Every picture has a story to tell
More data from more sources, coupled with intelligent processing, unlocks limitless opportunities enabled by IoT.
Satellite and drone data as a source of information, whether it is combined with other source of information or not, can be of inestimable value. Satellites measure the entire planet and can provide insights into changes over time.
Drones allow people to photograph, video, map, survey and inspect a limited sized area e.g. a farm or a forest.
Drone and satellite imagery can all provide useful information, however the value of the information is highly dependent on the quality of the imagery. Drones support a higher resolution than satellite images.
SATELLITE AND DRONE REMOTE SENSING
How it works and what to expect.
The absolute basics:
Earth observation satellites are designed for applications such as monitoring and predicting climatic patterns, estimating crop yields, assessing damage during natural disasters, availability of water sources, and surface geology mapping. It works at a high altitude using remote-sensing techniques to collect data on the earth system’s chemical, physical, and biological aspects. This method also helps to manage the earth’s natural resources in a resourceful manner.
Basically it’s all about “Spectral Imaging Systems”. In Mangosat’s specific cases about drones and satellites in space tracking change on the Earth’s surface over time. Its fundamentals maybe a bit overwhelming but not that difficult to understand. The technology (indeed a kind of rocket science) behind it is of a different kind. We are just – and only – users, hence we leave the true complexity up to the experts.
At Mangosat our job is first channelizing information and discovering trends. Next is simplifying the models and creating ready to implement, easy to understand solutions which are affordable and accessible to potential users in the field. Users who are in their own bubble, not necessarily technical, and only interested in how we possibly can fix THEIR problems.
In short: our (optical) remote sensing satellites do not take “photographs” as we know it…well they do that too. “Our” satellites use different kinds of sensors to collect electromagnetic radiation reflected from the Earth. On-board passive sensors collect radiation which the Sun emits and the Earth reflects. On-board active sensors emit radiation themselves and analyse it after it is reflected back from the Earth to the satellite. That information is first stored and at regular intervals transmitted to a receiving station on Earth (aka gateway). Raw data is then processed into an useful image before shared with the end-user.
A bit more in depth: Spectral imaging refers to a group of analytical techniques that collect spectroscopic information and imaging information at the same time. The spectroscopic information tells us about the chemical makeup at the individual points of the image (pixels) allowing a “chemical map” of the imaged area to be produced. A variety of chemometric techniques are used to interpret the results. Analytical techniques include:
- Spatial resolution describes the number of pixels per square are unit.
- Spectral resolution defines the spectral variation a system can support.
- Radiometric accuracy defines how accurate a system is if about measuring the spectral reflectance percentage.
The difference between hyperspectral and multispectral imaging
Hyperspectral imaging collects spectral information across a wide range of wavelengths (colours) at the same time as recording the spatial information in an image. Hyperspectral images are 3 dimensional. The greater information provided from these wavelengths enables more detailed chemical information to be mapped, but requires more expensive instrumentation and more complex data processing than the simpler multispectral imaging.
Multispectral imaging collects spectral information from a limited number of wavelengths (colours) whilst simultaneously recording the image. These wavelengths can be customised for a particular application, which leads to cheaper instrumentation and simpler data processing, but without the flexibility of hyperspectral imaging systems.
The most used way to achieve spectral imaging is to take an image for each desired band, using a narrowband filters. This leads to a huge number of images and large bank of filters when a significant spectral resolution is required.
If you are interested in a little more detail and background than please click Landsat Satellites – a short but comprehensive tutorial on the Landsat series. Or go here: Fundamentals of Remote Sensing – tons of really good information from the Canada Centre for Mapping and Earth Observation.
What to expect from remote sensing satellites:
Remote sensing satellites orbit the earth year-round, snapping “photos” or better “images” as regularly as every week. Most commercial imagery falls between 1 and 5 meter resolution, with high resolution sensors capturing 70, 50 and 30 cm resolution.
<strong>What about resolution</strong>
Public satellite image services may be limited to a maximum spatial resolution of 26 cm to 30 cm, which is plenty for personal use. Commercial services go beyond that, with some offering 15 cm and up for the highest resolution sensors. The update frequency for these satellite images can vary depending on their original source. Public and commercial services update their imagery database once every few days, but it differs from company to company.
Since centimeters provide the best quality and details for satellite imagery, why don’t we use the highest resolution for every situation possible?
One technicality holds us back. Equipping and deploying each satellite with robust high-resolution sensors is expensive. Aside from that, the higher the resolution, the overall file size of the images increases. Maintaining and updating large satellite image files would be a monumental task, but it wouldn’t be impossible. As this technology advances, we will likely see even higher-resolution images from public and commercial satellite image services.
Good to know is that our planet have been mapped many times over the past decades which means that you can often find historical satellite images which provide a high-level assessment.
Illustrating example: Chernobyl, Ukraine 1986-2013
A nuclear accident devastated the region near Chernobyl, Ukraine, on April 26, 1986. Farm fields are the bright green and tan shapes in the 1986 image. Those farms have converted to natural vegetation, the flat gray-green in the 2013 image.
The USGS/NASA Landsat 5 image from April 29, 1986, was the first civilian satellite image of the accident. The data from Landsat were used to help confirm that an explosion had happened at Chernobyl and that the plant had been shut down. Click here for the source.
Why satellite images have “unnatural” colours:
While humans can perceive only a small portion of the electro magnetic spectrum (visible light), satellite sensors can use other types, like infrared light, ultraviolet light, or even microwaves. When satellite and drone images are made, these invisible types of light are assigned a visible colour. That is why satellite images often have “unnatural” colours.
Note: Many types of spectroscopy can be used for spectral imaging. Remote sensing of the Earth operates in the visible and short-wave near infrared regions, whilst applications for precision agriculture are within the NIR region (780–2500 nm).
How useful are satellite images:
Satellite images are useful because different surfaces and objects can be identified by the way they react to radiation. For instance, smooth surfaces, such as roads, reflect almost all of the energy which comes at them in a single direction. This is called specular reflection. Meanwhile, rough surfaces, such as trees, reflect energy in all directions. This is called diffuse reflection. Also, objects react differently to different wavelengths of radiation. For instance, there is a frequency of infrared light which can be used to determine plant health. Healthy leaves reflect this frequency well while unhealthy ones do not.
<strong>More about spectral imaging</strong>
Spectral imaging is the detection of light reflected by an object or e.g. a crop with the use of specialized sensors. It is measured in spectral bands. The higher the number of bands the higher the accuracy, the flexibility and information content.
Currently satellites and drone fitted cameras are using multispectral imaging technology. Multispectral technology (5–7 bands) can offer a good general overview, but fails to do so if more detail or a higher level of accuracy is required. Hyperspectral technology with its higher number of spectral bands (40+) can provide solutions for almost every problem encountered in the field.
More spectral bands result in a significantly higher information content of the data acquired. High spectral resolution of data allows for detection and identification of for example inferring biological and chemical processes in crops, which opens up a full range of applications in precision agriculture.
Technology is fascinating!
Free sources for raw satellite images:
All the free options below offer different spatial resolutions for satellite or aerial imagery, depending on its use case. So if you want to discover satellite imageries from all over the world. you’ll love this list. And also all are downloadable!
According to the official website, the European Space Agency (ESA) operates missions under Sentinel-2, which consists of two polar-orbiting satellites. They track the Earth’s surface for changes, including climate change and land monitoring. You can locate the satellite data via their Copernicus Open Access Hub. The highest spatial resolution they offer goes up to 10 meters. Downloading the raw data offline or from the cloud via their official website is also available.
Landsat 8 and 9
USGS and NASA created a partnership for the unique Landsat missions. Landsat 9 is the newest, but it also offers raw data you can view from the older Landsat 8. Located on the official Landsat Data Access website, you can download this information offline or use cloud-based services. They offer a spatial image resolution of up to 15 meters, but 30 meters is the most common for each use case.
MODIS is short for Moderate Resolution Imaging Spectroradiometer and is owned and operated by NASA. It offers a detailed view of weather events worldwide, including climate change and cloud cover. The highest spatial resolution is 250 meters, more than enough to track these weather events. You can download the raw data from their official website using the Level-1 and Atmosphere Archive & Distribution System.
An open source collection of aerial imagery, OpenAerialMap allows anyone to upload and contribute to the database as a professional. Once you interact with the OpenAerialMap viewer, you can find a source file that interests you. From there, you can download its raw data instantly. Since this is a community-led effort, the spatial resolution varies, but it can sometimes go up to 1 cm. New sources are constantly added to the database for you to check out.
The National Agriculture Imagery Program (NAIP) is headed by the United States Department of Agriculture (USDA) under the Farm Service Agency (FSA). The NAIP uses aerial imagery to offer insights during peak agriculture season, helping federal farm programs and general education about land changes. They provide a 1-meter spatial resolution for satellite images but may also use centimeters for other use cases. You can visit the official Geospatial Data Gateway to download the raw data.
USGS Earth Explorer – is the one of best satellite imagery source. And it’s not just for the United States. You can choose and download any imagery.
Paid sources for raw satellite images:
All the paid options below offer at least a 1-meter spatial resolution for satellite or aerial imagery.
Maxar Imagery Mosaics
The most comprehensive commercial satellite imagery provider, Maxar Imagery Mosaics, offers high-resolution, virtually seamless images, as the official website states. It creates imagery basemaps with unrivaled accuracy to provide the best experience possible. Used to monitor borders, plan future expansion for land development, or track land changes across the globe, Maxar is perfect for anyone looking for robust satellite image tools. You can learn how to download the raw data offline through their official Imagery Download website.
Airbus OneAtlas is another massive commercial-based satellite imagery provider offering high-resolution basemaps to view and download. From 3D textured models and radar tasking to viewing multi-resolution images via the Living Library archive, Airbus OneAtlas has everything you need. You can even request a specific day to communicate with the satellites to create a custom dataset tailored to your needs. You can learn how to download the raw data from the official OneAtlas Data Download Service website.
Hexagon Geospatial provides high-resolution aerial images taken by aircraft during the best weather conditions. This allows it to get the cleanest and highest quality data possible for aerial satellite imagery. Since aircraft are closer to the ground than satellites, you can expect highly detailed images with even greater accuracy. It also takes 45-degree shots to reveal more structural and building details. You can learn how to download the data and software from their official Customer License Portal.
With Vexcel Imaging, you can access high-resolution satellite imagery that can be used to track natural disaster responses, different building or land attributes, and smart city planning with 3D modeling. It’s also possible to accurately measure a roof’s surface area to get a repair estimate from anywhere in the world. Everything is cloud-based and powered by UltraCam technology, allowing for easy future expansion of the Vexcel service. Check out the Vexcel Data Program to access more specific data. The official website says it’s “the world’s largest aerial imagery and geospatial data program.”
EagleView is an aerial imagery provider offering high-resolution satellite images for various use cases. It can be used for construction purposes or building planning, big government design decisions, and can help insurance companies make property claim estimates from anywhere globally. On top of that, you can get accurate roofing measurements via satellite images for easy solar panel installation on any property. More features are available than the few mentioned, making EagleView a one-stop shop for all your aerial imagery needs. You can get started by visiting their official Place an Order website.
Another big player in the aerial image provider game, Nearmap Orthographics, provides much higher-quality images than traditional satellite images. As it put its, businesses and government officials can “conduct virtual site visits” using Nearmap 3D city models without leaving the office. Since Nearmap is a cloud-based platform, it integrates everything you need via a web browsing experience. You also have the option to bring the data into another imaging or CAD platform as you see fit for offline use. To get started with Nearmap Orthographics, you can get a quote today.
What is the market opportunity and what are the cost for using satellite imagery:
Due to this growing demand for the availability of information, there is a global market for satellite applications that both governments and commercial parties are interested in. At the same time, both governments and a growing number of commercial parties are the suppliers of satellite data, and the quantity and quality of the available satellite data is rapidly increasing. This creates opportunities for developers and suppliers of services, but also for governments and markets to make use of the services.
more to come
What about drones.
Drone data, be it RGB, multispectral or hyper spectral, have a higher spatial and temporal resolution than satellite data and is cheaper than manned-aircraft flights for the same data.
Drone data output depends on the sensor and processing software. Select a sensor based on your data requirements. Your sensor choice will determine the drone you need to carry this payload.
Satellites and drones compared:
Satellites are recommended to cover large amounts of spatial temporal data to identify trends but do not offer the same level of precision as other tools such as drones or sensors on the ground.
Drones with multispectral imaging sensors are the recommended data source for the purposes of precision farming for specialty crops. Drone-acquired imagery has >100x resolution than plane-acquired imagery and the flights can be scheduled according to the phenology of the crop as well as ideal conditions for accurate readings. The higher resolution allows for segmentation of the tree canopy only, including immature trees, and excludes all noise.
“Instead of just seeing a plant, hyperspectral imaging can go deeper and see the chlorophyll content of the plant, or the composition of the soil.
Mangosat service offerings
Mangosat offers a total concept solution, which is the combination of image processing and on the ground IoT solutions for overall farm, tree, and on-tree fruit monitoring. The system is flexible and expandable and basically every sensor can be included. Our concept has low economic investment and is not subject to geographical restriction. Moreover, your farm can be efficiently managed from anywhere in the world.
If required we take care of electrical power, connectivity, implementation, and maintenance.
Collecting and combining data
Analysing the moment or looking backward
Detecting change and prediciting future development
Collecting and processing
Satellite data knows no boundaries. drones have a higher resolution, ground sensors are very precise. Together with our partners, we collect and combine all available data in one point for further processing.
The moment or the past
We have both near real-time and historical data available, allowing greater accuracy in monitoring. Powerful AI algorithm generate alarms, do predictions and facilitate decision making processes.
Useful and understandable
Raw data is not what we want to present to our customers and users. Our data is converted into handy dashboards or individual maps so that everyone can implement the data in his own preferred way.
What to expect?
We provide you with our data in all possible formats. CSV and TIFF files to get started right away.
You will receive a report with explanation of our findings. Including maps and graphs for the right insights.
Access your data anytime, anywhere. Your own online dashboard, with your own data.