MAIA M2 is the new lightest modular multispectral camera

The potential offered by RPAS (Remotely Piloted Aircraft Systems) in environmental prevention and monitoring is related to the possibility for sensors to fly over areas of interest. In the last decade proximal sensing technology saw a great development both with regard to sensors (lightweight multispectral and iperspectral sensors) and platforms (aircraft, helicopters, RPAS). Research and development in photogrammetric and multispectral surveys is offering new innovative solutions in sensors and technologies to monitor our environmental resources with very high frequency, precision and reliability.

MAIA is the most advanced multispectral camera designed to be employed onboard UAV systems, airplanes and terrestrial rovers as well, jointly developed and made in Italy by SAL Engineering, that designs and manufactures systems for data acquisition in sea, air, land environments, EOPTIS, specialized in designing and manufacturing opto-electronic measurement instruments, and 3DOM Research Unit of Fondazione Bruno Kessler, that is actively involved in accurate measurements and reality-based 3D reconstruction issues. In this team the Italian excellence in the fields of physics, optics, geomatics, 3D modeling and remote sensing have been concentrated: a consolidated know-how was made available for the construction of a multispectral imagery acquisition instrument that could ensure scientific rigor and total control of geometrical and radiometric data for a correct multispectral survey.

Regarding the differentiation of wavelenght intervals along the electromagnetic spectrum, MAIA has been designed according to two main sets: MAIA WV and MAIA S2. Nevertheless, thanks to the profitable collaboration with agronomic consulting companies and environmental protection agencies, or with universities and research institutes, a fully customizable modular solution was subsequently developed.

MAIA M2, in fact, is the new modular multispectral camera that the user can customize with a large portfolio of VIS-NIR bandpass filters, according to his needs.

The MAIA M2 single module can be composed using a pair of available band-pass filters. The choice of the pre-selected filter pairs will be made according to the most widely used multispectral indexes with two single bands, or on the basis of the aim of the multispectral survey. In the following table you can see the selected filters that are available in stock:

Each module has stand-alone capability with external trigger and strobe or free run mode, and presents several inputs/outputs for external devices interfacing such as trigger, strobe, serial port, USB and two aux port. The module/camera is based on a double global shutter CMOS sensor with 8/12 bits resolution and automatic exposure with selectable target value.

Single module of MAIA M2 has the lowest values in the market of modular multispectral cameras in terms of size (48 mm X 33 mm X 23 mm), weight (70 g) and price (1990 € until July 15th), but it presents the highest values in terms of resolution and sensitivity of sensors.

Multi-module management, up to 8 modules, is possible using the external MAIA M2 Control Unit that manages the images synchronization and geo-referencing, the powering of modules, the reading of PWM inputs, the light sensor input, two outputs with customizable variable advance for delay compensation of any connected DSLR cameras. An RTK version of MAIA M2 Control Unit is supplied including the GNSS antenna, the UHF antenna, the Lux Sensor and the connection cables for batteries, PWM inputs and DSLR shutter input. Multispectral raw images and parameters are stored in a removable SD card, and they can be downloaded from USB in order to be pre-processed with MultiCam Stitcher Pro, the MAIA images pre-processing software.

The following table shows some combinations of 2 or more MAIA M2 modules, useful to allow the calculation of many of the main multispectral indexes:

MAIA is basically the proper instrument for your multispectral survey.

You can detect VIS-NIR informations through 9 global shutter sensors with high resolution and top sensitivity; next, you have the total control on creating your dataset of undistorted and geometrically corrected images for reflectance analysis, indexes calculation and photogrammetric processing. You can then make decisions on monitoring crops, wineyards, forests and coastal environments, in order to safeguard ecosystems and to make your agronomic system more efficient. Along with the camera, an image processing software will be provided for correction of geometric and radial distorsion, for coregistration (pixel-pixel convergence) of RAW multispectral images acquired with MAIA, with tools for indexes calculation and for band combinations.

Since its foundation, SAL Engineering has participated, contributing with the design and management of data acquisition, synchronization and processing systems, to several projects with agronomic and precision farming companies, or environmental protection agencies that deal with natural environments such as coastal dunes, forests, reclaimed sites, areas with high environmental risk.

SAL Engineering is a company specialized in photogrammetric surveys based in Italy: visit our website www.salengineering.it.

For any further information about services and products, send us an email at info@salengineering.it.

For any detailed information about products and technologies that deal with multispectral surveys, please contact us. SAL Engineering is providing accurate multispectral data to companies specialized in agronomic consulting thanks to our integrated systems based on platform, control system and sensors.

MAIA WV is the multispectral camera equipped with the same wavelenght intervals of the WorldView-2™ satellite owned by DigitalGlobe. Now, you can compare satellite data with high-resolution maps obtained through a multispectral survey conducted with MAIA WV mounted on your UAV, getting centimeters-level precision and accuracy. WorldView-2™ is a commercial earth observation satellite that provides eight-band multispectral imagery with 1.84 m resolution, in support of services such as agriculture, forest monitoring, land cover changes and natural disaster management. MAIA WV multispectral camera is based on an array of 9 sensors (1 RGB and 8 monochrome with relative band-pass filters) to detect multispectral imagery in the VIS-NIR spectrum from 390 nm to 950 nm: MAIA WV is the most advanced broadband multispectral camera for RPAS, aircrafts, terrestrial rovers available today, with bands in Coastal and Blue spectrum region.

MAIA S2 is the multispectral camera equipped with the same wavelenght intervals of the European Spatial Agency‘s Sentinel-2™ satellite. Sentinel-2™ is an earth observation mission developed by ESA as part of the Copernicus Programme to perform observations in support of services such as precision agriculture, forest monitoring, land cover changes detection, and natural disaster management. Now, you can compare free satellite data with high-resolution maps obtained through a multispectral survey conducted with MAIA S2, the multispectral camera with two narrow spectral bands both in Red Edge and in NIR region.

The key features of the new-born MAIA M2, that presents the same quality of sensors, filters and optics of the standard versions WV and S2, are basically linked to the modular system and to the freedom to customize the set of bandpass filters, with excellent cost/benefit ratio and perfect physical adaptability onboard data acquisition platforms.

Mapping of anthropogenic stress in vegetation and soil

Techniques of acquisition, processing and interpretation of multispectral data related to key environmental processes such as chlorophylline photosynthesis and plant nutrition, have been refined, especially in relation to the identification and mapping of anthropogenic stress caused by soil infiltration or dispersion of polluting material on the surface. Recent research suggests that there is a distinction recognizable by data acquired through multispectral relief, between natural stress due to, for example, drought and an induced or anthropogenic stress due to soil contamination: this difference is visible in a different physiological response of plants (Zinnert and others, 2012).

A series of computations between multispectral bands, known as vegetation indices, have been developed and applied in agronomy and environmental sciences to optimize information from multispectral data, which has an increasing geometric, spectral, radiometric and temporal resolution since new acquisition technologies such as RPAS and new sensors have seen significant technological development over the last decades (Thenkabail, 2000).

For environmental and agronomic applications, the bands most involved in the calculations are Red (630 nm to 690 nm), Red Edge (705 nm to 745 nm), NIR (750 nm to 950 nm), Green (525 nm to 575 nm) and Violet (390 nm to 450 nm). The main vegetation indices are the Normalized Differentiation Vegetation Index (NDVI) and its optimizations or transformations such as the Green Normalized Differentiation Vegetation Index (GNDVI), Soil Adjusted Vegetation Index (SAVI) and also in this case including its optimizations such as the TSAVI (Transformed Soil Adjusted Vegetation Index) or the Modified Soil Adjusted Vegetation Index (MSAVI). A very important index for evaluating water content and quality in vegetation and soil is NDWI (Normalized Difference Water Index). By applying the multispectral survey to environmental monitoring of soil matrix or vegetation matrix, these indices are useful for defining growth rates and the vegetative quality of leafy vegetation.

The study of one of these spectral bands, the Red Edge, allows specifically to classify vegetation contaminated by presence in the soil by inflow or gaseous hydrocarbon suspension.

To apply this method of analysis, it is necessary to first correctly classify the vegetation present on the soil and evaluate the moisture content, the species present, the vegetation cover, the leaf cover index, and the surface temperature in different lighting stages and in different seasons. Vegetation growing in soil contaminated by hydrocarbons has visible damages in the Red Edge band, which is the name given to the sudden change in the spectrum region ranging from 680 nm to 730 nm and is caused by a combined effect of a strong incident radiation absorption and strong inner reflection and scattering of the leaf called “leaf internal scattering”.

The shift in the Red Edge reflection of vegetation, which indicates a reduction in plant health or a stress condition linked to anthropogenic contamination, has long been studied and applied to the agronomic and environmental study of cultivated or vegetated areas.

Important tests in this field were carried out by our R&D team on a soil with heavy hydrocarbon contamination: applied remote sensing outputs that were compared in GIS environments were the multi-spectral orthophoto of the 8 bands of MAIA WV, one thermal orthophoto obtained with high resolution thermal camera and an RGB orthophoto always obtained with MAIA’s RGB sensor. With regard to site investigations in collaboration with environmental engineers to which these products serve as mapping basis for macro-assessment of contamination problems, it was possible to identify and map vegetal anomalies related to the presence of hydrocarbons in the soil, also found in a variation in surface temperature, as well as in a different reflection in the NIR band and, as previously mentioned, particularly in the Red Edge band, and in some indexes.

Figure 1: Detail of the Red Edge band image of a terrain with hydrocarbon contamination.

Figure 2: Detail of a thermal orthophoto of a terrain that is contaminated by hydrocarbons.

Figure 3: Detail of a RGB Orthophoto of a terrain that is contaminated by hydrocarbons.

The field activity is supported by valid scientific publications (Noomen, 2003 & 2008), in which the high-quality method is certified and in which there is evidence of high susceptibility of crops in the presence of gaseous hydrocarbons deposited on the ground or surface stagnation, and more specifically in the presence of Ethane gas (C2H6): cultures exposed to this gas are more spectrally reflected in bandwidths ranging from 570 nm to 700 nm (Noomen 2008).

Thanks to the multispectral survey made with MAIA WV, it is also possible to detect an accentuated concentration of metals in the soil matrix. Wu et al. (2007) demonstrated that spectroscopy in visible and near infrared regions has a strong negative correlation with certain metals (Cadmium, Chrome, Copper, Mercury, Lead, Zinc) in contaminated soils, depending on the presence iron oxide and carbon content. Chloe et al. (2008) and Wu et al. (2008) continued fruitful research on the use of multispectral remote sensing to diagnose the presence of high concentrations of certain metals in contaminated soils.

Other scientific studies (Asmaryan et al 2014) confirm the positive correlations between the presence of chromium, lead and zinc measured in the site and detected by multispectral survey, on non-vegetated soil (whose NDVI index by definition goes from 0 to 0.3). The same studies have identified specific reflectance values at certain wavelengths of certain metals in the soil matrix, as shown in the table.

Figure 4: Correlation in different spectral ranges between metal content in non-plant soil and spectral values derived from a WorldView-2 satellite image.

Concerning the multispectral knowledge of the ground matrix, MAIA WV images, related to the wavelength ranges of satellites investigating the spectrum from visible to infrared such as Landsat TM, World-View 2 and Sentinel-2, can be processed to locate and map a large set of minerals, including iron oxides, clays, and other hydroxyl minerals that are often in nature at hydrothermal alterations in the outcrops (Source: Andrea G. Fabbri, Gabor Gaál, Richard B. McCammon, Deployment and Geoenvironmental Models for Resource Exploitation and Environmental Security, Springer Science & Business Media, 2012).