Showing items 1 - 7 of 7

Your selections:

SPEED: A deep learning assisted privacy-preserved framework for intelligent transportation systems

- Usman, Muhammad, Jan, Mian, Jolfaei, Alireza

  • Authors: Usman, Muhammad , Jan, Mian , Jolfaei, Alireza
  • Date: 2021
  • Type: Text , Journal article
  • Relation: IEEE Transactions on Intelligent Transportation Systems Vol. 22, no. 7 (2021), p. 4376-4384
  • Full Text: false
  • Reviewed:
  • Description: Roadside cameras in an Intelligent Transportation System (ITS) are used for various purposes, e.g., monitoring the speed of vehicles, violations of laws, and detection of suspicious activities in parking lots, streets, and side roads. These cameras generate big multimedia data, and as a result, the ITS faces challenges like data management, redundancy, and privacy breaching in end-to-end communication. To solve these challenges, we propose a framework, called SPEED, based on a multi-level edge computing architecture and machine learning algorithms. In this framework, data captured by end-devices, e.g., smart cameras, is distributed among multiple Level-One Edge Devices (LOEDs) to deal with data management issue and minimize packet drop due to buffer overflowing on end-devices and LOEDs. The data is forwarded from LOEDs to Level-Two Edge Devices (LTEDs) in a compressed sensed format. The LTEDs use an online Least-Squares Support-Vector Machines (LS-SVMs) model to determine distribution characteristics and index values of compressed sensed data to preserve its privacy during transmission between LTEDs and High-Level Edge Devices (HLEDs). The HLEDs estimate the redundancy in forwarded data using a deep learning architecture, i.e., a Convolutional Neural Network (CNN). The CNN is used to detect the presence of moving objects in the forwarded data. If a movement is detected, the data is forwarded to cloud servers for further analysis otherwise discarded. Experimental results show that the use of a multi-level edge computing architecture helps in managing the generated data. The machine learning algorithms help in addressing issues like data redundancy and privacy-preserving in end-to-end communication. © 2000-2011 IEEE.

Melanoma classification using efficientnets and ensemble of models with different input resolution

- Karki, Sagar, Kulkarni, Pradnya, Stranieri, Andrew

Efficient brain tumor segmentation with multiscale two-pathway-group conventional neural networks

- Razzak, Muhammad, Imran, Muhammad, Xu, Guandong

  • Authors: Razzak, Muhammad , Imran, Muhammad , Xu, Guandong
  • Date: 2019
  • Type: Text , Journal article
  • Relation: IEEE Journal of Biomedical and Health Informatics Vol. 23, no. 5 (2019), p. 1911-1919
  • Full Text:
  • Reviewed:
  • Description: Manual segmentation of the brain tumors for cancer diagnosis from MRI images is a difficult, tedious, and time-consuming task. The accuracy and the robustness of brain tumor segmentation, therefore, are crucial for the diagnosis, treatment planning, and treatment outcome evaluation. Mostly, the automatic brain tumor segmentation methods use hand designed features. Similarly, traditional methods of deep learning such as convolutional neural networks require a large amount of annotated data to learn from, which is often difficult to obtain in the medical domain. Here, we describe a new model two-pathway-group CNN architecture for brain tumor segmentation, which exploits local features and global contextual features simultaneously. This model enforces equivariance in the two-pathway CNN model to reduce instabilities and overfitting parameter sharing. Finally, we embed the cascade architecture into two-pathway-group CNN in which the output of a basic CNN is treated as an additional source and concatenated at the last layer. Validation of the model on BRATS2013 and BRATS2015 data sets revealed that embedding of a group CNN into a two pathway architecture improved the overall performance over the currently published state-of-the-art while computational complexity remains attractive. © 2013 IEEE.

A blockchain-based deep-learning-driven architecture for quality routing in wireless sensor networks

- Khan, Zahoor, Amjad, Sana, Ahmed, Farwa, Almasoud, Abdullah, Imran, Muhammad, Javaid, Nadeem

  • Authors: Khan, Zahoor , Amjad, Sana , Ahmed, Farwa , Almasoud, Abdullah , Imran, Muhammad , Javaid, Nadeem
  • Date: 2023
  • Type: Text , Journal article
  • Relation: IEEE Access Vol. 11, no. (2023), p. 31036-31051
  • Full Text:
  • Reviewed:
  • Description: Over the past few years, great importance has been given to wireless sensor networks (WSNs) as they play a significant role in facilitating the world with daily life services like healthcare, military, social products, etc. However, heterogeneous nature of WSNs makes them prone to various attacks, which results in low throughput, and high network delay and high energy consumption. In the WSNs, routing is performed using different routing protocols like low-energy adaptive clustering hierarchy (LEACH), heterogeneous gateway-based energy-aware multi-hop routing (HMGEAR), etc. In such protocols, some nodes in the network may perform malicious activities. Therefore, four deep learning (DL) techniques and a real-time message content validation (RMCV) scheme based on blockchain are used in the proposed network for the detection of malicious nodes (MNs). Moreover, to analyse the routing data in the WSN, DL models are trained on a state-of-the-art dataset generated from LEACH, known as WSN-DS 2016. The WSN contains three types of nodes: sensor nodes, cluster heads (CHs) and the base station (BS). The CHs after aggregating the data received from the sensor nodes, send it towards the BS. Furthermore, to overcome the single point of failure issue, a decentralized blockchain is deployed on CHs and BS. Additionally, MNs are removed from the network using RMCV and DL techniques. Moreover, legitimate nodes (LNs) are registered in the blockchain network using proof-of-authority consensus protocol. The protocol outperforms proof-of-work in terms of computational cost. Later, routing is performed between the LNs using different routing protocols and the results are compared with original LEACH and HMGEAR protocols. The results show that the accuracy of GRU is 97%, LSTM is 96%, CNN is 92% and ANN is 90%. Throughput, delay and the death of the first node are computed for LEACH, LEACH with DL, LEACH with RMCV, HMGEAR, HMGEAR with DL and HMGEAR with RMCV. Moreover, Oyente is used to perform the formal security analysis of the designed smart contract. The analysis shows that blockchain network is resilient against vulnerabilities. © 2013 IEEE.

CNN and GRU based deep neural network for electricity theft detection to secure smart grid

- Ullah, Ashraf, Javaid, Nadeem, Samuel, Omaji, Imran, Muhammad, Shoaib, Muhammad

  • Authors: Ullah, Ashraf , Javaid, Nadeem , Samuel, Omaji , Imran, Muhammad , Shoaib, Muhammad
  • Date: 2020
  • Type: Text , Conference paper
  • Relation: 16th IEEE International Wireless Communications and Mobile Computing Conference, IWCMC 2020, Limassol, Cyprus, 15 to 19 June 2020, 2020 International Wireless Communications and Mobile Computing, IWCMC 2020 p. 1598-1602
  • Full Text: false
  • Reviewed:
  • Description: In this paper, a Hybrid Deep Neural Network (HDNN) is proposed in this work, which is the combination of Convolutional Neural Network (CNN), Gated Recurrent Unit (GRU) and Particle Swarm Optimization (PSO), termed as CNN-GRU-PSO HDNN. In this paper, real time electricity consumption data of consumers is used, which is taken from an easily available online source, named as State Grid Corporation of China (SGCC). The original dataset consists of actual values along with the erroneous and missing values. The pre-processing steps are performed initially to refine the data. After that, feature selection and extraction are performed using CNN, which reduce both the dimensionality and the redundancy present in the dataset. Furthermore, the classification of provided data into honest and fake consumers is done using GRU-PSO technique. The proposed HDNN model's performance is then compared with various benchmark techniques like Logistic Regression (LR), Support Vector Machine (SVM), Long Short Term Memory (LSTM) and GRU. The efficiency of the proposed model is validated using various performance parameters like Area Under the Curve (AUC), precision, accuracy, recall and F1-Score. The simulation results show that the proposed model outperforms the existing techniques in terms of ETD and class imbalanced issues. Moreover, the proposed model is also more robust and accurate than the existing methods. © 2020 IEEE.

Distortion robust image classification using deep convolutional neural network with discrete cosine transform

- Hossain, Md Tahmid, Teng, Shyh, Zhang, Dengsheng, Lim, Suryani, Lu, Guojun

  • Authors: Hossain, Md Tahmid , Teng, Shyh , Zhang, Dengsheng , Lim, Suryani , Lu, Guojun
  • Date: 2019
  • Type: Text , Conference proceedings
  • Relation: 2019 IEEE International Conference on Image Processing (ICIP);Taipei, Taiwan; 22-25 Sept, 2019 p. 659-663
  • Full Text: false
  • Reviewed:
  • Description: Convolutional Neural Networks are highly effective for image classification. However, it is still vulnerable to image distortion. Even a small amount of noise or blur can severely hamper the performance of these CNNs. Most work in the literature strives to mitigate this problem simply by fine-tuning a pre-trained CNN on mutually exclusive or a union set of distorted training data. This iterative fine-tuning process with all known types of distortion is exhaustive and the network struggles to handle unseen distortions. In this work, we propose distortion robust DCT-Net, a Discrete Cosine Transform based module integrated into a deep network which is built on top of VGG16 [1]. Unlike other works in the literature, DCT-Net is "blind" to the distortion type and level in an image both during training and testing. The DCT-Net is trained only once and applied in a more generic situation without further retraining. We also extend the idea of dropout and present a training adaptive version of the same. We evaluate our proposed DCT-Net on a number of benchmark datasets. Our experimental results show that once trained, DCT-Net not only generalizes well to a variety of unseen distortions but also outperforms other comparable networks in the literature.

Towards robust convolutional neural networks in challenging environments

- Hossain, Md Tahmid

  • Authors: Hossain, Md Tahmid
  • Date: 2021
  • Type: Text , Thesis , PhD
  • Full Text:
  • Description: Image classification is one of the fundamental tasks in the field of computer vision. Although Artificial Neural Network (ANN) showed a lot of promise in this field, the lack of efficient computer hardware subdued its potential to a great extent. In the early 2000s, advances in hardware coupled with better network design saw the dramatic rise of Convolutional Neural Network (CNN). Deep CNNs pushed the State-of-The-Art (SOTA) in a number of vision tasks, including image classification, object detection, and segmentation. Presently, CNNs dominate these tasks. Although CNNs exhibit impressive classification performance on clean images, they are vulnerable to distortions, such as noise and blur. Fine-tuning a pre-trained CNN on mutually exclusive or a union set of distortions is a brute-force solution. This iterative fine-tuning process with all known types of distortion is, however, exhaustive and the network struggles to handle unseen distortions. CNNs are also vulnerable to image translation or shift, partly due to common Down-Sampling (DS) layers, e.g., max-pooling and strided convolution. These operations violate the Nyquist sampling rate and cause aliasing. The textbook solution is low-pass filtering (blurring) before down-sampling, which can benefit deep networks as well. Even so, non-linearity units, such as ReLU, often re-introduce the problem, suggesting that blurring alone may not suffice. Another important but under-explored issue for CNNs is unknown or Open Set Recognition (OSR). CNNs are commonly designed for closed set arrangements, where test instances only belong to some ‘Known Known’ (KK) classes used in training. As such, they predict a class label for a test sample based on the distribution of the KK classes. However, when used under the OSR setup (where an input may belong to an ‘Unknown Unknown’ or UU class), such a network will always classify a test instance as one of the KK classes even if it is from a UU class. Historically, CNNs have struggled with detecting objects in images with large difference in scale, especially small objects. This is because the DS layers inside a CNN often progressively wipe out the signal from small objects. As a result, the final layers are left with no signature from these objects leading to degraded performance. In this work, we propose solutions to the above four problems. First, we improve CNN robustness against distortion by proposing DCT based augmentation, adaptive regularisation, and noise suppressing Activation Functions (AF). Second, to ensure further performance gain and robustness to image transformations, we introduce anti-aliasing properties inside the AF and propose a novel DS method called blurpool. Third, to address the OSR problem, we propose a novel training paradigm that ensures detection of UU classes and accurate classification of the KK classes. Finally, we introduce a novel CNN that enables a deep detector to identify small objects with high precision and recall. We evaluate our methods on a number of benchmark datasets and demonstrate that they outperform contemporary methods in the respective problem set-ups.
  • Description: Doctor of Philosophy
  • 1