Algorithms Explained



For more than 40 years agronomists and scientists have used Normalized Difference Vegetation Index (NDVI) to assess the health of crops and plants. But what is NDVI? And how do you know you’re getting the real deal?


NDVI was created when scientists learned that a plant’s unique reflection of a combination of visible red light and near-infrared (NIR) light gave a good indication of plant health — far superior to what the reflectance of visible light can reveal. Simply put, NDVI is a single measure that characterizes this information. In precision agriculture, NDVI data provides growers with an accurate measurement of crop vigor and allows them to zero in on problem areas that may need further attention.


The short answer is that an NDVI sensor processes wavelengths of light outside of those captured by regular RGB cameras, like you have on your mobile phone. NDVI wavelengths are slightly longer than those of visible light and are located in the near-infrared (NIR) band — and that’s why a “regular” sensor can’t capture NDVI. When you hear of “False NDVI” this means the data is being captured by a RGB sensor and the wavelengths are being synthetically processed to simulate NIR and produce a “best guess” NDVI value. That’s like taking a black and white photo and guessing the colour of someone’s shirt. It’s a guess! Don’t bet your operation on it.


Some of the first sensors that could generate NDVI measurements were integrated into ERTS/LANDSAT satellites. Although the orbital altitude of a satellite means that the pixel resolution of satellite-based imagery is relatively low, the satellite-based sensors are usually extremely accurate and highly calibrated. LANDSAT sensors carried equipment that could sense in both the visual and NIR band, and scientists analyzing LANDSAT data created NDVI as an effective, easy-to-use measure of vegetative vigor.


In the last 40 years, advances in technology have allowed for integration of high-quality NDVI imagers onto manned aircraft and, now, multispectral sensors that integrate onto UAV platforms. And when combined with sophisticated NDVI analytics, and comparative data from previous flights, significant progress can be made in crop planning, optimising inputs and recognising crop health issues before they become major problems.

This is huge, not only for growers and agronomists, who can dramatically increase the efficiency of their operations, but for the world as a whole as we seek to address the growing burden of hunger caused by our burgeoning global population.



Understanding your crops’ health status isn’t the easiest thing to do. Sure, you can use the “eye test”, and a number of foliar contact and direct measurement techniques. But, what if there was an easy, fast and efficient way to see the health of plants and their status and progress over time? That’s where Normalized Difference Vegetation Index (NDVI) data comes in.


In simple terms, NDVI is a measurement of plant health based on how a plant reflects light (usually sunlight) at specific frequencies. To be more specific, NDVI is a measurement of the reflectivity of plants expressed as the ratio of near-infrared reflectivity (NIR) minus red reflectivity (VIS) over NIR plus VIS.

The equation for NDVI was developed several decades ago to make use of satellite imagery in agriculture. The way the equation is built makes it insensitive to overall brightness or darkness of light — it essentially tracks the ratio of NIR to red reflectivity, which doesn’t change with overall brightness.

NDVI works because when sunlight reaches a plant, certain wavelengths are absorbed while others are reflected. Chlorophyll strongly absorbs visible light while the cell structure of leaves strongly reflects near-infrared light. A spongy layer along the bottom of a plant causes these reflections. When a plant becomes dehydrated, sick, affected by disease, etc. this spongy layer deteriorates and the plant absorbs more of that near-infrared light rather than reflecting it. Conversely, when near-infrared light hits a leaf on a healthy plant, it is reflected back. So, looking at how NIR varies compared to red light provides an accurate indication of chlorophyll, which correlates to plant health.

The equation explained above will always result in a NDVI plant health value between -1 and +1. A number between -1 and 0 suggests an inanimate or dead object, like roads, buildings, or dead plants. A NDVI plant health rating between 0 and 0.33 indicates unhealthy or stressed plant material, 0.33 to 0.66 is moderately healthy, and 0.66 to 1 is very healthy. These numbers are just rules of thumb, and vary based on type of plant and other conditions. But that’s enough science for now.


As mentioned, NDVI plant health values are between -1.0 and +1.0 But how does this translate to the colorful NDVI maps you’ve probably seen? Basically, certain ranges of NDVI values are mapped to a set of colors. One of the most common color maps is the “red-green” NDVI color map. In this map, NDVI plant health values from -1 to 0 range appear red, 0.0 to 0.33 are orange-ish red or yellow, 0.33 to 0.66 tint green, and above 0.66 appear green. There isn’t a “standard” color map. Some people don’t like these colors, some people want more colors and some want fewer.

Here we see a stitched plant health map showing various NDVI values. The large red areas along each side are inanimate material (roads, dirt, etc.). Focusing on the outlined portion of the field, we notice NDVI plant health values ranging from green (good) to red (bad). By navigating to the two areas of concern within the field, the grower was able to identify weeds and washout. As a result, he applied the appropriate herbicide and adjusted irrigation measures to avoid yield loss. He was also able to identify thriving areas of his field (green) and reallocate inputs to boost his ROI.


Overall, NDVI is a way to measure plant health. Multispectral sensors detect indicators invisible to the naked eye, utilising light reflections and absorptions to calculate an NDVI score. Healthy plants absorb most of the visible light while reflecting a large amount of the near-infrared light. Unhealthy plants do the opposite. NDVI is an extremely helpful tool to assess plant health, and understanding it is important. Stratus Imaging can work with you to unlock even more value, by incorporating high-precision NDVI data right into most digital agriculture platforms.


  • Canopy coverage & density detection
  • Produces accurate growth trending with frequent use
  • Frost Damage Detection
  • Large Scale Pest Outbreaks
  • Optimizing crop rotation times
  • Ecological Benefits
  • Vegetation dynamics or plant seasonal changes over time
  • Biomass production
  • Grazing management & impacts (e.g., stocking rates)
  • Changes in range land condition
  • Vegetation or land cover classification
  • Moisture content in the soil
  • Carbon sequestration or CO2 flux

NDRE -Chlorophyll Mapping

NDRE uses a red edge filter to view the reflectance from the canopy of the crop. The red edge marks the boundary between absorption by chlorophyll in the red visible region, and scattering due to leaf internal structure in the NIR region. This allows you to determine many different variables with crop management. Understanding the levels of chlorophyll can provide you with the ability to monitor photosynthesis activity.

With this information you can optimize harvest times based on transitions of photosynthesis activity. During crop harvest events like: hull split in almonds or max sugar content in grapes, a noticeable change in NDRE values occur. This change occurs because sugar molecules produced from photosynthesis are no longer needed in such a high demand since the fruit/nut has reached maturity. This provides you with a crop management tool for harvest scheduling to have the highest quality produce.

Other factors that can change chlorophyll levels and cause crop stress are insect infestations. By utilizing NDRE you can determine how severe a mite outbreak is for an almond field and then use a precise way to terminate the infestation. This not only allows you to monitor outbreaks, but also reduce costs associated with pest control.

Advanced fertilisation methods have become essential to agriculture due to the increased price in fertiliser/fuel and increased restrictions. GNDVI displays and values provide you with a comprehensive way to apply your fertiliser. This means that fertiliser can be applied more precisely to target the areas of interest. This not only results in an economical benefit, but also an environmental benefit.



Normalized difference vegetation index (NDVI) imagery products have become increasingly common in precision agriculture applications. More companies are offering NDVI and NDVI-like products, including “False NDVI” images from drone platforms. At Stratus Imaging, it is our mission to provide you only the most valuable crop health data, which is why we use a multispectral camera. It’s important to understand how these products are different before choosing a remote sensing comapny to provide data.


The camera payloads that come standard with most consumer and professional drones, like the DJI Phantom and DJI Inspire systems, provide a phenomenal value! These cameras and integrated gimbals provide excellent stabilisation and deliver smooth, high-resolution video and high-megapixel still images.

The sensors in these cameras are built specifically to be sensitive to the same frequency bands of light that the human eye sees as color — within the visual red, green, and blue (RGB) spectrum. This is how these sensors, and any other RGB camera, produce images that can recreate almost exactly what our eyes would see.


Agronomists, horticulturist, and scientists have used NDVI for the last 40 years to assess the health of plants. A NDVI sensor uses wavelengths of light outside of what regular RGB cameras can detect. These wavelengths are slightly longer than those of visible light and are located in the near-infrared (NIR) band. Light reflected in the NIR band correlates to the biological processes in plant material. The combination of the non-visual NIR band and the visible red band results in a NDVI value. This value provides growers with an accurate measurement of crop vigor and allows them to zero in on problem areas that need addressing.


NDVI is based on the combination of red light and near-infrared light, period. A sensor that cannot develop a separate measurement of near-infrared light cannot produce NDVI.

An imaging product described as “False NDVI” is exactly that — not NDVI. These products are usually formed by acquiring normal RGB images and manipulating them in a way that hopes to approximate NDVI. But if the sensor hardware can’t measure and separate near-infrared and red light, the result can never be more than an approximation. To attempt to generate NDVI from a standard RGB drone camera is like trying to figure out the colour of somebody’s shirt from a black and white photo — the information just isn’t there.

NDVI was created because it provides a great measure of vegetative vigor. RGB imagery was available from the same satellites, but NDVI was created because it’s a better measurement of vigor than reflectance using any combination of RGB.

False NDVI’s only real attraction is that it’s cheap to build. It can be generated based on an RGB image taken from any camera. But it’s an inferior measure of crop health, and users should expect that analytics and prescriptions generating using inferior data to yield inferior advice.


For those seeking the most accurate and detailed NDVI data, a multispectral camera that collects NIR-band is the clear choice. Available with a 8 centimeter resolution with calibrated images