One of major buzzwords of precision agriculture is the NDVI, short for the Normalized Difference Vegetation Index. It’s constantly thrown around – but do we really know what it means and how valuable it truly is? Let’s dive into one of our most popular algorithms and learn why it is such a powerful piece of information to have on the field.
Some History and Theory
The NDVI is far from new. Compton Tucker published a landmark paper in 1979 regarding the development and use of the index in satellite remote sensing [1]. By the, 1980’s researchers would begin to use the NOAA’s AVHRR satellite to collect data and [2] to map dense green vegetation throughout the entire world. Even then, various use cases were already evolving. The NDVI helped scientists develop things like famine alarm systems and it was used to improve climate models. Until today, research on correlates of the index continue to get published.
So how did Tucker come up with the NDVI [3]? He based it on the fact that objects reflect and absorb various wavelengths in different amounts. Colors emerge because of the interaction between objects and light. Plants are special because light is an integral part of their lives. They have been found to absorb a lot of visible light, an energy source they use in photosynthesis.
Conversely, their cell structure causes them to reflect a lot of near infrared (NIR) radiation. By comparing the reflectance of visual and NIR radiation, therefore, one can get a good idea of how dense the vegetation in an area is. This led to the now ubiquitous equation:
Mathematically, values should range from -1 to +1. In practice, the absence of vegetation in bare soil conditions typically result in a value close to zero while areas of dense vegetation will have values close to +1.
Putting it into Practice
Fast forward to today. From its beginnings as a tool developed for use in satellite remote sensing, the NDVI has now become an integral part of precision agriculture because of the level of insight it provides. Apart from its ability to discern the density of vegetation, the NDVI has also been found to be correlated to key agricultural outcomes such as crop yield [4,5], crop identification [4], leaf area index [6], and even plant health [7]. On the field, the NDVI will work best as a part of a robust agricultural data stack, one that brings together important information on various data points like plant height, soil type, soil moisture, and yield expectation. The company is currently working on developing the key elements of this data stack both internally and through its growing network of external partnerships.
Integrated with a smart UAS platform like PrecisionHawk’s Lancaster, the NDVI becomes an even more powerful tool. It combines the speed by which one can deploy a drone to collect data with the richness of insight that the index inherently has. All these considered, the NDVI serves as a useful supplement or even substitute to traditional scouting. Imagine being able to quickly determine problem areas in the field based on regions with low NDVI values instead of relying on time-consuming practices that are also prone to human error.
The ease and availability of data also means that you can then quickly deploy field teams to examine specific problem areas, allowing them to diagnose issues more efficiently rather than spending time and resources scouring the field for answers. This makes time-sensitive events like the onset of certain diseases much more manageable. In the end, all these will benefit your bottom line–you spend less time and less resources while getting as much or even more insight than traditional agricultural methods.
One Index among Many
One very important thing to consider is that the NDVI may not always be the best vegetation index to use. Depending on the type of crop or the growth stage, other indices may be more appropriate. Fortunately, the AlgoMarket has you covered here as well. We currently offer a wide suite of vegetation indices to respond to the diverse needs of our clients. As we integrate new algorithms and expand our network of university partners, you can be sure that will we be offering even more vegetation indices to give you the most incisive data.
[1] http://www.washingtonpost.com/wp-dyn/content/article/2009/08/03/AR2009080300357.html
[2] https://earthobservatory.nasa.gov/features/MeasuringVegetation
[3] Tucker, C.J. (1979) ‘Red and Photographic Infrared Linear Combinations for Monitoring Vegetation’, Remote Sensing of Environment, 8(2), 127-150.
[4] Bolton, D. K., & Friedl, M. A. (2013). ‘Forecasting crop yield using remotely sensed vegetation indices and crop phenology metrics’, Agricultural and Forest Meteorology, 173, 74-84.
[5]Huang, J., Wang, H., Dai, Q., & Han, D. (2014). ‘Analysis of NDVI Data for Crop Identification and Yield Estimation’, Selected Topics in Applied Earth Observations and Remote Sensing, IEEE Journal of, 7(11), 4374-4384.
[6] Viña, A., Gitelson, A. A., Nguy-Robertson, A. L., & Peng, Y. (2011). ‘Comparison of different vegetation indices for the remote assessment of green leaf area index of crops’, Remote Sensing of Environment, 115(12), 3468-3478.
[7] Mahlein, A. K., Rumpf, T., Welke, P., Dehne, H. W., Plümer, L., Steiner, U., & Oerke, E. C. (2013). ‘Development of spectral indices for detecting and identifying plant diseases’, Remote Sensing of Environment, 128, 21-30.
Did you like this post about index NDVI? Read more and subscribe to our monthly newsletter!
#
Next article
For many years, remote sensing was directly connected to collecting data with satellites and manned aircraft. Originally this was a truly game changing technology, but it had its limitations. These data gathering technologies, while effective, can be expensive and, in many cases, time inefficient.
Traditional remote sensing multispectral images
Over the years market needs have evolved and today’s decision makers require hyper accurate, hi-resolution data in a near real-time, which is difficult to achieve using traditional methods. This is where drones, previously used for military reconnaissance, entered the remote sensing arena. UAV flights can be conducted daily, for smaller areas, at low altitudes, resulting in higher resolution imagery at a fraction of the cost..
PrecisionHawk – North Carolina-based start-up was one of the first who realised the potential of drones in remote sensing. In 2008 they started flying their first aircraft, The Lancaster, for commercial applications in Canada, specifically in the viticulture industry. The vision behind it was however much different from most of the players on the market. PrecisionHawk understood early on that the use of small UAVs goes far beyond data collection; the key is turning around actionable information, so a platform needs to include data processing and analysis. In the past, these functions were separated. Data was often collected by one company and then handed over to another organization for processing and analysis.
PrecisionHawk took a different approach and decided to build an end-to-end solution that did not require a pilot or remote sensing expert to operate and understand, but would allow an average grower to gather field data in the matter of minutes. Five years ago the idea seemed crazy, but today the approach introduced by PrecisionHawk is recognized as industry standard.
All you need is a tablet or laptop with the dedicated map platform where you mark area you need to survey. Than you simply throw the plane into the air. Everything else is done automatically. The aircraft computes flight paths, survey parameters, take-off and landing paths on its own. Once the survey is complete, the on-board computers will automatically connect to Wi-Fi networks and transfer all remote sensing data, flight information and diagnostics to remote servers, which can be accessed via its DataMapper platform. Moreover the drone sensors are fully customisable. Depending on your needs, you can buy extra sensors like Lidar and just plug it in. Sounds cool?
That’s not all. The PrecisionHawk software platform, DataMapper, is used for storing, processing and analysing remote sensing data features a really unique thing – a marketplace where a user can buy and sell your remote sensing algorithms. It is one of the first places where a user can easily commercialize it’s analysis tools and developments. But this is not the only area where the company is taking the approach to create an aerial information ecosystem. In early 2015, PrecisionHawk acquired TerraServer – a popular web portal to buy satellite images. In the future using TerraServer technology you will be able not only to buy satellite imagery, but order drone services from PrecisionHawk and other companies to get a higher-resolution picture of your desired location.
But building the end-to-end drone platform is just a beginning. In 2014 PrecisionHawk raised $11m in seed funding with the aim to go beyond being just an outstanding drone start-up. The company developed the first, automated air traffic control system for drones called LATAS (Low Altitude Traffic and Airspace Safety) to help solve the safety issue presented when integrating drones into the airspace with competing obstacles.
The existing air traffic control system almost fully relies on ground radars. It works well with regular aircrafts, but small drones, flying at low altitudes are almost impossible to be detected. Besides, any system of human operators could not possibly scale to accommodate the millions of drones expected in years to come. LATAS on the other hand uses cellular and satellites technologies to manage millions of simultaneous connections between drones and other ground and air obstacles. By relaying on existing infrastructure the platform has the ability to scale and to accommodate the millions of drones expected in years to come.
The aim of the project is to safely integrate drones into the national airspace, and it is being tested together with United State’s FAA under the Pathfinder program.
From PrecisionHawk perspective LATAS is a strategic project as the safety requirements are still a key barrier for the industry. This doesn’t stop the company to expand from agricultural data collection business to new industries including construction, insurance and energy among other. Today PrecisionHawk’s client base includes several Fortune 500 companies in the US, Europe and Asia. Not to shabby.
“A million-dollar idea” for a start-up needs to have a clear vision which either solves an existing problem or generates a new desire. PrecisionHawk is a model example of that sort of thinking. The company’s founders had a clear vision and found a proper people to make it happen. Today PrecisionHawk is one of the industry leaders and it sets standards for everyone else.
Did you like this post? Read more and subscribe to our monthly newsletter!
