Abstract—In image processing feature extraction starts from
an initial set of measured data and builds derived values as
features which are intended as informative and non-redundant.
This facilitates the learning and leading to interpretations. Image
processing Algorithms are used to detect and differentiate among
various desired portions or shapes features of an image. In
particular texture, color and shape features of image are used to
define similarity among images. Image is a representation of visual
information. Objects and its shape are important aspect as image
contains information about various useful structures during
information analysis. Segmentation of image is one of the
necessary task in image processing. It involves generation of
various useful parts as set of selected pixels. Such division aims
automation in image objects identification and relevant
information analysis. For any image shape object is unique. Object
detection algorithms typically use extracted features and learning
algorithms to recognize instances of an object category. It is
commonly used in applications such as image retrieval, security,
surveillance and advanced driver assistance systems. During the
process of object extraction there is a need to separate image
background and foreground in efficient and accurate manner.
This paper implements context shape boundary extraction along
with color segmentation is done to identify objects and its region
color. This combined methodology of image segmentation and
context shape description algorithm increases accuracy of
detecting useful feature of image. We describe a method for
registering pairs of images based on thin-plate spline mappings.
This method is scale, translation and rotation invariant.