HyperSlit: an ultralight hyperspectral UAV system for collecting plant spectral signatures while recording sky spectra


Send your question(s) for discussion within the Q&A!

In order to provide meaningful discussions in the Q&A with Hamed Mehdipoor, you can send us your questions & we will cover them during the Q&A on July 28th, 14:00 CEST.

Simply send them by mail: ippn@plant-phenotyping.org


Click here to join live Q&A session on 28 July, 14:00 CEST


Hyperspectral remote sensing (HRS) is becoming a common technique to gain an insight into plant beyond the human eye. This is because HRS is a powerful proxy for the estimation of plant parameters (Bannari et al., 1995; Zhang et al., 2011). For example, plant water and chlorophyll content can be accurately derived from HRS measurements (Penuelas et al., 1997; Haboudane et al., 2002). Moreover, indices linked to diseases and photosynthetic apparatus can be retrieved from plants spectral signatures (PSS) (Delalieux et al., 2009). The PSS is defined as the variation of reflectance[1] or emittance of a plant with respect to wavelengths. Therefore, HRS measurements are of high interest for plant biometeorology. For this purpose, calibrated, portable and non-destructive HRS systems are highly demanding to collect PSS over large areas.

Several biometeorological studies have been conducted by HRS systems that are mounted on unmanned aerial vehicles (UAV)(Delalieux et al., 2009; Primicerio et al., 2012). This is because UAV based HRS systems are tempting due to their great degree of automation and fast throughput (Suomalainen et al., 2014). The recent development of small and lightweight UAVs such DJI Mavic series (e.g., diagonal size is less than 40cm and weight is less than 1kg) offer affordable (about 1000 $) and stable flying platforms for HRS systems. However, there is no small HRS system that fits these UAVs to collect PSS while recording sky spectra. Sky spectrum is dynamically changing because the atmospheric condition is constantly changing. This makes the calculation of reliable reflectance information from plants challenging (Richter et al., 2002; Gao et al., 2009). To address this, in this Webinar we present HyperSlit system solution.

[Full Abstract]

Speaker Information

Short Bio:

Hamed Mehdipoor is a Senior designer and developer focusing on design and develop of technological solutions for precision agriculture and environmental protection. Hamed’s expertise is in remote sensing, artificial intelligence, and high-performance cloud computing. He holds a Ph.D. in Geo-AI and plant phenology from the University of Amsterdam, and an GIS and remote sensing engineer diploma from the University of Twente. Since 2018, he founded the Spectro-AG knowledge-based company for Research and development. Hamed is involved in several national and international project using AI and drone technology for precision agriculture and forestry. since 2017, he has been focussing on the estimation of grass traits from sky and space. He has designed and developed a 3D NDVI solution which is used in forest and rangeland monitoring in Germany, the Netherlands, and Afghanistan. Hamed was the chair if SNP group of the international society of biometeorology from 2017 to 2020 where he lead new professionals and PhD student for biometeorology related group activities. 


Most relevant Publications:


Anderson, V., Leung, A. C., Mehdipoor, H., Jänicke, B., Milošević, D., Oliveira, A., ... & Zurita-Milla, R. (2021). Technological opportunities for sensing of the health effects of weather and climate change: a state-of-the-art-review. International Journal of Biometeorology, 1-25. https://doi.org/10.1007/s00484-020-02063-z


About Spectro AG:

Spectro-AG was found on 2018 in the Netherlands with focus on the design and development of AI-based solutions that use remote sensing data for Unmanned Ground Vehicle (UGV), Unmanned Aerial Vehicle (UAV) and satellite systems. Initially the company was established to develop solutions for detecting objects of interest for farmers, environmentalist, and inspectors in the Netherlands. For this purpose, we leverage sensors and artificial intelligence, including machine learning and deep learning algorithms, to generate new knowledge and solutions for various application domains including Precision agriculture, Environmental monitoring, Infrastructure inspection, Ground and Aerial surveillance. Our mission is to build ready-to-use software and hardware products for object detection and parameter estimation that users with minimum expertise in AI and remote sensing can still benefit from. Within these markets Spectro-AG develops, manufactures, and commercializes AI and remote sensing systems for:

  • Farmers (e.g., AI systems for real-time estimation of parameters, as well as detection of weeds in grasslands using multi- and hyperspectral sensors)
  • Forest inspectors and managers (e.g., AI system for real-time detection and localization of unhealthy tree from drone images and AI)
  • Safety and security applications (LoRa UTM, motion- and thermal triggers, and automatic UAV deployment) and agricultural application (e.g. pest monitoring and control, wildlife detection, counting and rescue).