R. Sivagami ¹, V. Vaithiyanathan ¹, V. Sangeetha ¹, M. Ifjaz Ahmed ¹, K. Joseph Abraham Sundar ¹, K. Divya Lakshmi ¹

Affiliations
1 School of Computing, SASTRA University, Thanjavur - 613402, Tamil Nadu, IN

Abstract

Low resolution Multispectral images obtained from earth observation satellite interpreted directly have less information which is not suitable for remote sensing applications. The advancement of sensors onboard satellites provides panchromatic images which have more image details and low-resolution MS images lead the way for the researchers to develop algorithms suitable for fusing multi-sensor images which can improve the resolution of the MS images. In this paper review on various pixel based fusion algorithms and evaluation metrics are presented. From the literature review, it is inferred that the Multi resolution techniques will give better accuracy than all other traditional algorithms.

Keywords

Evaluation Metrics, Fusion Algorithms, Fusion Transforms, Image Fusion, Remote Sensing

Full Text

   

V. Sangeetha ¹, V. Vaithiyanathan ¹, R. Sivagami ¹, K. Divyalakshmi ¹, K. Joseph Abraham Sundar ¹, M. Ifjaz Ahmed ¹

Affiliations
1 School of Computing, SASTRA University, Thanjavur - 613402, Tamil Nadu, IN

Abstract

Skeletonisation is an effective way to provide an intact representation of an object’s shape. This approach has great applications in the fields of motion planning, object representation, tracking, information retrieval, topology representation, sensor distribution and location, facility location, forestry application, building constructions etc. Skeletons naturally provide a pixel to boundary mapping through which the path between points can be found in a terrain map. Many approaches for determining the skeleton of a terrain exist. In this paper, the different methods for Skeletonisation are discussed. Initially the various categorizations of Skeletonisation is discussed which is further followed by application of those algorithms for path determining. Quantitative measures for validating the skeleton are also discussed. Finally, the applications of Skeletonisation in various domains are elaborated.

Keywords

Distance Map, Distance Transform, Skeletonisation, Thinning, Voronoi Diagram

Full Text

   

V. Vaithiyanathan ¹, K. Divya Lakshmi ¹, K. Joseph Abraham Sundar ¹, M. Ifjaz Ahmed ¹, V. Sangeetha ¹, R. Sivagami ¹

Affiliations
1 School of Computing, SASTRA University, Thanjavur - 613402, Tamil Nadu, IN

Abstract

Local features of an image are used in many computer vision applications such as object detection and scene matching. The gradient orientation histogram is used by many local features such as Scale Invariant Feature Transform (SIFT), a widely used image local feature. This paper discusses various distance functions that can be used to measure the similarity between the local features described by the gradient orientation histogram. A distance function, based on the quadratic form is proposed for the SIFT descriptor. The state of the art distance functions - Euclidean, Chi-square, Manhattan and the proposed quadratic form based distance function are calculated between the features extracted from the images. Nearest neighborhood ratio strategy is used to find the corresponding features based on the distance measure. Correct matches are estimated using the ground truth transformation function between the images, present in the form of homograph matrix. It is experimentally found that the proposed distance function has an execution time reduced by 21% compared to the Euclidean distance for a similar accuracy performance. The proposed distance retrieves more number of correct matches compared to the modified Earth Mover distance which is fastest among the evaluated distance functions. The future work will be aimed at improving the time taken for computing the distance matrix between the feature sets and a better strategy for computing the matches.

Keywords

Distance Functions, Gradient Orientation Histogram, Image Matching, Local Features, SIFT

Full Text

   

K. Joseph Abraham Sundar ¹, K. Divyalakhsmi ¹, M. Ifjaz Ahmed ¹, R. Sivagami ¹, V. Sangeetha ¹, V. Vaithiyanathan ¹

Affiliations
1 School of Computing, SASTRA University, Thanjavur - 613 401, Tamil Nadu, IN

Abstract

Super resolution image reconstruction is defined as generating a high resolution image from a low resolution image or sequence of low resolution images captured from identical scene apparently a video. An algorithm for reconstructing a high resolution image from a low resolution image by altering the frequency components is discussed in this paper. In this method the high frequency components of the zoomed low resolution image is modified so as to increase the resolution of the low resolution image. The evaluation for the experiments is based on the performance measure matrix peak signal to noise ratio. The experimental results shown proves that the algorithm is highly advantageous and computationally fast compared to the other interpolation methods. The algorithm will be helpful in practical applications of medical imaging diagnosis, remote sensing and military applications.

Keywords

Filters, Frequency Spectrum, Interpolation, Image Reconstruction, Super Resolution

Full Text

   

M. Ifjaz Ahmed ¹, R. Sivagami ¹, K. Joseph Abraham Sundar ¹, K. Divyalakshmi ¹, V. Sangeetha ¹, V. Vaithiyanathan ¹

Affiliations
1 School of Computing, SASTRA University, Thanjavur – 613401, Tamil Nadu, IN

Abstract

Image fusion, the process of combining multi-modal or multi-focus images to form a new informative image than input images. The fused image contains more information than their input images which helps remote sensing applications for image analysis efficiently. Many image fusion algorithms have been developed, but all of them introduced a term called as color distortion. The huge difference between the intensity of input images leads to the color distortion. The Intensity-Hue-Saturation fusion technique gives more access to the user to work on pixel by pixel. This method accessibility delivers the good fusion technique base to reduce the color distortion. We proposed an algorithm which solves the color distortion problem effectively and it can produce an output image as highly informative. The difference in intensity of input images reduced by injecting the intensity of the high resolution image. The proposed work performs better than existing algorithms and it is proved by statistical measures like a correlation coefficient, mean, standard deviation, ERGAS, UIQI and spectral angle mapper.

Keywords

Color Distortion, Enhancement, IIHS Transform, Image Fusion, Spatial Injection

Full Text