Research Highlights Student projects

Image Processing

Current research projects being carried out involve the development of new image processing algorithms including real-time implementations and machine learning techniques. The researches target multiple areas such as medical field, cosmetic industry, meat and agriculture industries among other.


Skin Image Analysis

Current works involve the development of a multispectral imaging device to extract and measure the different elements composing the skin. The research focuses on two aspects: hardware development and signal/image processing algorithm. The hardware development aims to create a low-cost multispectral camera allowing the acquisition of skin area at multiple wavelengths not visible by the human eye. The development of signal/image processing intends to analyse data acquired by imaging devices. The ultimate goal is to objectively quantify different skin components for the assessment of skin lesions and treatment efficacy.

Texture imaging platform

Visual image’s texture is a key to important information that is often used in various image analysis applications e.g., biomedical image analysis, terrains classification via satellite images and wood species recognition. To describe the texture of the image, different features can be extracted from the image. Some common features that are extracted from the images include grayscale statistics (mean, standard deviation, max, min, Kurtosis, skewness, histogram), gray-level run length matrix (GLRL), gray-level Co-occurrence matrix (GLCM), frequency spectrum, etc. The aim of this project is to develop a platform that can analyze textures from images and extract only features that are relevant for specific applications.

DigiKup - Digital Make-up testing platform

In recent years, research has been shifting to the digital world for many aspect of daily life, one being make-up testing. BU-CROCCS has developed Digikup, our own digital makeup. The system is offering a unique experience for makeup customers by virtually testing makeup on them in a realistic fashion using a 3d lookalike rendering. Therefore, testing multiple makeups is fast and avoids potential unhygienic contact. The project involves the development of image processing algorithms for face and makeup area detection as well as realistic rendering. The project aim to provide a new tool for shopper wanting to ease their makeup testing.

Skin Image Analysis

Texture imaging platform

DigiKup - Digital Make-up testing platform