TY - BOOK
T1 - Morphological analysis of wear debris
T2 - Current state of available solutions
AU - Hietanen, Seija
N1 - Project code: H1SU00229
PY - 2002
Y1 - 2002
N2 - For a computer to automatically identify wear and
contaminant particles it is essential to develop image
processing software robust enough to extract the shape
and surface features of the particle accurately and
independently of the quality and content of the image.
The texture, colour and shape of the particle reveal the
wear mode that generated the particle in a machine.
Usually what we see on the computer screen is the result
of an attack while the particle is in the lubricant, and
distortions generated by the microscope and vision system
used to produce the image. Wear particles and surfaces
are three-dimensional objects and their numerical
characterisation and classification is still largely an
unresolved problem. Usually a set of various parameters
is used to describe the surface topography. These
parameters are, however, of limited use, especially when
dealing with anisotropic surfaces.
In this report the main emphasis is on the current state
of optical, microscope and camera based image analysis
techniques and different approaches used with them to
analyse wear debris morphology. LaserNet Fines combines
the standard oil analysis techniques of particle counting
and shape classification into a single analytical
instrument, by combining imaging technology and neural
net shape classification. LaserNet is an optically based
debris monitoring technology that determines the
existence, type, severity and rate of progression of
mechanical faults by measuring the size distribution,
shape characteristics and rate of production of debris
particles. Other methods used are computer aided vision
engineering (CAVE); a program developed to analyse
particle morphology quantitatively, a software-based wear
debris classification system - SYCLOPS (systematic
classification of oil-wetted particles), Jetscan; a SEM
and EDX system in which the process of particle
identification, measurement, characterisation and
diagnosis is automatic, and EDAX SEM - EDS method.
AB - For a computer to automatically identify wear and
contaminant particles it is essential to develop image
processing software robust enough to extract the shape
and surface features of the particle accurately and
independently of the quality and content of the image.
The texture, colour and shape of the particle reveal the
wear mode that generated the particle in a machine.
Usually what we see on the computer screen is the result
of an attack while the particle is in the lubricant, and
distortions generated by the microscope and vision system
used to produce the image. Wear particles and surfaces
are three-dimensional objects and their numerical
characterisation and classification is still largely an
unresolved problem. Usually a set of various parameters
is used to describe the surface topography. These
parameters are, however, of limited use, especially when
dealing with anisotropic surfaces.
In this report the main emphasis is on the current state
of optical, microscope and camera based image analysis
techniques and different approaches used with them to
analyse wear debris morphology. LaserNet Fines combines
the standard oil analysis techniques of particle counting
and shape classification into a single analytical
instrument, by combining imaging technology and neural
net shape classification. LaserNet is an optically based
debris monitoring technology that determines the
existence, type, severity and rate of progression of
mechanical faults by measuring the size distribution,
shape characteristics and rate of production of debris
particles. Other methods used are computer aided vision
engineering (CAVE); a program developed to analyse
particle morphology quantitatively, a software-based wear
debris classification system - SYCLOPS (systematic
classification of oil-wetted particles), Jetscan; a SEM
and EDX system in which the process of particle
identification, measurement, characterisation and
diagnosis is automatic, and EDAX SEM - EDS method.
KW - wear debris analysis
KW - morphological analysis
M3 - Report
T3 - VTT Manufacturing Technology. Research Report
BT - Morphological analysis of wear debris
PB - VTT Technical Research Centre of Finland
CY - Espoo
ER -