Manish Kashyap

Learn More
In this paper we present an efficient approach for clustering analysis to detect embedded and nested adjacent clusters using concept of density based notion of clusters and neighborhood difference. Basically our proposed algorithm is improved version basic DBSCAN algorithm, proposed to address the clustering problem with the use global density parameters in(More)
`Maximally Stable Extremal Regions' (MSER) based interest points are frequently used for medical image registration on account of their robustness to noise, better localization, and good repeatability. However, if the objects in the image do not have sharp boundaries (as is the case with medical images), the number of MSER's detected is low. Also, MSER's(More)
Through this paper a computationally efficient approach to object extraction in fuzzy settings is presented. The proposed approach is a `potential alternative' (in terms of computational complexity) to the two most popular and reliable image segmentation algorithms namely `Udupa and Samarasekera's Fuzzy connectedness algorithm' and the `level set(More)
Histogram based multilevel thresholding is one of the most aggressive methods to realize image segmentation. We have used histogram based multilevel thresholding as a gray image applying Firefly and social spider algorithm. For thresholding, we have worked on Kapur's and Otsu's methods to maximize the objective function values. We have used standard images,(More)
We present seeded region growing algorithm using notion of `affinity' as region growth mechanism for segmentation of medical images. Affinity between pair of pixels captures the idea of nearness in location and similarity between their gray scale values. Affinity has the capability to separate different segments of the images depending upon its functional(More)
Organizing data into sensible groups is called as ‘data clustering.’ It is an open research problem in various scientific fields. Neither a universal solution nor an absolute strategy for its evaluation exists in the literature. In this context, through this paper, we make following three contributions: (1) A new method for finding ‘natural groupings’ or(More)
Various studies on interest point (IP) detection have concluded that maximally stable extremal region (MSER)-based IPs outperform others on repeatability, localization accuracy, robustness, efficiency and covariance to global and local image distortions. Since medical images lack sharp detail, corner IPs are not a suitable choice for them. Instead, MSERs(More)
The correlation between the environmental features and image features of cotton bolls is the necessary step for the pattern recognition and translate those features for machine understanding is the main challenge to distinguish mature cotton boll from immature one. Present work is tried to solve this problem using shape based features. The fuzzy based(More)
There are many methods to find Interest Points (IPs) in images for image registration. However, the underlying heuristics for finding them is different for each. Due to this, their behavior towards different image distortions is expected to vary. Through this paper, we attempt to investigate the truth of the following hypothesis - "Global Transform is(More)
  • 1