Visual Inspection

Contact: Professor Heikki Kälviäinen

Visual inspection research at MVPR is focused on imaging, image processing, and image analysis methods for machine vision applications. The research topics include visual quality assessment, image-based process control, human quality perception and machine vision measurements, and visual object categorization. The applications are focused on industrial machine vision, especially forest and printing industry, image-content based solutions, and medical image processing. One of the main objectives in forest industry applications is resource-efficient and environmentally sound production with the known quality, using less raw material, water, and energy.

Scene Understanding

Principal Investigator: Prof. Joni Kamarainen

Our research aims at total scene understanding. Relevant sub-topics are object class detection (visual object categorisation, VOC), specific object detection and event analysis from video. These are some of the most actively investigated topics in computer vision. We develop methods which can be used for efficient visual search and indexing in large scale image and video databases and Web.

Our main interest for the past few years have been part-based methods for object class representation, learning and detection. Our methods are general, but particularly good results have been achieved for face and car license plate detection and localisation. Our main scientific contribution is a novel local feature referred to as "simple Gabor feature space" or "multi-resolution Gabor feature", which provides robust and invariant feature for learning and detecting local object parts. Our main focus is now on spatial (constellation) models of local parts and on semi-supervised and completely unsupervised methods for visual object categorisation.

Biomolecular Vision

Contact: Professor Lasse Lensu

Molecular computing is a relatively new field of science where novel computing approaches are searched from the domain of molecules and atoms. More specifically, the aim is to understand how to control molecular reactions for information processing. Despite the fact that the Moore's "law" still holds, these studies are motivated by the increasing technical difficulties to further develop the CMOS transistors as the building blocks of computing devices. These difficulties are already realising because the microelectronics industry has already pushed multicore and other parallel architectures to the market.

Biomolecules offer several advantages over synthetic ones. In their natural environment, their functionality and robustness is usually close to optimal due to evolutionary steps during the development of their structure and function. Therefore, many things can be learned from nature by studying the biomolecules and their interactions. Most of the studies concerning information processing using biomolecules have concentrated in DNA and photoactive biomolecules, for example, rhodopsins, chloroplasts, photosynthetic reaction centers and light-harvesting complexes, and retinal proteins. Bacteriorhodopsin (BR) is a retinal protein which has been intensively studied and proposed for various applications.

Medical Imaging

Contact: Professor Lasse Lensu

Medical imaging is the most important application area of computer vision and image analysis. In the recent years, MVPR has conducted research on processing microscopy images and automatic detection of lesions of diabetic retinopathy from fundus images. The laboratory has published the DiaRetDB1 database for benchmarking diabetic retinopathy detection methods and the database has become one of the standard benchmarks in the field.

Laboratory Machine Vision and Pattern Recognition (MVPR)
Head of Laboratory Heikki Kälviäinen

tel:+358 40 586 7552
fax:+358 5 621 2899

Professor Lasse Lensu

tel:+358 40 759 1720
fax:+358 5 411 7201

Associate prof. Arto Kaarna

tel:+358 40 576 2525

email to personnel: