Image Vectorization and Significant Point Detection

PREETI, JOSHI (2019) Image Vectorization and Significant Point Detection. In: International Conference on Advances In Engineering And Technology - ICAET 2014, 24 - 25 May, 2014, RIT, Roorkee, India.

20140721_103901.pdf - Published Version

Download (655kB) | Preview
Official URL:


In this paper we present a novel method for compact representation of images as a list of lines and splines. The method uses vectorization of images based on significance measure. The image could be represented as a list of dominant points (spline control points) instead of the m × n array of pixels. This is a compact representation of images and experiments have shown that the compression achieved in vector images is 99% and for curved images also we could represent the image as a list of spline curves. These dominant points are called the significant points using which the original image could be reconstructed without any significant loss of information. The method is affine transformation invariant. This method has many advantages such as compression, faster processing, transmission and less storage. The compact representation of image could be used for pattern matching such as character recognition. The paper also discusses piece-wise linear reconstruction of image from these significant points. Low reconstruction error is a measure of goodness of proposed technique.

Item Type: Conference or Workshop Item (Paper)
Uncontrolled Keywords: image vectorization, reconstruction, segmentation.
Depositing User: Mr. John Steve
Date Deposited: 21 May 2019 10:54
Last Modified: 21 May 2019 10:54

Actions (login required)

View Item View Item