This preview shows page 1. Sign up to view the full content.
Unformatted text preview: could employ:
The length of the sensed edge must be less than or equal to the length of the
model edge under consideration.
The angle between two adjacent sensed edges must agree with that between the
two corresponding matched model edges.
Faculty of Engineering Robotics Technology MECH 4041 B. Eng (Hons.) Mechatronics S. Venkannah Mechanical and Production Engineering Department Let
represent the range of vectors from any point on sensed edge a to
any point on sensed edge b. In an interpretation which respectively pairs sensed
edges a and b with model edges i and j, this range of vectors must be compatible
with the range of vectors produced by i and j.
For Surface we could employ:
Angles between planes.
Area of planes.
Measures of curvature. Model-based recognition
All object recognition systems contain the following modules to some extent:
Selection. What subset of the data corresponds to an object?
This problem, also known as the perceptual grouping problem, aims to organise the features that
come from a single object into a single set. The types of features considered might be edges,
corners, lines, curves represented as splines, or regional features such as texture. The grouping is
usually accomplished using cues such as proximity, parallelism, collinearity, and continuity in
Indexing. Which object model corresponds to the data subset?
Correspondence. Which individual model features correspond to each data
Thus object recognition is a process of hypothesizing an object-to-model correspondence and
then verifying that the hypothesis is correct. Generally an hypothesis is considered successful if
the error between the projected model features and the corresponding image features is below
some threshold, and a reasonable fraction of the object outline is covered by the image features.
For the two approaches mentioned above, that of estimating the transformation undergone in the
imaging process has complexity
is the number of models, i is the number of image features, m is the
number of features per model, and k is the number of features needed to determine the objectimage transformation. Typically, k is about 4.
The approach that uses transformation-invariant measurements of the object in the image for
recognition has complexity O(ik),where k is the number of features required to form the
indexing. In this case, recognition need not be proportional to the number of models in the
library. This can be a considerable advantage when the number of models is large.
Model Based Recognition
Faculty of Engineering Robotics Technology MECH 4041 B. Eng (Hons.) Mechatronics S. Venkannah Mechanical and Production Engineering Department Recognition is a matching problem between scene and model description.
Matching is a Classic AI problem.
Many methods applied to try to solve this problem.
Computationally complex and intensive.
We will look at three broad methods here. Relaxation Labeling Methods
Scene labeling and constraint propagation
The goal of scene labeling is to assign a label (a meaning) to each image object to achieve an
appropriate image interpretation. Assume that regions have been detected in an image that
correspond to objects or other image entities, and let the objects and their inter-relationships be
described by a region adjacency graph and/or a semantic net. Object properties are described by
unary relations, and inter-relationships between objects are described by binary (or n-ary)
The resulting interpretation should correspond with available scene knowledge. The labeling
should be consistent, and should favor more probable interpretations if there is more than one
option. Consistency means that no two objects of the image appear in an illegal configuration e.g. an object labeled house in the middle of an object labeled lake will be considered
inconsistent in most scenes. Conversely, an object labeled house surrounded by an object
labeled lawn in the middle of a lake may be fully acceptable.
Two main approaches may be chosen to achieve this goal.
Discrete labeling allows only one label to be assigned to each object in the final
labeling. Effort is directed to achieving a consistent labeling all over the image.
Probabilistic labeling allows multiple labels to co-exist in objects. Labels are
probabilistically weighted, with a label confidence being assigned to each object label.
Relaxation labeling has been applied to many problems in computer vision, from edge detection
to scene interpretation on the basis of labeled scene components. Early work on scene labeling of
line drawings employed a discrete relaxation approach in which each scene component was
assigned a set of possible interpretations, and inconsistent labeling were removed by examining
firstly label pairs on connected segments, and by ensuring secondly that the locally defined
consistencies could be linked together in a continuous closed path. This type of algorithm co...
View Full Document
- Spring '14