Showing papers by "Asifullah Khan published in 2019"

Recent Progress on Generative Adversarial Networks (GANs): A Survey

Abstract: Generative adversarial network (GANs) is one of the most important research avenues in the field of artificial intelligence, and its outstanding data generation capacity has received wide attention. In this paper, we present the recent progress on GANs. First, the basic theory of GANs and the differences among different generative models in recent years were analyzed and summarized. Then, the derived models of GANs are classified and introduced one by one. Third, the training tricks and evaluation metrics were given. Fourth, the applications of GANs were introduced. Finally, the problem, we need to address, and future directions were discussed.

Transfer learning based deep CNN for segmentation and detection of mitoses in breast cancer histopathological images.

Abstract: Segmentation and detection of mitotic nuclei is a challenging task. To address this problem, a Transfer Learning based fast and accurate system is proposed. To give the classifier a balanced dataset, this work exploits the concept of Transfer Learning by first using a pre-trained convolutional neural network (CNN) for segmentation, and then another Hybrid-CNN (with Weights Transfer and custom layers) for classification of mitoses. First, mitotic nuclei are automatically annotated, based on the ground truth centroids. The segmentation module then segments mitotic nuclei and also produces some false positives. Finally, the detection module is trained on the patches from the segmentation module and performs the final detection. Fine-tuning based Transfer Learning reduced training time, provided good initial weights, and improved the detection rate with F-measure of 0.713 and 76% area under the precision-recall curve for the challenging task of mitosis detection.

A Survey of the Recent Architectures of Deep Convolutional Neural Networks

Abstract: Deep Convolutional Neural Network (CNN) is a special type of Neural Networks, which has shown exemplary performance on several competitions related to Computer Vision and Image Processing. Some of the exciting application areas of CNN include Image Classification and Segmentation, Object Detection, Video Processing, Natural Language Processing, and Speech Recognition. The powerful learning ability of deep CNN is primarily due to the use of multiple feature extraction stages that can automatically learn representations from the data. The availability of a large amount of data and improvement in the hardware technology has accelerated the research in CNNs, and recently interesting deep CNN architectures have been reported. Several inspiring ideas to bring advancements in CNNs have been explored, such as the use of different activation and loss functions, parameter optimization, regularization, and architectural innovations. However, the significant improvement in the representational capacity of the deep CNN is achieved through architectural innovations. Notably, the ideas of exploiting spatial and channel information, depth and width of architecture, and multi-path information processing have gained substantial attention. Similarly, the idea of using a block of layers as a structural unit is also gaining popularity. This survey thus focuses on the intrinsic taxonomy present in the recently reported deep CNN architectures and, consequently, classifies the recent innovations in CNN architectures into seven different categories. These seven categories are based on spatial exploitation, depth, multi-path, width, feature-map exploitation, channel boosting, and attention. Additionally, the elementary understanding of CNN components, current challenges, and applications of CNN are also provided.

Bio-inspired heuristics for layer thickness optimization in multilayer piezoelectric transducer for broadband structures

Abstract: Recently, enhancement of sensitivity of multilayered piezoelectric transducer and reduction in electrical impedance has gained importance due to development of stacked active element designs. This work presents mathematical optimization of layer thicknesses for broadband structures using piezo-composite with ceramic and single crystal as active material for underwater SONAR. The proposed technique employs bio-inspired heuristics-based genetic algorithms by invoking one-dimensional thickness model. Initially, optimization has been performed for monolithic materials in the stack for various acoustic media, and then, the results were validated by comparing with the published data. In the proposed scheme, optimization is carried out for two-phase 1–3 piezo-composite stacks with same active elements for better mechanical output and broadband structure while preserving the minima among first three harmonics under $$-\,3$$ and $$-\,6$$ dB from the peaks in the frequency spectrum. The results show that the optimized single-crystal-based transducers have higher mechanical output and lower electrical impedance than their counterparts using piezo-ceramic in single- and two-phase materials.

Detecting Driver Drowsiness in Real Time Through Deep Learning Based Object Detection

Abstract: Vehicle accidents due to drowsiness in drivers take thousands of lives each year worldwide. This fact clearly exhibits a need for a drowsiness detection application that can help prevent such accidents and ultimately save lives. In this work, we propose a novel deep learning methodology based on Convolutional Neural Networks (CNN) to tackle this problem. The proposed methodology treats drowsiness detection as an object detection task, and from an incoming video stream of a driver, detects and localizes open and closed eyes. MobileNet CNN architecture with Single Shot Multibox Detector (SSD) is used for this task of object detection. A separate algorithm is then used to detect driver drowsiness based on the output from the MobileNet-SSD architecture. In order to train the MobileNet-SSD Network a custom dataset of about 6000 images was compiled and labeled with the objects face, eye open and eye closed. Out of these, 350 images were randomly separated and used to test the trained model. The trained model was evaluated on the test dataset using the PASCAL VOC metric and achieved a Mean Average Precision (mAP) of 0.84 on these categories. The proposed methodology, while maintaining reasonable accuracy, is also computationally efficient and cost effective, as it can process an incoming video stream in real time on a standalone mobile device without the need of expensive hardware support. It can easily be deployed on cheap embedded devices in vehicles, such as the Raspberry Pi 3 or a mobile smartphone.

Transfer Learning and Meta Classification Based Deep Churn Prediction System for Telecom Industry

Abstract: A churn prediction system guides telecom service providers to reduce revenue loss. However, the development of a churn prediction system for a telecom industry is a challenging task, mainly due to the large size of the data, high dimensional features, and imbalanced distribution of the data. In this paper, we present a solution to the inherent problems of churn prediction, using the concept of Transfer Learning (TL) and Ensemble-based Meta-Classification. The proposed method TL-DeepE is applied in two stages. The first stage employs TL by fine-tuning multiple pre-trained Deep Convolution Neural Networks (CNNs). Telecom datasets are normally in vector form, which is converted into 2D images because Deep CNNs have high learning capacity on images. In the second stage, predictions from these Deep CNNs are appended to the original feature vector and thus are used to build a final feature vector for the high-level Genetic Programming (GP) and AdaBoost based ensemble classifier. Thus, the experiments are conducted using various CNNs as base classifiers and the GP-AdaBoost as a meta-classifier. By using 10-fold cross-validation, the performance of the proposed TL-DeepE system is compared with existing techniques, for two standard telecommunication datasets; Orange and Cell2cell. Performing experiments on Orange and Cell2cell datasets, the prediction accuracy obtained was 75.4% and 68.2%, while the area under the curve was 0.83 and 0.74, respectively.

A Recent Survey on the Applications of Genetic Programming in Image Processing

Abstract: Genetic Programming (GP) has been primarily used to tackle optimization, classification, and feature selection related tasks. The widespread use of GP is due to its flexible and comprehensible tree-type structure. Similarly, research is also gaining momentum in the field of Image Processing, because of its promising results over vast areas of applications ranging from medical Image Processing to multispectral imaging. Image Processing is mainly involved in applications such as computer vision, pattern recognition, image compression, storage, and medical diagnostics. This universal nature of images and their associated algorithm, i.e., complexities, gave an impetus to the exploration of GP. GP has thus been used in different ways for Image Processing since its inception. Many interesting GP techniques have been developed and employed in the field of Image Processing, and consequently, we aim to provide the research community an extensive view of these techniques. This survey thus presents the diverse applications of GP in Image Processing and provides useful resources for further research. Also, the comparison of different parameters used in different applications of Image Processing is summarized in tabular form. Moreover, analysis of the different parameters used in Image Processing related tasks is carried-out to save the time needed in the future for evaluating the parameters of GP. As more advancement is made in GP methodologies, its success in solving complex tasks, not only in Image Processing but also in other fields, may increase. Additionally, guidelines are provided for applying GP in Image Processing related tasks, the pros and cons of GP techniques are discussed, and some future directions are also set.

Malware Classification using Deep Learning based Feature Extraction and Wrapper based Feature Selection Technique.

Abstract: In the case of malware analysis, categorization of malicious files is an essential part after malware detection. Numerous static and dynamic techniques have been reported so far for categorizing malware. This research presents a deep learning-based malware detection (DLMD) technique based on static methods for classifying different malware families. The proposed DLMD technique uses both the byte and ASM files for feature engineering, thus classifying malware families. First, features are extracted from byte files using two different Deep Convolutional Neural Networks (CNN). After that, essential and discriminative opcode features are selected using a wrapper-based mechanism, where Support Vector Machine (SVM) is used as a classifier. The idea is to construct a hybrid feature space by combining the different feature spaces to overcome the shortcoming of particular feature space and thus, reduce the chances of missing a malware. Finally, the hybrid feature space is used to train a Multilayer Perceptron, which classifies all nine different malware families. Experimental results show that proposed DLMD technique achieves log-loss of 0.09 for ten independent runs. Moreover, the proposed DLMD technique's performance is compared against different classifiers and shows its effectiveness in categorizing malware. The relevant code and database can be found at this https URL.

Cartography of rhodopsin-like G protein-coupled receptors across vertebrate genomes.

Abstract: We conduct a cartography of rhodopsin-like non-olfactory G protein-coupled receptors in the Ensembl database. The most recent genomic data (releases 90–92, 90 vertebrate genomes) are analyzed through the online interface and receptors mapped on phylogenetic guide trees that were constructed based on a set of ~14.000 amino acid sequences. This snapshot of genomic data suggest vertebrate genomes to harbour 142 clades of GPCRs without human orthologues. Among those, 69 have not to our knowledge been mentioned or studied previously in the literature, of which 28 are distant from existing receptors and likely new orphans. These newly identified receptors are candidates for more focused evolutionary studies such as chromosomal mapping as well for in-depth pharmacological characterization. Interestingly, we also show that 37 of the 72 human orphan (or recently deorphanized) receptors included in this study cluster into nineteen closely related groups, which implies that there are less ligands to be identified than previously anticipated. Altogether, this work has significant implications when discussing nomenclature issues for GPCRs.

Low complexity versatile video coding for traffic surveillance system

Abstract: The versatile video coding (VVC) standard can efficiently compress the video data of traffic surveillance system, however, the encoding complexity of VVC is quite high. In this paper, we propose an efficient motion estimation (ME) algorithm for improving the encoding complexity of VVC. First, we propose a content oriented adaptive search range setting algorithm, where the search range size of the children coding units (CUs) can be adaptively set by using the best motion vector information of their parent CU. In addition, based on the high spatial correlation and similar characteristics, we propose a fast reference frame direction decision algorithm to further reduce the ME encoding complexity. The simulation results show that the encoding complexity saving performance of the proposed algorithm is quite well, i.e., the total encoding time is saved by an average of 34.27%, and the ME encoding time is saved by an average of 40.79%.

Object detection using hybridization of static and dynamic feature spaces and its exploitation by ensemble classification

Abstract: This paper presents a learning mechanism based on hybridization of static and dynamic learning. Realizing the detection performances offered by the state-of-the-art deep learning techniques and the competitive performances offered by the conventional static learning techniques, we propose the idea of exploitation of the concatenated (parallel) hybridization of the static and dynamic learning-based feature spaces. This is contrary to the cascaded (series) hybridization topology in which the initial feature space (provided by the conventional, static, and handcrafted feature extraction technique) is explored using deep, dynamic, and automated learning technique. Consequently, the characteristics already suppressed by the conventional representation cannot be explored by the dynamic learning technique. Instead, the proposed technique combines the conventional static and deep dynamic representation in concatenated (parallel) topology to generate an information-rich hybrid feature space. Thus, this hybrid feature space may aggregate the good characteristics of both conventional and deep representations, which are then explored using an appropriate classification technique. We also hypothesize that ensemble classification may better exploit this parallel hybrid perspective of the feature spaces. For this purpose, pyramid histogram of oriented gradients-based static learning has been incorporated in conjunction with convolution neural network-based deep learning to produce concatenated hybrid feature space. This hybrid space is then explored with various state-of-the-art ensemble classification techniques. We have considered the publicly available INRIA person and Caltech pedestrian standard image datasets to assess the performance of the proposed hybrid learning system. Furthermore, McNemar’s test has been used to statistically validate the outperformance of the proposed technique over various contemporary techniques. The validated experimental results show that the employment of the proposed hybrid representation results in effective detection performance (an AUC of 0.9996 for INRIA person and 0.9985 for Caltech pedestrian datasets) as compared to the individual static and dynamic representations.

Ransomware Analysis using Feature Engineering and Deep Neural Networks.

Abstract: Detection and analysis of a potential malware specifically, used for ransom is a challenging task. Recently, intruders are utilizing advanced cryptographic techniques to get hold of digital assets and then demand a ransom. It is believed that generally, the files comprise of some attributes, states, and patterns that can be recognized by a machine learning technique. This work thus focuses on the detection of Ransomware by performing feature engineering, which helps in analyzing vital attributes and behaviors of the malware. The main contribution of this work is the identification of important and distinct characteristics of Ransomware that can help in detecting them. Finally, based on the selected features, both conventional machine learning techniques and Transfer Learning based Deep Convolutional Neural Networks have been used to detect Ransomware. In order to perform feature engineering and analysis, two separate datasets (static and dynamic) were generated. The static dataset has 3646 samples (1700 Ransomware and 1946 Goodware). On the other hand, the dynamic dataset comprised of 3444 samples (1455 Ransomware and 1989 Goodware). Through various experiments, it is observed that the Registry changes, API calls, and DLLs are the most important features for Ransomware detection. Additionally, important sequences are found with the help of the N-Gram technique. It is also observed that in the case of Registry Delete operation, if a malicious file tries to delete registries, it follows a specific and repeated sequence. However, for the benign file, it doesnt follow any specific sequence or repetition. Similarly, an interesting observation made through this study is that there is no common Registry deleted sequence between malicious and benign files. And thus this discernible fact can be readily exploited for Ransomware detection.

Adaptive transfer learning in deep neural networks: Wind power prediction using knowledge transfer from region to region and between different task domains

Abstract: Transfer Learning (TL) in Deep Neural Networks is gaining importance because in most of the applications, the labeling of data is costly and time-consuming. Additionally, TL also provides an effective weight initialization strategy for Deep Neural Networks . This paper introduces the idea of Adaptive Transfer Learning in Deep Neural Networks (ATL-DNN) for wind power prediction. Specifically, we show in case of wind power prediction that adaptive TL of Deep Neural Networks system can be adaptively modified as regards training on a different wind farm is concerned. The proposed ATL-DNN technique is tested for short-term wind power prediction, where continuously arriving information has to be exploited. Adaptive TL not only helps in providing good weight initialization, but is also helpful to utilize the incoming data for effective learning. Additionally, the proposed ATL-DNN technique is shown to transfer knowledge between different task domains (wind power to wind speed prediction) and from one region to another region. The simulation results show that the proposed ATL-DNN technique achieves average values of 0.0637,0.0986, and 0.0984 for the Mean-Absolute-Error, Root-Mean-Squared-Error, and Standard-Deviation-Error, respectively.