Hanaa S. Ali

Boosted Decision Trees for Vertebral Column Disease Diagnosis

Vertebral column diseases are of the main public health problems which cause a negative impact on patients. Disk hernia and spondylolisthesis are examples of pathologies of the vertebral column which cause intensive pain. Data mining tools play an important role in medical decision making and deal with human short-term memory limitations quite effectively. This paper presents a decision support tool that can help in detection of pathology on the vertebral column using three types of decision trees classifiers. They are Single Decision Tree (SDT), Boosted Decision Tree (BDT), and Decision Tree Forest (DTF). Decision Tree and Regression (DTREG) software package is used for simulation and the database is available from UCI Machine Learning Repository. The performance of the proposed structure is evaluated in terms of accuracy, sensitivity, specificity, ROC curves, and other metrics. The results showed that the accuracies of SDT and BDT in the training phase are 90.65 and 96.77 %, respectively. BDT performed better than SDT for all performance metrics. Value of ROC for BDT in the training phase is 0.9952. In the validation phase, BDT achieved 84.84 % accuracy, which is superior to SDT (81.94 %) and DTF (84.19 %). Results showed also that grade of spondylolisthesis is the most relevant feature for classification using BDT classifier.

Efficient enhancement and matching for iris recognition using SURF

Iris recognition is gaining more attention and the development of the field is increasing rapidly. This paper presents a complete iris recognition system. The iris features are obtained using Speeded Up Robust Features (SURF) after enhancing the image using Contrast Limited Adaptive Histogram Equalization (CLAHE). A novel matching algorithm based on applying fusion rules at different levels is proposed. The algorithm has the advantage of reduced data storage and fast matching. It can also handle efficiently the problem of rotation, scaling, illumination variation and occlusions. The proposed algorithm is implemented and tested using CASIA (V4) database. The recognition accuracies obtained are 99% using left images and 99.5% using right images. Results show that fusion of right and left images scores increases the recognition accuracy. The recognition accuracies obtained after fusion are 99.5% and 100% using minimum and sum rules respectively. Moreover, the proposed algorithm has an excellent robustness with respect to increasing the number of subjects.

Statistical representation for iris anti-spoofing using wavelet-based feature extraction and selection algorithms

The development of fake iris detection systems, which is one of the most important topics in the biometric field, is growing rapidly. In this paper, discriminative statistical features are used for differing between real and fake iris images. The multilevel 2-D wavelet decomposition is employed to obtain approximation and detail wavelet channels. For feature classification, Euclidean distance and suitable fusion rules are applied. Problems with numerous features require the use of feature selection. Thus, to reduce the computational cost and enhance the system performance, an effective feature selection algorithm is proposed. CASIA-Iris-Syn database, which consists of about 10000 synthesized images, is used. Results show that the variance measure is efficient for detecting deceived attacks with 100% classification accuracy. The kurtosis measure gives 90.4648 %, which is the lowest accuracy obtained. Other feature selection algorithms are applied for a comparison purpose. Results prove the high explanatory capability of the prediction method. The proposed feature selection/classifier ensemble not only achieves dimensionality reduction, but also carefully investigates the dependence between the statistical features, and does not neglect features with complementary information. A poor choice of features may lead to significant deterioration in system performance. The proposed system has the advantage of working with large size database, and thus ensures the generalization ability of the proposed algorithms. Results also show that working with original non-segmented images not only reduces the processing time, but also enhances the classification accuracy, noticeably.

Entropy with Local Binary Patterns for Efficient Iris Liveness Detection

Iris anti-spoofing is one of the most important topics, in which the development is increasing rapidly. This paper introduces an efficient system for detecting iris attacks. The system avoids the segmentation and the normalization stages employed traditionally in fake detection systems. Wavelet packets (WPs) are used to decompose the original image into wavelet approximation and detail channels. Entropy values are extracted from the wavelet channels, and also from the local binary pattern (LBP) images of the channels. These features are used for discriminating between real and fake iris images. Support vector machines are used for the classification purpose. The aim is to contribute for improved classification accuracy with less computational complexity and reduced processing time. Entropy of the WP channels gives 99.9237% classification accuracy, and the entropy of the LBP images yields 99.781%, using ATVS-FIr-DB. Fusion of these features yields 100% classification accuracy. Entropy of the wavelet channels is sufficient to obtain 100% accuracy using CASIA-Iris-Syn database, without fusion. All images in both databases are used, without the need to discard images with unsuccessful segmentation. Segmented images from both databases are used for comparison. Results show that more discriminative features can be obtained using the proposed algorithm. System complexity and processing time are reduced noticeably, and the system is robust to different types of fakes.

An Efficient Source–Channel Coding for Wireless Image Transmission Over Underwater Acoustic Channel

In this paper, a complete system for image transmission in harsh underwater environment is proposed. The key to increase the performance of the system is the use of an efficient image compression algorithm with a bandwidth-efficient modulation technique. The wavelet packet (WP) decomposition is used to get the best image representation and the set partitioning in hierarchical trees is applied on the WP coefficients. The parental conflicts are resolved, the parent–child relationships are adapted and thus the similarities between cross-subbands are preserved. Reed–Solomon is used for forward error correction to combat with the errors in wireless transmission. Orthogonal frequency division multiplexing with differential quadrature phase shift keying is used to transmit the generated bit stream. Effective image quality metrics are used for objective evaluation. Results show that the proposed system manages to transmit images over the limited bandwidth, and to effectively minimize the perceptual degradation.

Piecewise linear model for haze level estimation and an efficient image restoration technique

Abstract In this paper, a piecewise linear predication model that relates light extinction coefficient (bext) to log of image contrast/transmission is proposed. The objective is to calculate more accurate bext values compared with previous treatments which used linear model of contrast relationship only. A multivariate linear regression model is learned for each linear portion of the model. Results show superior behavior of the proposed model in terms of error calculated between estimated values and ground truth values. Moreover, an effective approach to dehaze and enhance outdoor images captured under adverse weather conditions is introduced. A raw transmission map is estimated using a dark channel prior, then guided image filtering is used in two stages; first for transmission map refinement, second for details enhancement. Guided filtering not only exhibits the edge-preserving smoothing, but also a structure transferring property. Details enhancement results in improved visibility with minimum information loss.

Effective visibility restoration and enhancement of air polluted images with high information fidelity

In this paper, an effective approach for dehazing and enhancing outdoor images is proposed. The dark channel prior is used to estimate a raw transmission map. This transmission map is then refined using guided image filtering under the guidance of the hazy image. For colors and details enhancement, the adaptive manifolds high-dimensional filtering is applied to the recovered scene radiance. The proposed approach is compared with other enhancement techniques and effective quality assessment methods are used for objective evaluation. The visual information fidelity (VIF) metric is used to quantify the loss of image information, since contrast overcompensation may cause information loss. The structural similarity index (SSIM) is used to quantify the degradation of structural information between the enhanced and recovered image. Other metrics including the ratio between the gradient of visible edges before and after enhancement are also computed. Experimental results show that combining guided image filtering for refining the raw transmission map, with adaptive manifolds filtering for image enhancement, produces images with improved colors, high contrast and minimum information loss.

An IMS-based LTE-WiMAX-WLAN architecture with efficient mobility management

In this paper, a framework is proposed for interworking LTE, WiMAX and WLAN using IP Multimedia Subsystem (IMS) on top of the three technologies. The aim is to provide high quality real-time multimedia services during handoff. The proposed mobility management technique uses Mobile IP (MIP) and Session Initiation Protocol (SIP) together with IMS, which proves to maintain the continuity of an on-going session and data access during roaming between the different radio access technologies. A comparison between the proposed tight coupled architecture using MIP-SIP and the same tight coupled networks using SIP-based mobility management is presented. OPNET Modeler 17.1 is used to simulate the networks. Results show that the total average packet loss is reduced considerably during VoIP sessions using MIP-SIP in the proposed network.

An integrated system for underwater wireless image transmission

This paper focuses on presenting an efficient underwater acoustic communication system for image transmission. Water nature slows propagation speed down, and corrupts the transmitted signal. Moreover, the channel capacity is limited. The proposed system aims at effectively withstanding channel effects, minimizing the bit error rate and obtaining images with acceptable quality at the receiver. This system includes four parts; image source coding, channel coding, modulation scheme and image denoising. Set partitioning in hierarchical trees (SPIHT) is used for efficient image compression. To detect transmission errors, reduce the decoder complexity and increase the efficiency of transmission, convolutional coding is employed. Quadrature amplitude modulation (4-QAM) is used with orthogonal frequency division multiplexing (OFDM) to provide a reliable spectrally efficient system and enhance its resistivity to channel effects. Differential coding is used with 4-QAM to handle phase errors, while avoiding complex carrier tracking. To improve the image quality, 2-D double-density dual-tree complex discrete wavelet transform (DWT) is used to denoise the received decompressed image. Simulation results show that the proposed system is capable of transmitting and receiving underwater images with acceptable quality.

Optimization of Recursive Least Square-Based Adaptive Linear Equalizer for ZigBee Transceiver

An efficient technique to compensate for the channel detrimental effects in ZigBee systems is introduced in this paper. The proposed methodology relies on adding a recursive least square (RLS) based adaptive linear equalizer (ALE) to the physical layer of the receiver side. The performance of the RLS based ALE is investigated inside the ZigBee system under different multipath fading situations: Rician and Rayleigh. Moreover, the paper proposes a methodology for deciding the RLS based ALE’s design parameters. The design procedure depends on solving multiple objectives optimizing function based on genetic algorithms (GAs). The ALE’s parameters are chosen, such that the system experiences minimum bit error rate (BER) with fast convergence response. For design verification purposes, the ZigBee transceiver is modeled in MATLAB Simulink and tested under different fading and signal to noise ratios. In addition, the performance of the RLS adaptation algorithm is compared with the least mean square (LMS) one. The results show that the RLS based ALE provides better ZigBee performance with less BER and fast adaptation response.