Stage Artifical intelligence / machine learning for image processing

Job description

Solvay is a science company whose technologies bring benefits to many aspects of daily life. With more than 24,100 employees in 64 countries, Solvay bonds people, ideas and elements to reinvent progress. The Group seeks to create sustainable shared value for all, notably through its Solvay One Planet plan crafted around three pillars: protecting the climate, preserving resources and fostering better life. The Group’s innovative solutions contribute to safer, cleaner, and more sustainable products found in homes, food and consumer goods, planes, cars, batteries, smart devices, health care applications, water and air purification systems. Founded in 1863, Solvay today ranks among the world’s top three companies for the vast majority of its activities and delivered net sales of €10.2 billion in 2019. Solvay is listed on Euronext Brussels (SOLB) and Paris and in the United States, where its shares (SOLVY) are traded through a Level I ADR program.The Research & Innovation function (R&I) leads Solvay into new growth territories with breakthrough innovation to generate new products, applications and processes in the field of sustainable chemistry. Partners with the GBUs to deliver more value and leverages the Group’s capabilities to improve competitiveness and create new business opportunities for Solvay and its customers.

Job Overview and Responsibilities:

The Solvay Laboratoire duFutur (LOF) at Pessac near Bordeaux is dedicated to speeding up the research atSolvay which is a big chemistry group.Onemethod to achieve this is high throughput (HT) screening using automatedworkflows either with commercial robots or developed in house coupled withautomated data analysis. In this context we have been using classical imageacquisition and processing for fully automated quantitative data analysis formany different applications for more than 15 years now.Oneimportant step in image analysis is segmentation where each pixel of an imageis classified as either background or belonging to a certain class. Propersegmentation is crucial for further image processing and with classical methodsif often only achieved using an optimized set of preprocessing parameters thatwork well for a certain scenario, but may be less adapted or fail altogetherfor (often only) slightly different scenarios.

Ouraim is now to make the step from classical to AI supported image processingusing our large number of properly segmented images as a training sets formachine learning.For each new application we can also quickly create newtraining sets.

We are hence looking for acandidate with experience in machine learning in general who wants to apply AImethods to improve image processing in a range of very different applicationfields.

Education and Experience:

The candidate will becomepart of a small but dedicated team of engineers and developers whose mission isto provide customized automation and analysis solutions for in house projectsas well as projects for other business units of the Solvay group. Apart from asolid technical and scientific background, we are looking for curious andcreative candidates with a strong motivation for interdisciplinary research ina multicultural environment.

Skills:

This internship is anopportunity to discover an industrial project in an applied researchenvironment.

Internshipduration and level: 5 month, master level

Start and end of internship: can benegotiated

Required skills: programmingexperience with AI in Matlab and/or python, experience in image processing is aplus but not obligatory