Muhammad Faisal Khan

Segmentation and Classification Using Logistic Regression in Remote Sensing Imagery

This paper presents techniques for segmentation and change classification using logistic regression. The research was conducted on SPOT 5 multispectral multitemporal images covering the 2010 floods in Pakistan. Segmentation was performed to extract the built up area (BUA) from the satellite images and change detection was performed to find the damaged BUA. The damaged area was classified into three categories based on the extent of damage. The segmentation results were validated using statistical measures like precision, recall, and dice coefficient on available ground truth. The results of change classification were compared and found consistent with the manual assessment report produced by UNO experts using Worldview 1 satellite imagery with submeter resolution. The proposed scheme and results give an indication that SPOT 5 imagery can be used for fast automatic damage assessment and classification immediately after a natural calamity. The proposed change detection technique was also applied on Unites States Geographical Survey dataset. We compared our change detection results with established methods like change vector analysis, Principal component analysis using K-means and commercially available software Erdas Imagine on both the above-mentioned datasets. The comparison results suggest that our proposed algorithm performs better than the other methods.

Deformable image registration based on elastodynamics

This paper presents a novel method for 2D inter-subject non-rigid image registration. It is motivated by the ideas derived from elastodynamics which is the subclass of linear elastic theory. We propose to model the non-rigid deformations as elastic waves which are characterized by elastodynamics wave equation. Our registration method is formulated in a multi-resolution manner such that it is able to recover significant deformations. The proposed scheme recovers local (non-linear) as well as global deformations. The results of image registration by the proposed technique were validated via similarity measures for real MR brain images, BrainWeb dataset and MR prostate images. We have also used overlap measures (Jaccard distance, Dice coefficient and Overlap coefficient) over the segmented deep-brain structures including: lateral and third ventricles, putamen and caudate nucleus for real image dataset to further validate our proposed registration scheme. The proposed scheme was also compared with demons, Level set and free-form deformations (cubic B-spline) algorithms implemented in ITK toolkit and symmetric normalization method available in ANTs software package. Our proposed algorithm shows marked improvement of the inter-subject non-rigid registration, both qualitatively and quantitatively, over the four above-mentioned methods. This comparison was performed on all the three types of datasets mentioned above. 2D synthetic experiments comprising of Patch-‘C’, Square–Rectangle and Circle-‘C’ were also performed with our algorithm and compared with other four methods. Both qualitative and quantitative results show that our method performs better than the other four methods. Results suggest that our proposed algorithm has improved accuracy over lemons, level set and free-form deformations (ITK implemented) and symmetric normalization method (ANTs based), tested over two types of brain and one prostate MRI. The commonly used synthetic experiments also suggest the superiority of our proposed scheme.

Topology preserving non-rigid image registration using time-varying elasticity model for MRI brain volumes

In this paper, we present a new non-rigid image registration method that imposes a topology preservation constraint on the deformation. We propose to incorporate the time varying elasticity model into the deformable image matching procedure and constrain the Jacobian determinant of the transformation over the entire image domain. The motion of elastic bodies is governed by a hyperbolic partial differential equation, generally termed as elastodynamics wave equation, which we propose to use as a deformation model. We carried out clinical image registration experiments on 3D magnetic resonance brain scans from IBSR database. The results of the proposed registration approach in terms of Kappa index and relative overlap computed over the subcortical structures were compared against the existing topology preserving non-rigid image registration methods and non topology preserving variant of our proposed registration scheme. The Jacobian determinant maps obtained with our proposed registration method were qualitatively and quantitatively analyzed. The results demonstrated that the proposed scheme provides good registration accuracy with smooth transformations, thereby guaranteeing the preservation of topology.

Non rigid image registration by modeling deformations as elastic waves

In this paper we propose an inter-subject non-rigid image registration method that is derived from the concept of elastodynamics. Non-rigid warps are modeled as elastic waves which tend to recover extensive deformations. The deformable transformation model is given by Navier PDE which is solved iteratively using finite difference approximations. The dynamical displacements so computed represent the warp field which is then used to bring the subject image into spatial correspondence with the atlas. The results were validated using Jaccard coefficient (J), Dice coefficient (D), Overlap coefficient (O) and Normalized Cross Correlation (NCC) with mean ± std values of 0.81 ± .04, 0.89 ± .03, 0.99 ± .01 and 0.97 ± .01 respectively. This quantitative assessment confirmed that the proposed registration method performs very well.

River extraction from high resolution satellite images

High resolution satellite imagery has large ground coverage and extraction of main geographical features such as rivers is of vital importance for numerous applications. In this paper, we propose an automatic technique for extraction of rivers from panchromatic satellite images using an iterative multi resolution decomposition of textural features, combined with morphological operations. The method was used for extraction of Kabul river of Pakistan from high resolution SPOT 5 satellite image.

Network weight adjustment in a fractional fourier transform based multi-channel brain computer interface for person authentication

Brain is composed of unique complex neural structure thus electrical activity between neurons referred to as electroencephalogram (EEG) in different brain regions varies from one user to another. In this paper EEG distinctiveness is exploited through application to person authentication system based on five mental imagery tasks. Seven electrodes placed at C3, C4, P3, P4, O1, O2 and EOG are used to record EEG signals. A parallel structure of Exact Radial Basis (RBE) neural networks are used as classifiers. Individual classifier response for each mental task is evaluated and a weighting approach is used to regulate contribution of each channel within a multi-channel Brain Computer Interface (BCI) system. The estimated and experimental results indicate an average increase of 14% in system performance when tested on 722 trials of 1sec duration for 7 subjects. Fractional Fourier Transform (FRFT) with order optimization is used for feature extraction, and special one dimensional case of k-means clustering algorithm is used to calculate the threshold for individual classifiers.

3D non-rigid image registration with inverse consistency constraint on elastodynamics wave equation

In this work, a new inverse consistent non-rigid image registration method is presented. The inter-subject deformations are modeled as elastic waves which tend to propagate over the image domain while recovering the nonlinear discrepancies between the two images. An inverse consistency constraint is introduced into the inertial force that is part of the elastodynamics wave equation which governs the underlying non-rigid deformations. The proposed registration method was analyzed over 3D MR brain scans both quantitatively and qualitatively. Normalized cross-correlation (NCC) was utilized to check the registration accuracy, and it revealed that the performance of proposed registration method with and without inverse consistency constraint is comparable in terms of NCC which increased by

Dynamic elasticity model for inter-subject non-rigid registration of 3D MRI brain scans

13\%