Friday, February 12, 2016

Times they are a Changin....

MidNight Comments - Since I posted this about a month ago I have continued to think about what the heck may be going on?

Here are my ruminations:

In the AlphaBet, I believe that E is for ESRI?  Why?  Well the arc-business model that I believe that is emerging includes an opportunity for the Dangermonds to sell their company into the dominating geo-search engine that Maps and Earth Search.  Maps without all information search of WWW content are essentially maps of ignorance.  Further the scale of free and open content is clearly on the increase.  But for me the ultimate next phase that will drive the GIS sciences and currency of map information will be VR and virtual presence and augmentation of the always navigating personal device.

I don't know for certain but I would be really quite excited if suddenly we were to discover that ESRI now could fully draw-on all of the www content that have rendered in Google Maps and Earth?  All of their imagery, all of their StreetView, all of the constructions made and to be made using Earth Enterprise Engine, as well as KML by the TB.  How might that arrive?

Well a replacement of the Google Earth 3D-Globe and its programmatic design likely needs to be made.  The old one from Keyhole-era got them their first two-billion of so users... an new generation that streams 3D and all sorts of VR goodness will be needed?  Enter ESRI ArcGIS Pro 64 and their Globe browser, ArcGIS Earth, totally free and essentially a 1:1 or darned near 1:1 re-build of Google Earth IMHO... only without the data.  There is a ArcGIS Earth API as I understand it that is also to be included in the ArcGIS Pro.  ArcGIS Pro 64 as a development tool is also to be free?  So the last issue is how does the "hook" work?  We it looks to me that only a real ArcGIS Pro on a desk can generate content that can be viewed on ArcGis Earth or be mounted to the cloud server.  Also there likely will be an annual subscription to access the ESRI-isze earth reference data layering?

Thoughts?


Google's Microsofts Move to Mobile Changes Future of Mapping

by Laurie Sullivan @lauriesullivan,

Google began to quietly shutter Google Earth Enterprise in the U.S. along with some of the most important business features for Google Earth Pro such as data for parcels, demographics, and traffic counts. By the end of January it shutdown Parcel (APN) Search, sending those relying on the data for daily business requirements scrambling to find alternatives.

The decision, which Google would say was brought on by consumer behavior, continues to alter the future of a market segment.

Google isn't alone. Microsoft killed MapPoint last year. As more consumers leave their desktop for mobile phones, both companies, along with Apple, have been building up their mobile mapping applications, changing the business model, and shuttering data services and tools for desktop.

Now some believe that Google and Microsoft have turned their attention away from business mapping technology and toward mapping services that generate advertising revenue.

Despite the vanishing act brought on by the move to mobile, these traditional desktop business mapping tools that provided the ability to search on assessor parcel numbers (APN) remain popular, especially among industries that require easy-to-use software sitting behind their firewall. The model continues to change, along with the range of options.

Google's tools were affordable, flexible, powerful, and had tangible business benefits, according to Stewart Berry, director of product management for mapping software at Caliper, which develops geographic information systems (GIS) and transportation software. "These products have gone away," he says. "All that's left are APIs, with limited functionality, but at a higher cost."

Parcel boundaries are viewable using the Google Maps API, which provides the ability to style the lines to make them more prominent. Along with the API came new tools, but these are for developers, not consumers of business analysts, per Berry. "When you kill a desktop product and don't offer a comparable cloud product, you leave your customers high and dry."

While Google points to alternatives for parcel data like the Google Maps API, Berry says users cannot search by the parcel ID. The alternatives are for developers and have costs per interaction.

Having the ability to search by parcel number remains a huge issue for some industries such as real estate and construction. "I see the frustration in the removal of the APN layers, so I will offer what my company uses," writes Eric Maturino, an engineer offering advice in the Google Maps and Earth help form. Maturino works at a photovoltaic construction company running through permits and other services multiple times per week.

"At the price of $400 per county, this can be ridiculous, seeing at times we operate in six different counties a month," Maturino explains, suggesting a Web-based service called Parcel Quest Navigator.

The $1,800 annual cost may be steep, but it avoids the hassle of repeatedly buying parcel information and navigating the search. It is a drop in the bucket, he writes. Parcels and addresses are searchable if you know the number.


>>>>>>>>>>>>>>>>>>>>>>>>>><<<<<<<<<<<<<<<<<<<<<<<<<<

Find out which Google mapping tools will help you change the world

The tools below are available to nonprofits to help raise awareness, increase decision-making and engage stakeholders. Explore the free versions of these tools, and if advanced features are needed, learn about our software grants program.

Google Earth

Google Earth lets you fly anywhere on Earth to view satellite imagery, terrain, 3D buildings, from galaxies in outer space to the canyons of the ocean. You can explore rich geographical content, save your toured places, and share with others in this free desktop and mobile application. Google Earth Pro provides additional capabilities such as Movie Maker and GIS data import capability.

Read a case study
Take a tutorial
Learn more about Google Earth Pro

Google Earth Engine

Google Earth Engine brings together the world's satellite imagery and makes it available online with tools for scientists, independent researchers, and nations, allowing them to mine this massive warehouse of data to detect changes, map trends and quantify differences on the earth's surface.

Earth Engine overview videos
Take a tutorial


Google My Maps

Google My Maps allows you to create stylish custom maps to share and publish online. You can add lines, shapes, or places on the map, or import data from Google Drive spreadsheets, as well as Excel, CSV and KML files.

Take a tutorial


Tour Builder

Tour Builder lets you tell stories in their geographic context on the web using the Google Earth Plugin. You can easily create an immersive story with places, photos and videos, and share it with your friends.

Take a tutorial
Visit Tour Builder
See example tours

Open Data Kit

Open Data Kit (ODK) is an open-source toolkit that allows organizations to collect data offline, then bring that data online for sharing, querying and mapping. ODK Collect runs on Google Android - the first free, open source, and fully customizable mobile platform.

Read a case study
Take a tutorial

Google Fusion Tables

Google Fusion Tables is a collaborative data tool in the cloud. Upload data from spreadsheets, CSV or KML files. Find and reuse public data. Merge data from multiple tables. Visualize your data on customized maps and charts that update when your data does.

Take a tutorial

Google Map Maker

Google Map Maker™ allows you to map a neighborhood, a city or even an entire country. You can draw roads and landmarks, add local businesses, and edit locations added by other map makers. Your changes will appear instantly on Google Map Maker for everyone to see.

Take a tutorial

Google Maps API

The Google Maps API lets you embed the robust functionality and everyday usefulness of Google Maps and Google Earth into your own website and applications, and overlay your own data on top of Google Maps. Incorporate Google Maps API for Work into your internal website to map sensitive or secure data, or take advantage of the increased geocoding limits.

Read a case study about free Maps API
Read a case study about Maps API for Work
View Showcase
Apply for a Maps API for Work grant

Google Street View

Google Maps with Street View lets you explore places around the world through 360-degree street-level imagery. You can check out restaurants, plan your next trip or even visit the Amazon!

Read a case study
Browse the Gallery

Spreadsheet Mapper

Spreadsheet Mapper is a tool built by Google Earth Outreach to let you dynamically publish a Google Earth layer (KML or KMZ file) directly from a Google Spreadsheet. Create different balloon templates for different parts of your dataset in Spreadsheet Mapper.

Read a case study
Take a tutorial

Saturday, January 2, 2016


Meet the world's smallest camera-equipped drone

It's so tiny, users don't have to register it with the FAA.
Jessica Conditt @JessConditt10h ago in Robots


The Axis Vidius is a quadcopter that fits in the palm of your hand -- it's roughly 1.5 inches square -- yet it's able to livestream and record video in 420p. Axis says its Vidius model is the smallest-ever camera-equipped drone, and it's so tiny that users don't have to register it with the Federal Aviation Administration (meaning it weighs less than .55 pounds).

The FAA launched its online drone-registration program in December. It requires all pilots, even hobbyists, to register their robots by February 19th -- it will generally cost $5 per registration, but the FAA is waiving this cost through January 20th. Information in the registry will be public record. In the program's first two days, the FAA collected 45,000 registrations.

The teensy Vidius drone may appeal to those who don't want to deal with federal paperwork before flying, though it does have limitations. Pilots can stream a live video feed from the 'copter via WiFi, though it has a flight time of just five to seven minutes on a 20-minute USB charge. Vidius comes with a 2.4 gHz controller, and users can pilot it with a smartphone or tablet, as well. Vidius is up for pre-order now with a ship date of January 29th at the latest. It's on sale for $75 before January 7th; afterwards, it'll be $95.

A tiny drone could be a nice way to have a little -- and we do mean little -- fun.

Friday, January 1, 2016

CMOS versus CCD Video Sensors

A Brief Digital Imaging Primer

Digital imaging has become so ubiquitous as to be nearly invisible these days. Smartphones with cameras dominate the creation of digital images; YouTube claims that one hour of video is uploaded per second. We take all this for granted, but how much do you know about the underlying technologies?
Before I go any further, I’m going to issue a pre-emptive engineering-lawyer warning: Some of the stuff I say here is going to be a gross oversimplification, and some of it, almost (but not quite) flat out wrong, but useful as a model for understanding the concept.
CCD Sensor vs a CMOS Sensor
Images (left) courtesy Wikimedia Commons User Anderzej w k 2, CC BY-SA 4.0, and (right) courtesy Wikimedia Commons User Filya1, CC BY-SA 3.0

Types of Imaging Sensors

Imaging sensors are based on one of two technologies, CMOS or CCD. Each has its advantages and disadvantages, and with modern technology, it’s not really appropriate to say one is flat-out better than the other.
I’ll also point out that there isn’t really such a thing as a “color” sensor – all imaging sensors simply record the number of photons striking some region of an object during a given time frame, irrespective of the color of those photons. To get a color image, filters are laid over the individual sensor regions to exclude certain wavelengths from those regions, and then the resulting data is extrapolated from neighboring pixels, which each allow different wavelengths to be recorded. The variety of methods with which this can be achieved could be an article unto themselves; I’m just going to stick to un-filtered (grayscale) imagers for the rest of the post.

CCD Imaging Sensors

The first truly successful imaging sensors were CCD type. CCD stands for “Charge Coupled Device,” which isn’t really all that useful.
A very useful mental model for a CCD array is a line of people holding buckets, standing in a rainstorm. Let’s say you want to characterize how much water (light) is falling on an area. A way to do this would be to gather up, say, 300 of your closest, most patient friends, give all of them a bucket, 2/3 of them umbrellas, and have them form a rectangle 10 rows high and 30 columns wide. You’ll also need a megaphone. We’re also going to count off these poor shmucks, giving the ones without umbrellas the number one, then two and three, repeating.
In a CCD imager, there’s a very similar situation going on, where you have regions of the device that are exposed (no umbrella), and regions that are covered. The exposed regions gather light, turning photons striking the surface into electrons and storing them up during the “exposure” phase. During the “readout” phase, those electrons are passed along from one region (bucket) to the next until they reach the location where the analog-to-digital converter (ADC) is located, where the value is digitized. Here’s an excellent animation showing this process of scooting the charge across from one region to the next.
Animation of charge transfer in a CCD
Animation courtesy Wikipedia User Schmid CC BY 2.5

Drawbacks to CCD Sensors

Complexity - One of the major drawbacks of the CCD is that it requires substantial external circuitry to operate. Because the nature of the semiconductor materials required to realize a CCD is different to those used for other types of integrated circuits, all of the control circuitry required must be located on other chips. Frequently, in addition to the timing signals required to synchronize the transfer of charge from one cell to the next and the ADC required to digitize the data, voltages outside of the normal battery range for mobile devices (~3-6V) are required, so those voltages must be created by large, expensive additional circuits.
Smear - Smear happens when a bright light saturates cells on the imager. Its cause varies based on the readout type of the CCD (another very complex topic), and CMOS imagers aren’t susceptible to this type of artifact. You can see a good example of smear in this video of a Flaming Lips concert; compare that to this video of the same show, from a similar angle, but shot with a CMOS imager. You’ll see that the same light flashing causes a smear in the first video, but not in the second.
Thermal noise - Heat causes noise in CCD sensors, too; the warmer the image, the more free electrons there are just floating around in the semiconductor, messing up the image. Keeping an imager cool using thermoelectric coolers, liquid nitrogen, or other means is often used for high performance imaging applications. In particular, CCD sensors with aggressive active cooling find a great deal of application in super long exposure astronomy.

CMOS Sensors

Newer to the application is the CMOS imager. CMOS stands for “complementary metal-oxide semiconductor,” and unlike CCD, it’s not unique to imaging – most ICs these days are made with CMOS processes.
If we’re going to keep with our “field full of friends” mental model, a CMOS imager has nobody covered by an umbrella. You can see right away that this provides an advantage; the active area of the imager can be smaller for a similar pixel count than with a CCD. Also, instead of passing the electrons from cell to cell to a converter, the CMOS sensor multiplexes the ADC internally to each pixel. The method of charge creation is different, as well: in a CCD, the region capturing light also stores the charge developed by the photons hitting the surface. In a CMOS imager, the light strikes a conventional photodiode and the current released from that interaction is stored in a capacitor.
Because the CMOS imager is made using the same techniques as the vast majority of other chips, it benefits directly from advances in IC manufacturing. Another huge benefit is the ability to integrate the entire chip in one package, one wafer, one production line. This drives the cost for a CMOS solution way down, and the size, as well. The picture at the top of the post shows a CCD (left) and a CMOS sensor (right), for comparison. You can see the extra circuitry around the perimeter of the imaging section of the CMOS device, which is absent on the CCD.
Drawbacks to CMOS Sensors
Rolling shutter - The biggest drawback to CMOS imagers is that most of them use a “rolling shutter” method of capturing the image. This means that not all of the image elements capture data at the same time; basically, pixels are captured sequentially, with pixels next to each other being captured relatively closely in time to one another but pixels at the top of the image are captured substantially later than pixels at the bottom of the image.
Smeared image due to rolling shutter effect
Image courtesy Wikimedia commons user Jonen CC A-SA 3.0
In the picture above, you can see how the helicopter’s blades appear warped and misshapen while the rest of the image looks normal. The blades were moving fast enough relative to the capture times of the various pixels that different pixels caught the blades in different positions, creating the illusion of a physical warping.
Picture showing a partial exposure of a lightning strike
Image courtesy Wikipedia user Kalleboo CC A-SA 3.0
Here, you can see what happens when the illumination of a scene changes dramatically in the middle of image capture. Half of the image is significantly brighter because of the lightning.
Finally, the video above shows me playing with a Van De Graaf generator and a fluorescent bulb. Each time the charge builds up sufficiently to jump to the tube, the tube briefly lights up. As I turned my camera from landscape to portrait, you can see that the tube flashes all at once when it’s horizontally oriented, but appears to flash almost bottom to top when it’s vertically oriented. The horizontal pixels are captured fairly closely in time to one another, but from top to bottom, the timing varies greatly, creating the illusion that the light moves when it is, in fact, the exposure “window” moving across the lit tube.

Digital imaging is a very complex subject, of course, and this brief introduction really fails to do it adequate justice. If there’s suitable interest, perhaps I’ll do a larger, more in-depth tutorial covering the topic. If you’ve got questions, I’ll attempt to address them in the comments.

Sunday, December 13, 2015

Friday, December 4, 2015

Emerging Personal UAV/UAS Rules

General rules
  • Small unmanned aircraft must give way to all manned aviation activities: airplanes, gliders, parachutists, hang gliders, the Goodyear blimp, etc. If it flies or glides, it has the right of way.
  • The operator must remain within visual line of sight of the small unmanned aircraft. You can’t control or remain clear of other aircraft when you can’t see your own small unmanned aircraft.
  • Small unmanned aircraft may not operate over any persons not directly involved in the operation.