Research

    The Center for Mathematical Morphology is the image processing laboratory of Mines ParisTech. Our research is focused on the development of mathematical morphology, a theory invented at CMM in 1964 that has many applications in fields including image processing or random sets.

    In recent years, we have dedicated a large part of our research to the joint application of machine learning and mathematical morphology techniques to classic problems in image processing, including image segmentation and denoising or object detection. In particular, we carried out numerous studies on the application of these techniques to new imaging modalities, such as hyper-spectral and multi-modal images, point clouds or non-Euclidean images. We are also developing research activities on the theory of random sets and its applications to the physics of heterogeneous media.

    Historically, we draw our inspiration from collaborations with the industrial world. The fields of application of our research are therefore very diverse: cosmetology, materials engineering, medical and ophthalmological imaging, non-destructive testing or analysis of urban scenes are thus in the spotlight in the studies conducted in our laboratory.

    Image processing

    • Mathematical morphology
    • Image Segmentation
    • Machine learning
    • Convolutional networks

    Imaging modalities

    • Hyper-spectral imaging
    • Multi-modal imaging
    • Points clouds
    • Non-Euclidean imaging

    Random sets

    • Heterogeneous media
    • Stochastic geometry
    • Homogenization
    • FFT Computations

    Applications

    • Materials engineering
    • Cosmetology
    • Ophthalmology
    • Non-destructive testing