According one is Image Classification. This implicates
to the research project, the system should detect objects which help to water
remain and grow mosquito larvae. The system uses a video frames to do that and
uses video processing to identify those places. Therefore there are more issues
should have understand. The necessity for retrieving relevant information from
video, and the technological advancements have led fundamental research looking
for novel processing techniques in image enhancement for reconstruction, restoration,
de-noising, improvement of resolution and reduction of artifacts and noise;
image segmentation for extracting an object from a background; recognition for
identification and classification; pattern recognition for monitoring and
diagnosis; security and compression for improving the protection of
information, optimizing its storage, and maintaining its quality; image
classification for object recognition, description, and matching of feature
points; image annotation for interpretation of contents through words, keywords
or comments; and image modeling to detect depth information of a target object1.
recognition and image segmentation is the most critical thing in the video
processing. In here the system should have a capable to identify patterns and
regularities in data. As well as Image Segmentation allows locating objects and
boundaries in images by grouping pixels that share certain characteristics 1.
The second one is Image Enhancement. This plays a significant role in modern
applications such as auto-focusing, remote sensing of the environment, object
tracking, noise removal, object sharpening, quality assessment and improvement
1. Because of some fog, cloudy sky…etc there may be an information loss.
Therefore image or video to use there should be use image enhancement. Then the
next one is Image Classification. This implicates the detection, description,
and matching of feature points, which forms the basis of many applications such
as image registration, object recognition, and image retrieval, among many
to make accurate identifications of objects, the system should be included all
2.2.1 Object detection system
for detecting instances of an object in a digital image
This is an object detection system for detecting instances of
an object in a digital image includes an image integrator and an object
detector, which includes a classifier (classification function) and image
scanner 2.An input image is received by image integrator and then calculate
internal image representation of the input image. Then the image scanner scans
the image in same sized subwindows. The each subwindow should contain an
instance of the object. To classify the subwindow, use homogeneous
classification functions. . Each classifier evaluates one or more features of
the object to determine the presence of such features in a subwindow that would
indicate the likelihood of an instance of the object in the subwindow 2.
are more methods to detect objects in image. The prior art approaches are the
most existing method for this. This use a learning algorithm based on a
training data set that contains many examples of the object. In this approach,
an object detector uses a scanning process to enumerate all possible patches (subwindows)
within a given image 2. The
classification function is run against all such patches to detect the possible
presence of an instance of the object in the patch. Although this prior art
approach have a high rate of accurate detection of objects in an image, but may
perform the object detection process in a relatively slow manner compared to this
method. As example to identify an object in an image, prior art approach take
more time because it has to use many number of patches.
detector engages in much less initial image processing. By using classifier and
scanners, this object detector creates and uses an integral image while prior
art approaches use an image pyramid. This object detector invention computes
the integral image in less than about 10 operations per pixel. Nevertheless,
the object detector detects objects at any scale and location. This model uses
a feature representation, which detects objects by looking for the appearance
of features, which have basic geometric shapes. These simple features, combined
with the integral image, allow for a computationally efficient approach to
identifying whether a given area in a subwindow has a feature of interest (that
may identify an instance of the object, usually along with other features).
This approach of the invention is more powerful and more efficient than looking
at the pixels of the image itself as is done in many prior art approaches 2.
And also this classification and scan method is faster than others because it
quickly determines object at a given scale and location. As well as this method
can be used in real-time applications.