Les commentaires sont clos.

Post Doctorat - AGRIPOLHYS: AGRiculture with Intelligent POLarimetric and HYperspectral Sensing

27 Octobre 2023

Catégorie : Post-doctorant

The subject of this postdoctoral position is led by L@bISEN and is part of the ANR AGRIPOLHYS project in collaboration with Vegenov (a specialized technological resource center for plants) and Photon Lines (an optical solutions architect). This project aims to develop advanced tools for smart agriculture using hyperspectral and polarimetric imaging assisted by Artificial Intelligence.

Job Profile:
  • Affiliated Institution: Yncréa Ouest, a private higher education institution of general interest (EESPIG), under contract with the Ministry of Higher Education and Research
  • Research Unit: L@bISEN
  • Research Team: LSL (Light – Scatter – Learning)
  • Workplace: Brest campus
  • Contract Duration: 18 months
  • Salary: 35k€/year, subject to experience
  • Opportunity to participate in teaching activities at Yncréa Ouest
About the Research Team:
The research activities of the LSL (Light Scatter Learning) team are focused on innovative optical imaging and object modeling through artificial intelligence. Their research domain comprises four main aspects:
  • Development of advanced optical models.
  • Utilization of hyperspectral and polarimetric data to power artificial intelligence models.
  • Application of these advancements in real-world scenarios.
  • Exploration of unconventional optical imaging techniques.
To realize these endeavors, the team has developed a hyperspectral imaging bench as well as a multispectral imaging polarimeter. Additionally, several models are currently under exploration. The research applications encompass fields such as remote sensing, food product inspection, medical diagnostics, and more.
Within the scope of this project, the postdoctoral researcher will collaborate with the team, which consists of experts in optics, mathematics, and artificial intelligence.
Project Context:
Smart agriculture is crucial for the success of ecological transition, but it faces unresolved technological and scientific challenges. Hyperspectral imaging (HSI), with its ability to monitor plant health in a non-destructive and rapid manner, holds promise for smart agriculture. HSI records light in numerous narrow spectral channels, covering a wide range of wavelengths, making it more suitable than traditional cameras for detecting plant diseases when symptoms are visible to the human eye [1,2].
However, the goal is to detect diseases earlier before they become visible. In this context, in addition to HSI, there is also polarimetric imaging (PIM), which has recently shown significant potential for plant disease detection [3]. Polarimetric imaging (PIM) records the polarization properties of light transmitted/reflected by objects or surfaces. These data include details on how light interacts with object structures, which can be valuable for characterizing plants and detecting diseases at an early stage, even before symptoms visible to the human eye appear.
Project Objective:
The objective of this project is to provide a decision support tool for smart agriculture using hyperspectral and polarimetric imaging, assisted by artificial intelligence. This tool will enable early disease detection, reducing plant losses and the need for chemical treatments.
Experiments will be conducted on tomato leaves with the aim of early detection and identification of four tomato diseases: powdery mildew, downy mildew, gray mold, and leaf mold. These research efforts will begin with the analysis of each disease individually and progress toward a consortium approach to identify them collectively. Keywords: Smart agriculture, hyperspectral imaging (HSI), polarimetric imaging (PIM), machine learning, sustainable agriculture, plant monitoring.
The candidate must possess:
  • A Ph.D. in experimental physics/computational physics/optics and photonics.
  • Strong skills in mathematics, science, and data analysis...
  • Programming experience (preferably in Python).
  • Understanding of machine learning for image segmentation.
  • Self-taught, he/she can work independently and creatively solving problems.
  • Excellent oral and written communication skills, with the ability to present research results clearly and concisely.
  • Previous experience in Artificial Intelligence and/or hyperspectral field would be a significant asset.
  • Motivated and passionate about the field of optics.


  • Opportunity to participate in a cutting-edge collaborative research project that integrates advanced technologies, enabling you to acquire advanced skills.
  • Interdisciplinary collaboration with innovative companies at the forefront of technological advancements.
  • Contribution to Smart Agriculture and Ecological Transition: By participating in this project, you will make a significant contribution to smart agriculture and ecological transition, reducing crop losses and minimizing the use of chemicals for a positive environmental impact.
  • Possibility of publishing in reputable conferences and journals.
  • Competitive compensation and attractive benefits.
  • Opportunity for a permanent position as a teaching and research faculty member within our institution at the end of the contract.
To apply :
Please submit the following documents:
  • Curriculum vitae (CV).
  • Cover letter.
  • Any other documents deemed useful to enhance the application (letters of recommendation, scientific articles, thesis report, etc.).
via email to the following addresses:
Application deadline : January 8th, 2023.
Références :
[1] J. Abdulridha, “Laboratory and UAV-Based Identification and Classification of Tomato Yellow Leaf Curl, Bacterial Spot, and Target Spot Diseases in Tomato Utilizing Hyperspectral Imaging and Machine Learning”, Remote Sens., 12, 2732, 2020.
[2] N. Zhang, “A Review of Advanced Technologies and Development for Hyperspectral-Based Plant Disease Detection in the Past Three Decades”, Remote Sens., 12, 3188, 2020.
[3] C. Rodríguez et al., “Polarimetric observables for the enhanced visualization of plant diseases”, Nature Scientific Report, 12, pp. 14743, 2022.
[4] A. Pierangelo et al., "Multispectral Mueller polarimetric imaging detecting residual cancer and cancer regression after neoadjuvant treatment for colorectal carcinomas," Journal of Biomedical Optics, 18(4), pp. 046014, 2013.
[5] D. Wang et al., “Early Detection of Tomato Spotted Wilt Virus by Hyperspectral Imaging and Outlier Removal Auxiliary Classifier Generative Adversarial Nets (OR-AC-GAN)”. Sci Rep 9, 4377, 2019.
[6] C. Baskiotis et al., "Selecting Hyperspectral Bands for Leaf Mass Per Area Prediction by Means of Neural Networks," IGARSS 2022 - 2022 IEEE International Geoscience and Remote Sensing Symposium, Kuala Lumpur, Malaysia, pp. 1408-1411, 2022.
[7] C. Baskiotis et al., “Information extraction in polarization-resolved second harmonic generation microscopy of human tissues for automatic cancer diagnosis”, in proc. SPIE Photonic West, BiOS, , San Francisco, paper 12382-24, oral presentation, 2023.
[8] L.W. Kuswidiyanto et al., “Plant Disease Diagnosis Using Deep Learning Based on Aerial Hyperspectral Images: A Review”. Remote Sens., 14, 6031, 2022.
[9] R.Q. Zhou et al., “Early Detection of Magnaporthe oryzae-Infected Barley Leaves and Lesion Visualization Based on Hyperspectral Imaging”, Front Plant Sci., 15;9:1962, 2019.
[10] S. Iwasaki et al., “Simultaneous Detection of Plant- and Fungus-Derived Genes Constitutively Expressed in Single Pseudoidium neolycopersici-Inoculated Type I Trichome Cells of Tomato Leaves via Multiplex RT-PCR and Nested PCR”, Agriculture 12, 254, 2022.
[11] R.Q. Zhou et al., “Early Detection of Magnaporthe oryzae-Infected Barley Leaves and Lesion Visualization Based on Hyperspectral Imaging”, Frontiers in Plant Science, 15;9:1962, 2019.
[12] C. Römer et al., “Early drought stress detection in cereals: simplex volume maximisation for hyperspectral image analysis”, Functional Plant Biology., 39(11):878-890, 2012.
[13] K. Kersting et al., “Pre-Symptomatic Prediction of Plant Drought Stress Using Dirichlet-Aggregation Regression on Hyperspectral Images”, Proceedings of the AAAI Conf