A5002608156- Image and Signal Denoising Methods
- Fire Detection and Safety Systems
- Image Enhancement Techniques
- Neural Networks and Applications
- Video Surveillance and Tracking Methods
- Medical Image Segmentation Techniques
- Sparse and Compressive Sensing Techniques
- Physical Unclonable Functions (PUFs) and Hardware Security
- COVID-19 diagnosis using AI
- Blind Source Separation Techniques
- PAPR reduction in OFDM
- Advanced MRI Techniques and Applications
- Brain Tumor Detection and Classification
- Speech and Audio Processing
- Advanced Neural Network Applications
- Advanced Chemical Sensor Technologies
- Radiomics and Machine Learning in Medical Imaging
- CCD and CMOS Imaging Sensors
- Machine Fault Diagnosis Techniques
- Neural Networks and Reservoir Computing
- Digital Radiography and Breast Imaging
- AI in cancer detection
- Internet of Things and AI
- Cloud Data Security Solutions
- Security and Verification in Computing
Northwestern University
2024-2025
Intel (United States)
2024-2025
University of Illinois Chicago
2019-2024
Pacific Northwest Research Station
2022
China National Petroleum Corporation (China)
2015
In this paper, we propose a deep convolutional neural network for camera based wildfire detection. We train the via transfer learning and use window analysis strategy to increase fire detection rate. To achieve computational efficiency, calculate frequency response of kernels in dense layers eliminate those filters with low energy impulse response. Moreover, reduce storage edge devices, compare Fourier domain discard similar using cosine similarity measure domain. test performance variety...
Physically unclonable functions (PUFs) are a class of hardware-specific security primitives based on secret keys extracted from integrated circuits, which can protect important information against cyberattacks and reverse engineering. Here, we put forward an emerging type PUF in the electromagnetic domain by virtue self-dual absorber-emitter singularity that uniquely exists non-Hermitian parity-time (PT)-symmetric structures. At this singular point, reconfigurable emissive absorptive...
In this paper, we present a wireless ECG-derived Respiration Rate (RR) estimation using an autoencoder with DCT Layer. The wearable system records the ECG data of subject and respiration rate is determined from variations in baseline level data. A straightforward Fourier analysis obtained may lead to incorrect results due uneven breathing. To improve precision, propose neural network that uses novel Discrete Cosine Transform (DCT) layer denoise decorrelates has trainable weights...
In this paper, we propose a novel layer based on fast Walsh-Hadamard transform (WHT) and smooth-thresholding to replace 1 × convolution layers in deep neural networks. the WHT domain, denoise domain coefficients using new non-linearity, smoothed version of well-known soft-thresholding operator. We also introduce family multiplication-free operators from basic 2×2 Hadamard implement 3 depthwise separable layers. Using these two types layers, bottleneck MobileNet-V2 reduce network's number...
Convolution has been the core operation of modern deep neural networks. It is well known that convolutions can be implemented in Fourier Transform domain. In this article, we propose to use binary block Walsh–Hadamard transform (WHT) instead transform. We WHT-based layers replace some regular convolution utilize both one-dimensional (1D) and 2D WHTs article. 1D layers, compute WHT input feature map denoise domain coefficients using a nonlinearity obtained by combining soft-thresholding with...
In this article, we describe an ember detection method in infrared (IR) video. Embers, also called firebrands, can act as wildfire super-spreaders. We develop a novel neural network with Walsh-Hadamard Transform (WHT) layer to process the IR The WHT is used temporal dimension of video data model high-frequency activity due movements. insert ResNet-18 and obtained higher accuracy compared standard single slice processing entire block. repeat experiments on ResNet-34, but found that sufficient...
We propose a kernel-PCA-based method to detect anomaly in chemical sensors. use temporal signals produced by sensors form vectors perform the principal component analysis (PCA). estimate kernel-covariance matrix of sensor data and compute eigenvector corresponding largest eigenvalue covariance matrix. The can be detected comparing difference between actual reconstructed from dominant eigenvector. In this letter, we introduce new multiplication-free kernel, which is related <inline-formula...
In this article, we present a computationally efficient algorithm to detect Volatile Organic Compounds (VOC) leaking out of components used in chemical processes petrochemical refineries and plants. A VOC plume from damaged component appears as dynamic dark cloud infrared videos. We describe two-stage deep neural network structure, taking advantage both spatial temporal structure the texture regions created by plume. first moving pixels which are darker then their neighboring pixels. extract...
The edge processing of deep neural networks (DNNs) is becoming increasingly important due to its ability extract valuable information directly at the data source minimize latency and energy consumption. Although pruning techniques are commonly used reduce model size for computing, they have certain limitations. Frequency-domain compression, such as with Walsh–Hadamard transform (WHT), has been identified an efficient alternative. However, benefits frequency-domain often offset by increased...
In this paper, we present a set of efficient dimensionality reduction methods for array signal processing using <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\ell_{1}-$</tex> Kernel-based multiplication-free PCA ( xmlns:xlink="http://www.w3.org/1999/xlink">$\ell_{1}$</tex> -MF-PCA) techniques. Our proposed -MF-PCA utilize -norm kernels, which enhance the robustness approach compared to classical...
We consider a family of vector dot products that can be implemented using sign changes and addition operations only. The are energy-efficient as they avoid the multiplication operation entirely. Moreover, induce <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\ell_{1}$</tex> -norm, thus providing robustness to impulsive noise. First, we analytically prove yield symmetric, positive semi-definite generalized covariance matrices, enabling...
<p>The edge processing of deep neural networks (DNNs) is becoming increasingly important due to its ability extract valuable information directly at the data source minimize latency and energy consumption. Although pruning techniques are commonly used reduce model size for computing, they have certain limitations. Frequency-domain compression, such as with Walsh-Hadamard transform (WHT), has been identified an efficient alternative. However, benefits frequency-domain often offset by...
Power iteration is a fundamental algorithm in data analysis. It extracts the eigenvector corresponding to largest eigenvalue of given matrix. Applications include ranking algorithms, principal component analysis (PCA), among many others. Certain use cases may benefit from alternate, non-linear power methods with low complexity. In this paper, we introduce multiplication-avoiding (MAPI). MAPI replaces standard ℓ <inf xmlns:mml="http://www.w3.org/1998/Math/MathML"...
In this paper, we propose a deep convolutional neural network for camera based wildfire detection. We train the via transfer learning and use window analysis strategy to increase fire detection rate. To achieve computational efficiency, calculate frequency response of kernels in dense layers eliminate those filters with low energy impulse response. Moreover, reduce storage edge devices, compare Fourier domain discard similar using cosine similarity measure domain. test performance variety...
Accurate segmentation of organs from abdominal CT scans is essential for clinical applications such as diagnosis, treatment planning, and patient monitoring. To handle challenges heterogeneity in organ shapes, sizes, complex anatomical relationships, we propose a \textbf{\textit{\ac{MDNet}}}, an encoder-decoder network that uses the pre-trained \textit{MiT-B2} encoder multiple different decoder networks. Each connected to part via multi-scale feature enhancement dilated block. With each...
Central to the Transformer architectures' effectiveness is self-attention mechanism, a function that maps queries, keys, and values into high-dimensional vector space. However, training attention weights of non-trivial from state random initialization. In this paper, we propose two methods. (i) We first address initialization problem Vision Transformers by introducing simple, yet highly innovative, approach utilizing Discrete Cosine Transform (DCT) coefficients. Our proposed DCT-based marks...
Magnetic field inhomogeneity correction remains a challenging task in MRI analysis. Most established techniques are designed for brain by supposing that image intensities the identical tissue follow uniform distribution. Such an assumption cannot be easily applied to other organs, especially those small size and heterogeneous texture (large variations intensity), such as prostate. To address this problem, paper proposes probabilistic Hadamard U-Net (PHU-Net) prostate bias correction. First,...
In this article, we propose a set of transform-based neural network layers as an alternative to the <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$3\ttimes3$</tex-math> </inline-formula> Conv2D in convolutional networks (CNNs). The proposed can be implemented based on orthogonal transforms, such discrete cosine transform (DCT), Hadamard (HT), and biorthogonal block wavelet (BWT). Furthermore, by taking...