Machine Vision

Assignment - 2
- 15691A04C8 - DS Prasanth Kumar

1. Find the magnitude of Edge and direction using

(a) Sobel operator (b) Prewitt (c) Robert

Edges:

 Edges are significant local changes of intensity in an image.

 Edges typically occur on the boundary between two different regions in an image.
Edge detection using derivatives
- Calculus describes changes of continuous functions using derivatives.
- An image is a 2D function, so operators describing edges are expressed using partial
derivatives.
- Points which lie on an edge can be detected by:
 detecting local maxima or minima of the first derivative
 detecting the zero-crossing of the second derivative
2. What are factor that affect object recognition system?
The central issues that should be considered in designing an object recognition system
are:
Object or model representation: How should objects be represented in the model database?
What are the important attributes or features of objects that must be captured in these
models? For some objects, geometric descriptions may be available and may also be efficient,
while for another class one may have to rely on generic or functional features.
The representation of an object should capture all relevant information without any
redundancies and should organize this information in a form that allows easy access by
different components of the object recognition system.
Feature extraction: Which features should be detected, and how can they be detected
reliably? Most features can be computed in two dimensional images but they are related to
three-dimensional characteristics of objects.
Feature-model matching: How can features in images be matched to models in the
database? In most object recognition tasks, there are many features and numerous objects. An
exhaustive matching approach will solve the recognition problem but may be too slow to be
useful. Effectiveness of features and efficiency of a matching technique must be considered
in developing a matching approach.
Hypotheses formation: How can a set of likely objects based on the feature matching be
selected, and how can probabilities be assigned to each possible object? The hypothesis
formation step is basically a heuristic to reduce the size of the search space. This step uses
knowledge of the application domain to assign some kind of probability or confidence
measure to different objects in the domain. This measure effects the likelihood of the
presence of objects based on the detected features.
Object verification:

 How can object models be used to select the most likely object from the set of
probable objects in a given image?
 The presence of each likely object can be verified by using their models.
 One must examine each plausible hypothesis to verify the presence of the
object or ignore it.
 If the models are geometric, it is easy to precisely verify objects using camera
location and other scene parameters.
 In other cases, it may not be possible to verify a hypothesis.

3. What is neural network? How we train, validate, and test the network?
 Neural Networks:
Neural networks are a new method of programming computers. They are
exceptionally good at performing pattern recognition and other tasks that are very
difficult to program using conventional techniques. Programs that employ neural
nets are also capable of learning on their own and adapting to changing
conditions.
 1. In the training mode, the neuron can be trained to fire (or not), for particular input
patterns.

2. In the using mode, when a taught input pattern is detected at the input, its associated
output becomes the current output. If the input pattern does not belong in the taught
list of input patterns, the firing rule is used to determine whether to fire or not.

3. The firing rule is an important concept in neural networks and accounts for their high
flexibility. A firing rule determines how one calculates whether a neuron should fire
for any input pattern. It relates to all the input patterns, not only the ones on which the
node was trained on previously.

4. An important application of neural networks is pattern recognition. Pattern

recognition can be implemented by using a feed-forward neural network that has been
trained accordingly.

5. During training, the network is trained to associate outputs with input patterns.

6. When the network is used, it identifies the input pattern and tries to output the
associated output pattern.

7. The power of neural networks comes to life when a pattern that has no output
associated with it, is given as an input.

In this case, the network gives the output that corresponds to a taught input pattern that is
least different from the given pattern.

 Feed-forward networks

1. Feed-forward NNs allow signals to travel one way only; from input to output.
There is no feedback (loops) i.e. the output of any layer does not affect that same
layer.

2. Feed-forward NNs tend to be straight forward networks that associate inputs

with outputs. They are extensively used in pattern recognition.

3. This type of organization is also referred to as bottom-up or top-down.

 Feedback networks

1. Feedback networks can have signals traveling in both directions by introducing

loops in the network.

2. Feedback networks are dynamic; their 'state' is changing continuously until they
reach an equilibrium point.

3. They remain at the equilibrium point until the input changes and a new
equilibrium needs to be found.
 4. Feedback architectures are also referred to as interactive or recurrent, although
the latter term is often used to denote feedback connections in single-layer
organizations.

Back prop algorithm:

 The Back prop algorithm searches for weight values that minimize the total error of
the network over the set of training examples (training set).

 Back prop consists of the repeated application of the following two passes:

– Forward pass: in this step the network is activated on one example and the
error of (each neuron of) the output layer is computed.

– Backward pass: in this step the network error is used for updating the
weights. Starting at the output layer, the error is propagated backwards
through the network, layer by layer. This is done by recursively computing the
local gradient of each neuron.

4. Define Image range and range radar?

Machine vision relies on the pinhole camera model which models the geometry of
perspective projection but omits the effects of depth of field, based on the fact that only those
points within a certain depth range are in focus on the image plane.
Perspective projection assumes that the view volume is an infinite pyramid limited only by
the top, bottom, and sides of the viewable rectangle on the image plane.
Optical effects such as depth of field and light attenuation limit the view volume by
introducing limits on the distances to viewable objects. Objects may not be seen clearly if
they are too close or too far away.
The pinhole camera model assumes that the perspective rays pass through an infinitesimal
aperture at the front of the cam*era. In practice, the aperture must be larger to admit more
light.

Lenses are placed in the aperture to focus the bundle of rays from each point in the scene
onto the corresponding point in the image plane as indicated by the geometry of perspective
projection. The scene point, the projected point on the image plane, and the center of the
aperture are on the ray for the line of sight in perspective projection. A lens gathers more
light, allowing the camera to work with less ambient illumination or with a faster shutter
speed, but the depth of field is limited.

With the exception of limited depth of field and the camera calibration problems discussed in
Section 12.9, lenses do not introduce any effects that violate the assumptions of ideal
perspective projection, and most machine vision algorithms do not depend on the design of
the optical system for image formation.
5. Make 3- cluster of given data using K-mean clustering: Data-
{2, 3, 4, 6, 8, 10, 11, 13, 14, 18, 19}
6. What is relation between world coordinate and image coordinate?