Memory Efficient Neural Architecture Search for Image Classification

Amrita Rana,Kyung Ki Kim 대한전자공학회 2021 대한전자공학회 학술대회 Vol.2021 No.6

Robust architecture search using network adaptation

( Amrita Rana ),( Kyung Ki Kim ) 한국센서학회 2021 센서학회지 Vol.30 No.5

Experts have designed popular and successful model architectures, which, however, were not the optimal option for different scenarios. Despite the remarkable performances achieved by deep neural networks, manually designed networks for classification tasks are the backbone of object detection. One major challenge is the ImageNet pre-training of the search space representation; moreover, the searched network incurs huge computational cost. Therefore, to overcome the obstacle of the pre-training process, we introduce a network adaptation technique using a pre-trained backbone model tested on ImageNet. The adaptation method can efficiently adapt the manually designed network on ImageNet to the new object-detection task. Neural architecture search (NAS) is adopted to adapt the architecture of the network. The adaptation is conducted on the MobileNetV2 network. The proposed NAS is tested using SSDLite detector. The results demonstrate increased performance compared to existing network architecture in terms of search cost, total number of adder arithmetics (Madds), and mean Average Precision(mAP). The total computational cost of the proposed NAS is much less than that of the State Of The Art (SOTA) NAS method.

A Novel Spiking Neural Network for ECG signal Classification

( Amrita Rana ),( Kyung Ki Kim ) 한국센서학회 2021 센서학회지 Vol.30 No.1

The electrocardiogram (ECG) is one of the most extensively employed signals used to diagnose and predict cardiovascular diseases (CVDs). In recent years, several deep learning (DL) models have been proposed to improve detection accuracy. Among these, deep neural networks (DNNs) are the most popular, wherein the features are extracted automatically. Despite the increment in classification accuracy, DL models require exorbitant computational resources and power. This causes the mapping of DNNs to be slow; in addition, the mapping is challenging for a wearable device. Embedded systems have constrained power and memory resources. Therefore full-precision DNNs are not easily deployable on devices. To make the neural network faster and more power-efficient, spiking neural networks (SNNs) have been introduced for fewer operations and less complex hardware resources. However, the conventional SNN has low accuracy and high computational cost. Therefore, this paper proposes a new binarized SNN which modifies the synaptic weights of SNN constraining it to be binary (+1 and -1). In the simulation results, this paper compares the DL models and SNNs and evaluates which model is optimal for ECG classification. Although there is a slight compromise in accuracy, the latter proves to be energy-efficient.

Cardiac Disease Detection Using Modified Pan―Tompkins Algorithm

( Amrita Rana ),( Kyung Ki Kim ) 한국센서학회 2019 센서학회지 Vol.28 No.1

The analysis of electrocardiogram (ECG) signals facilitates the detection of various abnormal conditions of the human heart. The QRS complex is the most critical part of the ECG waveform. Further, different diseases can be identified based on the QRS complex. In this paper, a new algorithm based on the well-known Pan-Tompkins algorithm has been proposed. In the proposed scheme, the QRS complex is initially extracted by removing the background noise. Subsequently, the R-R interval and heart rate are calculated to detect whether the ECG is normal or has some abnormalities such as tachycardia and bradycardia. The accuracy of the proposed algorithm is found to be almost the same as the Pan-Tompkins algorithm and increases the R peak detection processing speed. For this work, samples are used from the MIT-BIH Arrhythmia Database, and the simulation is carried out using MATLAB 2016a.

Comparison of Artificial Neural Networks for Low-Power ECG-Classification System

( Amrita Rana ),( Kyung Ki Kim ) 한국센서학회 2020 센서학회지 Vol.29 No.1

Electrocardiogram (ECG) classification has become an essential task of modern day wearable devices, and can be used to detect cardiovascular diseases. State-of-the-art Artificial Intelligence (AI)-based ECG classifiers have been designed using various artificial neural networks (ANNs). Despite their high accuracy, ANNs require significant computational resources and power. Herein, three different ANNs have been compared: multilayer perceptron (MLP), convolutional neural network (CNN), and spiking neural network (SNN) only for the ECG classification. The ANN model has been developed in Python and Theano, trained on a central processing unit (CPU) platform, and deployed on a PYNQ-Z2 FPGA board to validate the model using a Jupyter notebook. Meanwhile, the hardware accelerator is designed with Overlay, which is a hardware library on PYNQ. For classification, the MIT-BIH dataset obtained from the Physionet library is used. The resulting ANN system can accurately classify four ECG types: normal, atrial premature contraction, left bundle branch block, and premature ventricular contraction. The performance of the ECG classifier models is evaluated based on accuracy and power. Among the three AI algorithms, the SNN requires the lowest power consumption of 0.226 W on-chip, followed by MLP (1.677 W), and CNN (2.266 W). However, the highest accuracy is achieved by the CNN (95%), followed by MLP (76%) and SNN (90%).

FPGA Implementation of an Artificial Intelligence Signal Recognition System

( Amrita Rana ),( Kyung Ki Kim ) 한국센서학회 2022 센서학회지 Vol.31 No.1

Cardiac disease is the most common cause of death worldwide. Therefore, detection and classification of electrocardiogram (ECG) signals are crucial to extend life expectancy. In this study, we aimed to implement an artificial intelligence signal recognition system in field programmable gate array (FPGA), which can recognize patterns of bio-signals such as ECG in edge devices that require batteries. Despite the increment in classification accuracy, deep learning models require exorbitant computational resources and power, which makes the mapping of deep neural networks slow and implementation on wearable devices challenging. To overcome these limitations, spiking neural networks (SNNs) have been applied. SNNs are biologically inspired, event-driven neural networks that compute and transfer information using discrete spikes, which require fewer operations and less complex hardware resources. Thus, they are more energy-efficient compared to other artificial neural networks algorithms.

Lightweight image classifier for CIFAR-10

( Akshay Kumar Sharma ),( Amrita Rana ),( Kyung Ki Kim ) 한국센서학회 2021 센서학회지 Vol.30 No.5

Image classification is one of the fundamental applications of computer vision. It enables a system to identify an object in an image. Recently, image classification applications have broadened their scope from computer applications to edge devices. The convolutional neural network (CNN) is the main class of deep learning neural networks that are widely used in computer tasks, and it delivers high accuracy. However, CNN algorithms use a large number of parameters and incur high computational costs, which hinder their implementation in edge hardware devices. To address this issue, this paper proposes a lightweight image classifier that provides good accuracy while using fewer parameters. The proposed image classifier diverts the input into three paths and utilizes different scales of receptive fields to extract more feature maps while using fewer parameters at the time of training. This results in the development of a model of small size. This model is tested on the CIFAR-10 dataset and achieves an accuracy of 90% using .26M parameters. This is better than the state-of-the-art models, and it can be implemented on edge devices.