even continents.

Michael Baker employees typically access the BEAST via Citrix MetaFrame interface. This allows users to access the data through a thin client with all data and software residing “in our private cloud”.

Software and database solutions that we use for LiDAR processing and feature extraction or GIS product production include Optech’s LiDAR Mapping Suite (LMS), ArcGIS (in multiple flavors), Orbit GT, SQL Server, FME, MicroStation, GeoCue, TopoDOT and LizardTech GeoExpress along with custom developed applications. By utilizing fully virtualized environments, the BEAST allows for unparalleled flexibility to rapidly scale additional resources for large-scale project delivery. In many cases, we will also utilize Amazon and Microsoft Cloud services for surge needs; while we continually evaluate overall cost effectiveness of those and other cloud platforms considering the practicalities of uploading and storing terabytes of LiDAR data each day to a cloud service.

Two 56 TB storage servers at a local production office.

In our last posting, Aaron briefly questioned how you would manage 4 terabytes of new data on a daily basis. One of our solutions is the management of data and core processing functions at the BEAST. Where more robust processing is required and a thin client simply will not do, we push data to offices for further processing. Our regionally based processing centers have robust LiDAR server solutions with 50 – 200 Terabytes of storage each. Due to increase network traffic to support the rotund datasets, we have deployed Cisco Meraki network infrastructure for Multi-gigabit connectivity.

Cheers!
Art

Art Morris is a Systems Analyst in Michael Baker's Harrisburg office.  Art joins the Mobile LiDAR team from IT Services, to focus his talents on the acquisition, transfer, and processing of the vast amount of LiDAR data collected by our vehicles.

Art's IT background helps him to act as a liaison between departments, offices and regions, to ensure synergistic collaboration between the Mobile LiDAR team and Michael Baker's IT organization.

Memorial Day

On behalf of Michael Baker International, the Mobile LiDAR Team would like to take the opportunity this Memorial Day to recognize and remember our country’s military members who made the ultimate sacrifice to protect the very freedoms we enjoy today. As you spend this day with loved ones, please think about those who fought and gave their lives to secure the liberties that allow us to live, work and make a difference. 

As a firm that employs veterans from many countries and delivers mission-critical services in support of our country’s military around the globe, we’d like our active duty, reserve, guard and contractor friends and family to know they are in our thoughts during their extended deployments and separation from friends and family.

If you are able and so inclined, please consider making a contribution to organizations that support the surviving spouses and children of our fallen heroes. Such organizations include (but not limited to):

• Special Operations Warrior Foundation - http://www.specialops.org/
• Children of Fallen Patriots - http://www.fallenpatriots.org/
• Intrepid Fallen Heroes Fund - http://www.fallenheroesfund.org/

Thank you,

The Mobile LiDAR Team

Launching a Fleet

It’s been a while since our last posting, and we’ve certainly appreciated your patience during this brief hiatus – but rest assured, plenty of new content (and contributors) is on the way. The expansion of our Fleet –4 Mobile LiDAR systems, plus an integrated Laser Crack Measurement System (LCMS), plus an integrated Ground Penetrating Radar (GPR), plus multiple static LiDAR units – temporarily diverted our attention from our social media content while we focused on scaling other areas of the operation to handle the workload simultaneously coming from four systems. Luckily our faithful followers have been treated to a regular stream of content to  Instagram thanks to our new Fleet Manager.

Quadrupling your collection capacity in less than 8 months looks like a pretty straight forward process on paper, but launching a fleet of the most sophisticated mobile mapping systems at times felt like we were choreographing logistics for NASA’s Mission Control. Four times the number of collection systems means four times the amount of logistics, maintenance, personnel, proposals, storage capacity, and last but not least…Data!

It’s no small feat effectively handle up to 4 Terabytes of new data each and every day. Any way you slice it, that’s a lot of data. So there’s your parting thought for the day – How would you transfer 4 TBs of data from various locations throughout North America on a daily basis?

Cheers!
Aaron
LiDAR Services Director

Direction of our "Light"

No longer only a tool for science, landform mapping and Geomatics Engineering, we have expanded our LiDAR (Light Detection and Ranging) usage by further analyzing, managing and manipulating the LiDAR data sets across a host of practical applications. This understanding of how to fully utilize and interact with LiDAR data is allowing us to design and create structures and systems – notably in the building, infrastructure and in situ environments – that are safer and more resilient.

But what’s fascinating is that we are still just beginning to achieve LiDAR’s potential for information gathering and usage. One emerging market that Michael Baker is focusing on is in enabling autonomous vehicles while building safer, smarter, real-time information systems within our transportation networks. The impact and applications for this type of ubiquitous, low-cost form of mass-produced LiDAR technology associated with automotive, rail, airborne, marine and other vehicles can change the way the world moves people and things.

What you see is what you get (WYSIWYG) with Mobile LiDAR.  Completely immersive scenes are generated along transportation corridors for an array of applications.

The LiDAR data we collect at Michael Baker is of a very precise and dense variety. We apply LiDAR to the purpose of very selective illumination of phenomena in the line-of-sight of our sensor systems. We are selective in where and how we deploy our systems and we carefully plan collections. We primarily direct our LiDAR sensors at our customers’ infrastructure or assets for the comprehensive collection of a point cloud that will be analyzed to extract very specific information. Combined with the acquisition of panoramic and other imagery, our LiDAR and imagery data becomes quite massive in terms of the individual artifacts of data created within the collection process. Now we’re developing methods and algorithms that interact directly with these point clouds to perform conformity analysis with design standards and to analyze stress and perhaps load forecasting on infrastructure. This is the way we can extend the potential for that infrastructure though effective operation and maintenance practices, as well as predict the infrastructure’s ability to bear existing or additional demand loads.

Critical infrastructure is accurately mapped with application of sound surveying principles.  

With our Mobile LiDAR capabilities a single vehicle has the potential to collect in excess of a terabyte of data each day. Billions of LiDAR shots are recorded, tens of millions of photographs are taken and hundreds of thousands of individual LiDAR-based files are generated in a relatively short time period as part of the overall process of creating LiDAR deliverables for a typical project. We currently operate four extraordinarily hi-tech Mobile LiDAR systems, running a collective 800 days per year. We annually collect LiDAR to create more than a petabyte of raw engineering quality data to produce discrete, extracted information about specific infrastructure, assets or ecosystems. After collection, we may double the amount of data we produce and manage for the deliverables we derive from the raw LiDAR or images. Accurately and cost-effectively creating and managing such information is a constant challenge for a large scale LiDAR operation such as ours.

The laser sensors we employ are generally indiscriminate when collecting data. A pulse is emitted and returned. We collect everything in the line-of-sight potential of our eye-safe lasers with great precision. Then we analyze the aggregation of billions of LiDAR points within a dataset to determine any phenomena and its relevancy to an intended purpose.

The use of highly detailed Mobile LiDAR datasets provide the mechanism for feature extraction and attribution when used in conjunction with 360° spherical photography.

As we continue to find new data to mine and explore all of the possibilities that LiDAR technology affords the transportation industry, it’s exciting to ponder the industry’s evolution and what the future looks like.

I invite you to join me at the 2016 MAPPS Winter Conference for a presentation titled: “Big Data or Data That’s Big? The Pervasiveness of LiDAR in the Approaches to Engineering” as I am further addressing these topics and how we are managing data. And, continue to follow this blog for future insights.

Cheers!

Bob