IU Indianapolis ScholarWorks :: Browsing by Author "Rizkalla, Maher E."

Browsing by Author "Rizkalla, Maher E."

Now showing 1 - 10 of 47

A 2D PLUS DEPTH VIDEO CAMERA PROTOTYPE USING DEPTH FROM DEFOCUS IMAGING AND A SINGLE MICROFLUIDIC LENS
(2011-08) Li, Weixu; Christopher, Lauren; Rizkalla, Maher E.; Salama, Paul
A new method for capturing 3D video from a single imager and lens is introduced in this research. The benefit of this method is that it does not have the calibration and alignment issues associated with binocular 3D video cameras, and allows for a less expensive overall system. The digital imaging technique Depth from Defocus (DfD) has been successfully used in still camera imaging to develop a depth map associated with the image. However, DfD has not been applied in real-time video so far since the focus mechanisms are too slow to produce real-time results. This new research result shows that a Microfluidic lens is capable of the required focal length changes at 2x video frame rate, due to the electrostatic control of the focus. During the processing, two focus settings per output frame are captured using this lens combined with a broadcast video camera prototype. We show that the DfD technique using Bayesian Markov Random Field optimization can produce a valid depth map.
3D EM/MPM MEDICAL IMAGE SEGMENTATION USING AN FPGA EMBEDDED DESIGN IMPLEMENTATION
(2011-08) Liu, Chao; Christopher, Lauren; Rizkalla, Maher E.; Salama, Paul
This thesis presents a Field Programmable Gate Array (FPGA) based embedded system which is used to achieve high speed segmentation of 3D images. Segmenta- tion is performed using Expectation-Maximization with Maximization of Posterior Marginals (EM/MPM) Bayesian algorithm. In this system, the embedded processor controls a custom circuit which performs the MPM and portions of the EM algorithm. The embedded processor completes the EM algorithm and also controls image data transmission between host computer and on-board memory. The whole system has been implemented on Xilinx Virtex 6 FPGA and achieved over 100 times improvement compared to standard desktop computing hardware.
3D ENDOSCOPY VIDEO GENERATED USING DEPTH INFERENCE: CONVERTING 2D TO 3D
(2013-08-20) Rao, Swetcha; Christopher, Lauren; Rizkalla, Maher E.; Salama, Paul; King, Brian
A novel algorithm was developed to convert raw 2-dimensional endoscope videos into 3-dimensional view. Minimally invasive surgeries aided with 3D view of the invivo site have shown to reduce errors and improve training time compared to those with 2D view. The novelty of this algorithm is that two cues in the images have been used to develop the 3D. Illumination is the rst cue used to nd the darkest regions in the endoscopy images in order to locate the vanishing point(s). The second cue is the presence of ridge-like structures in the in-vivo images of the endoscopy image sequence. Edge detection is used to map these ridge-like structures into concentric ellipses with their common center at the darkest spot. Then, these two observations are used to infer the depth of the endoscopy videos; which then serves to convert them from 2D to 3D. The processing time is between 21 seconds to 20 minutes for each frame, on a 2.27GHz CPU. The time depends on the number of edge pixels present in the edge-detection image. The accuracy of ellipse detection was measured to be 98.98% to 99.99%. The algorithm was tested on 3 truth images with known ellipse parameters and also on real bronchoscopy image sequences from two surgical procedures. Out of 1020 frames tested in total, 688 frames had single vanishing point while 332 frames had two vanishing points. Our algorithm detected the single vanishing point in 653 of the 688 frames and two vanishing points in 322 of the 332 frames.
3D Image Segmentation Implementation on FPGA Using EM/MPM Algorithm
(2010-12) Sun, Yan; Christopher, Lauren; Rizkalla, Maher E.; Salama, Paul
In this thesis, 3D image segmentation is targeted to a Xilinx Field Programmable Gate Array (FPGA), and verified with extensive simulation. Segmentation is performed using the Expectation-Maximization with Maximization of the Posterior Marginals (EM/MPM) Bayesian algorithm. This algorithm segments the 3D image using neighboring pixels based on a Markov Random Field (MRF) model. This iterative algorithm is designed, synthesized and simulated for the Xilinx FPGA, and greater than 100 times speed improvement over standard desktop computer hardware is achieved. Three new techniques were the key to achieving this speed: Pipelined computational cores, sixteen parallel data paths and a novel memory interface for maximizing the external memory bandwidth. Seven MPM segmentation iterations are matched to the external memory bandwidth required of a single source file read, and a single segmented file write, plus a small amount of latency.
Advancing profiling sensors with a wireless approach
(2013-11-20) Galvis, Alejandro; Russomanno, David J.; Li, Feng; Rizkalla, Maher E.; King, Brian
In general, profiling sensors are low-cost crude imagers that typically utilize a sparse detector array, whereas traditional cameras employ a dense focal-plane array. Profiling sensors are of particular interest in applications that require classification of a sensed object into broad categories, such as human, animal, or vehicle. However, profiling sensors have many other applications in which reliable classification of a crude silhouette or profile produced by the sensor is of value. The notion of a profiling sensor was first realized by a Near-Infrared (N-IR), retro-reflective prototype consisting of a vertical column of sparse detectors. Alternative arrangements of detectors have been implemented in which a subset of the detectors have been offset from the vertical column and placed at arbitrary locations along the anticipated path of the objects of interest. All prior work with the N-IR, retro-reflective profiling sensors has consisted of wired detectors. This thesis surveys prior work and advances this work with a wireless profiling sensor prototype in which each detector is a wireless sensor node and the aggregation of these nodes comprises a profiling sensor’s field of view. In this novel approach, a base station pre-processes the data collected from the sensor nodes, including data realignment, prior to its classification through a back-propagation neural network. Such a wireless detector configuration advances deployment options for N-IR, retro-reflective profiling sensors.
AI on the Edge with CondenseNeXt: An Efficient Deep Neural Network for Devices with Constrained Computational Resources
(2021-08) Kalgaonkar, Priyank B.; El-Sharkawy, Mohamed A.; King, Brian S.; Rizkalla, Maher E.
Research work presented within this thesis propose a neoteric variant of deep convolutional neural network architecture, CondenseNeXt, designed specifically for ARM-based embedded computing platforms with constrained computational resources. CondenseNeXt is an improved version of CondenseNet, the baseline architecture whose roots can be traced back to ResNet. CondeseNeXt replaces group convolutions in CondenseNet with depthwise separable convolutions and introduces group-wise pruning, a model compression technique, to prune (remove) redundant and insignificant elements that either are irrelevant or do not affect performance of the network upon disposition. Cardinality, a new dimension to the existing spatial dimensions, and class-balanced focal loss function, a weighting factor inversely proportional to the number of samples, has been incorporated in order to relieve the harsh effects of pruning, into the design of CondenseNeXt’s algorithm. Furthermore, extensive analyses of this novel CNN architecture was performed on three benchmarking image datasets: CIFAR-10, CIFAR-100 and ImageNet by deploying the trained weight on to an ARM-based embedded computing platform: NXP BlueBox 2.0, for real-time image classification. The outputs are observed in real-time in RTMaps Remote Studio’s console to verify the correctness of classes being predicted. CondenseNeXt achieves state-of-the-art image classification performance on three benchmark datasets including CIFAR-10 (4.79% top-1 error), CIFAR-100 (21.98% top-1 error) and ImageNet (7.91% single model, single crop top-5 error), and up to 59.98% reduction in forward FLOPs compared to CondenseNet. CondenseNeXt can also achieve a final trained model size of 2.9 MB, however at the cost of 2.26% in accuracy loss. Thus, performing image classification on ARM-Based computing platforms without requiring a CUDA enabled GPU support, with outstanding efficiency.
Antenna Design and SAR Analysis on Human Head Phantom Simulation for Future Clinical Applications
(Scientific Research Publishing, 2017-09) Perez, Felipe Pablo; Bandeira, Joseph Paul; Morisaki, Jorge J.; Krishna Peddinti, Seshasai Vamsi; Salama, Paul; Rizkalla, James; Rizkalla, Maher E.; Medicine, School of Medicine
Background The rapid development of a variety of devices that emit Radiofrequency Electromagnetic fields (RF-EMF) has sparked growing interest in their interaction with biological systems and the beneficial effects on human health. As a result, investigations have been driven by the potential for therapeutic applications, as well as concern for any possible negative health implications of these EM energies [-]. Recent results have indicated specific tuning of experimental and clinical RF exposure may lead to their clinical application toward beneficial health outcomes []. Method In the current study, a mathematical and computer simulation model to analyze a specific RF-EMF exposure on a human head model was developed. Impetus for this research was derived from results of our previous experiments which revealed that Repeated Electromagnetic Field Stimulation (REMFS) decreased the toxic levels of beta amyloid (Aβ) in neuronal cells, thereby suggesting a new potential therapeutic strategy for the treatment of Alzheimer's disease (AD). Throughout development of the proposed device, experimental variables such as the EM frequency range, specific absorption rate (SAR), penetration depth, and innate properties of different tissues have been carefully considered. Results RF-EMF exposure to the human head phantom was performed utilizing a Yagi-Uda antenna type possessing high gain (in the order of 10 dbs) at a frequency of 64 MHz and SAR of 0.6 W/Kg. In order to maximize the EM power transmission in one direction, directors were placed in front of the driven element and reflectors were placed behind the driven element. So as to strategically direct the EM field into the center of the brain tissue, while providing field linearity, our analysis considered the field distribution for one versus four antennas. Within the provided dimensions of a typical human brain, results of the Bioheat equation within COMSOL Multiphysics version 5.2a software demonstrated less than a 1 m˚K increase from the absorbed EM power.
ASIC-Implemented MicroBlaze-Based Coprocessor for Data Stream Management Systems
(2020-05) Balasubramanian, Linknath Surya; Lee, John J.; Christopher, Lauren A; Rizkalla, Maher E.
The drastic increase in Internet usage demands the need for processing data in real time with higher efficiency than ever before. Symbiote Coprocessor Unit (SCU), developed by Dr. Pranav Vaidya, is a hardware accelerator which has potential of providing data processing speedup of up to 150x compared with traditional data stream processors. However, SCU implementation is very complex, fixed, and uses an outdated host interface, which limits future improvement. Mr. Tareq S. Alqaisi, an MSECE graduate from IUPUI worked on curbing these limitations. In his architecture, he used a Xilinx MicroBlaze microcontroller to reduce the complexity of SCU along with few other modifications. The objective of this study is to make SCU suitable for mass production while reducing its power consumption and delay. To accomplish this, the execution unit of SCU has been implemented in application specific integrated circuit and modules such as ACG/OCG, sequential comparator, and D-word multiplier/divider are integrated into the design. Furthermore, techniques such as operand isolation, buffer insertion, cell swapping, and cell resizing are also integrated into the system. As a result, the new design attains 67.9435 µW of dynamic power as compared to 74.0012 µW before power optimization along with a small increase in static power, 39.47 ns of clock period as opposed to 52.26 ns before time optimization.
Attached Learning Model for First Digital System Design Course in ECE Program
(American Society for Engineering Education, 2016-06) Shayesteh, Seemein; Rizkalla, Maher E.; Christopher, Lauren; Miled, Zina Ben; Department of Electrical and Computer Engineering, School of Engineering and Technology
Control of Non-minimum Phase Power Converters
(2012-05) Gavini, Sree Likhita; Izadian, Afshin; Li, Lingxi; Rizkalla, Maher E.
The inner structural characteristics of non-minimum phase DC-DC converters pose a severe limitation in direct regulation of voltage when addressed from a control perspective. This constraint is reflected by the presence of right half plane zeros or the unstable zero dynamics of the output voltage of these converters. The existing controllers make use of one-to-one correspondence between the voltage and current equilibriums of the non-minimum phase converters and exploit the property that when the average output of these converters is the inductor current, the system dynamics are stable and hence they indirectly regulate the voltage. As a result, the system performance is susceptible to circuit parameter and load variation and require additional controllers, which in turn increase the system complexity. In this thesis, a novel approach to this problem is proposed for second order non-minimum phase converters such as Boost and Buck-Boost Converter. Different solutions have been suggested to the problem based on whether the converter is modeled as a linear system or as a nonlinear system. For the converter modeled as a linear system, the non-minimum phase part of the system is decoupled and its transfer function is converted to minimum phase using a parallel compensator. Then the control action is achieved by using a simple proportional gain controller. This method accelerates the transient response of the converter, reduces the initial undershoot in the response, and considerably reduces the oscillations in the transient response. Simulation results demonstrate the effectiveness of the proposed approach. When the converter is modeled as a bilinear system, it preserves the stabilizing nonlinearities of the system. Hence, a more effective control approach is adopted by using Passivity properties. In this approach, the non-minimum phase converter system is viewed from an energy-based perspective and the property of passivity is used to achieve stable zero dynamics of the output voltage. A system is passive if its rate of energy storage is less than the supply rate i.e. the system dissipates more energy than stores. As a result, the energy storage function of the system is less than the supply rate function. Non-minimum phase systems are not passive, and passivation of non-minimum phase power converters is an attractive solution to the posed problem. Stability of non-minimum phase systems can also be investigated by defining the passivity indices. This research approaches the problem by characterizing the degree of passivity i.e. the amount of damping in the system, from passivity indices. Thus, the problem is viewed from a system level rather than from a circuit level description. This method uses feed-forward passivation to compensate for the shortage of passivity in the non-minimum phase converter and makes use of a parallel interconnection to the open-loop system to attain exponentially stable zero dynamics of the output voltage. Detailed analytical analysis regarding the control structure and passivation process is performed on a buck-boost converter. Simulation and experimental results carried out on the test bed validate the effectiveness of the proposed method.

Browsing by Author "Rizkalla, Maher E."

Results Per Page

Sort Options